jfa journal

AND option

OR option



S. Tejiram1,2, C. Galet3, J. Cartwright4, V.H. Hatcher3, D.A. Skeete3, C. Cocanour2, K.S. Romanowski1


1. Department of Surgery, Division of Burn Surgery, University of California, Davis, Sacramento, CA. 2315 Stockton Blvd, Sacramento, CA, USA; 2. Department of Surgery, Division of Trauma Surgery, University of California, Davis, 2315 Stockton Blvd, Sacramento, CA, USA; 3. Department of Surgery, Division of Acute Care Surgery, University of Iowa, 200 Hawkins Dr., Iowa City, IA, USA; 4. University of Michigan School of Medicine. 1500 E Medical Center Dr, Ann Arbor, MI, USA.

Corresponding Author: Kathleen S. Romanowski, MD, University of California, Davis, and Shriners Hospitals for Children Northern CaliforniaN Stockton Blvd. Suite 718, Sacramento, CA 95762, Phone: 916-453-5022, Email: ksromanowski@ucdavis.edu

J Frailty Aging 2021;in press
Published online June 17, 2021, http://dx.doi.org/10.14283/jfa.2021.25



Purpose: The older population is particularly vulnerable to traumatic injury. Frailty scores, used to estimate the physiologic status of an individual, are key to identifying those most at risk for injury. Global health measures such as the Veterans RAND 12 Item Health Survey (VR-12) are quality of life measures that assess older adults’ overall perception of their health and may serve as a useful adjunct when predicting frailty. Herein, we evaluated whether components of the VR-12 correlated with worse frailty scores over time.
Methods: Older adults (≥65) admitted to burn, trauma, or emergency general surgery services were prospectively enrolled. Demographics, frailty determined using the Trauma Specific Frailty Index (TSFI), and VR-12 surveys were collected at enrollment and 3, 6, 9, and 12-month follow-ups. A physical component score (PCS) and mental component score (MCS) was produced by VR-12 surveys for comparison purposes.
Results: Fifty-eight patients were enrolled, of which 8 died. No significant changes in PCS (p = 0.25) and MCS (p = 0.56) were observed over time. PCS (p = 0.97) and MCS (p = 0.78) at enrollment did not predict mortality. PCS (OR = 0.894 [0.84-0.95], p = 0.0004) and age (OR = 1.113 [1.012-1.223], p = 0.03) independently predicted enrollment frailty.
Conclusion: These global measures of health could be utilized in lieu or in addition to frailty scores when assessing patients in the setting of acute injury. Studies are warranted to confirm this association.

Key words: Frailty, older adults, trauma, VR-12.



Health related quality of life measures obtained from individuals can provide information on a wide range of outcomes such as disease burden and treatment effectiveness (1, 2). These self-reported outcomes can be collected and tracked to better predict resource utilization and mortality (3, 4). Global health measures are typically generic and concise to allow application to a wide variety of patient populations (2). Though these measures are not explicitly used in economic evaluations, there is increasing interest in incorporating global health perceptions into cost analysis to better inform health care decision making (5, 6). The Centers for Medicare & Medicaid Services have adopted global health perception assessments in their cost-effectiveness analysis and the Health Outcomes Survey (HOS) became the first Medicare managed outcomes measure in the Healthcare Effectiveness Data and Information Set (HEDIS) (7). In this context, surveys such as the Veteran RAND 12-Item Health Survey (VR-12) have become commonly used in population-based studies like the Medicare HOS to more efficiently utilize health care resources (8, 9).
Global health measures like the VR-12 are quality of life measures that assess an individual’s overall perception of their health and were originally developed from the Short Form-36 (SF-36) Health Survey. The SF-36 is a 36-item questionnaire that assesses patient health across eight dimensions such as health perceptions, physical and social functioning, and mental health to produce summary scores for physical and mental health. Individuals provide self-assessment and perceptions of their health by answering questions about their perceived limitations. For example, one may answer a question about their ability to ambulate as limited by a lot, a little, or not at all. These answers correspond to a score of 1, 2, or 3, respectively, to calculate a score range from 10 to 30 that is later transformed to a 100-point scale for comparison purposes. In this survey, each answer is given equal weight for score determination. Previous literature has established the significance of these scores to detect meaningful differences following intervention among clinic patient groups with various disease processes (2, 5, 10-12). The Veterans RAND 36-Item Health Survey (VR-36) was developed from the SF-36 and both were reduced and adapted from a 36-item survey to 12 items with minimal information loss to create the VR-12 survey (13-17).
The VR-12 is freely available to assess health related quality of life and disease burden through a 12 item questionnaire focused again on eight dimensions: general health, physical functioning, changes and limitations in physical and emotional health, pain, energy, mental health, and social functioning (2, 15, 18). Answers to these questions produce a physical component score (PCS) and a mental component score (MCS). The PCS focuses more on questions about general health, physical functioning and pain whereas the MCS places more emphasis on emotional, mental health, and social functioning questions (18).
Quality of life measures that assess an individual’s overall perception of their health may serve as a useful adjunct when predicting frailty. Frailty utilizes the physiologic, environmental, and behavioral characteristics of an individual to determine their vulnerability to an adverse event. Frailty scores determined by these characteristics can be used to estimate the physiologic status of an individual and are key in identifying those most at risk. Frailty has been identified as a major predictor of poor outcomes and has been incorporated as a prognostication tool for an adverse event in the acute injury setting (19-23). The older population is particularly vulnerable to traumatic injury. Fall-related injuries represent a significant portion of traumatic injury experienced by older adults (24-26). Fall related injury and frailty are inter-related. Frailty itself may help better identify those more at risk for fall injury, but fall injuries can also be considered an adverse event secondary to frailty (27). For those individuals who sustain a fall injury, their frailty may worsen and lead to increasing recidivism with repeat traumatic injuries related to falls or other adverse events. The fear of falling itself can also lead to decreased activity, thereby increasing frailty. With the older population expected to double in size by the year 2060, the incidence of trauma in older adults is expected to increase considerably (28). There is a myriad of reasons for this, related to multiple dynamics that include the increasing amount of care needed for this rising population, increased use of facility assistance like nursing homes and assisted living facilities, increasing need to recognize therapy interactions like benzodiazepine use, limited staff, and geriatrician availability, and limited resources that increase the risk of traumatic injury in this population. Regardless, the rising population density of older individuals has resulted in an increased number of geriatric trauma overall. As a result, there is increasing interest in identifying models and scoring systems that better predict and identify older individuals most at risk for falls.
While frailty scores are typically assessed by providers, patient self-assessment by global health surveys may provide additional information on their risk burden at the time of presentation in the acute injury setting. Much like frailty assessments, self-reported measures of global health usually occur at the time of injury or admission and does not consider the long-term dynamics of patient management or rehabilitation as their course and disposition evolve over time. As a result, there is a paucity of literature examining global health measures in those with acute injury. The aim of this study was to evaluate measures of global health in acutely injured adults by VR-12 assessment, to assess whether components of the VR-12 changed over time, and whether VR-12 assessment correlated with frailty scores.



This prospective observational study was approved by our Institutional Review Board. After determining eligibility by assessing whether patients met the study’s inclusion and exclusion criteria, they were consented and observed for one year following hospital discharge. A written informed consent was obtained from all participants.

Inclusion and Exclusion Criteria

Patients included for study were those aged 65 years and older who presented with an acute injury and admitted to our institution’s burn, trauma, or emergency general surgery (EGS) services from December 2016 to June 2017. Patients that were non-English speakers, unable to participate in, discuss, or assess their own health and social situation, or those who did not have participating family members that could answer questions about a patient’s functional status or medical history were excluded from study.

Study Design

A questionnaire was provided to a patient or appropriate surrogate at enrollment and at 3, 6, 9, and 12 months intervals following discharge to collect information on demographics, general health and comorbidities, and objective information necessary for global health assessment. Patients completed questionnaires by mail, email, phone, or at scheduled follow up appointments. Data collected from the questionnaires was utilized to determine frailty using the Trauma Specific Frailty Index (TSFI) and VR-12 scoring. The TSFI is a validated fifteen variable assessment that considers patient attributes such as comorbidities, ability to perform daily activities, and general life attitudes to ultimately determine frailty (29). Each variable can score up to one point depending on the severity of disease and the total score is divided by 15 to determine a final score. The score can range from 0 to 1 with a score of 0.27 or higher designating frailty. Patients identified as frail by TSFI score have been shown to be more likely to have an unfavorable discharge disposition (29, 30). For VR-12 determination, PCS and MCS was produced for comparison purposes.

Demographic Data Collection

Medical records and surveys were reviewed for demographic information that included age, gender, race, co-morbidities, substance abuse history, occupation, height, and weight. Additional information pertaining to hospital course and management included length of stay, follow up and repeat admissions, any associated complications related to their hospital stay, nutritional status and assessments, participation with therapy, and any trauma-related data were collected.

Statistical Analysis

SAS statistical software, version 9.4 (SAS Institute, Cary, NC, USA) was used to analyze the data. Student’s t-test was used to differentiate differences between groups for continuous data. Simple linear regression was performed for PCS and MCS at each time point measured with respect to TSFI categorical frailty. Pearson correlation coefficients were utilized to examine the relationship between the continuous variables (age, PCS, MCS, and TSFI) at each time point. Multivariate logistic regression was used to evaluate for an association of the VR-12 components, PCS and MCS, and frailty as determined by TSFI at enrollment. Age was included as a covariate to control for this variable in this analysis. To account for correlation within each subject, generalized estimating equations were used to estimate parameters and the covariance matrix estimated with a robust sandwich estimator. An independence structure was assumed for the working correlation matrix. Age was centered at 65. Weeks post-discharge was modeled as a continuous variable as were all frailty metrics. Models were fit for each frailty metric using Proc Genmod. Frail patients were defined as patients with a TSFI score of 0.27 or greater. P < 0.05 was considered significant. A linear marginal model with Gaussian errors was used to relate each VR-12 metric to time expressed in months as a categorical variable. To account for the correlation of observations from the same subject, the correlation structure was modeled. Several correlation structures (unstructured, compound symmetry, autoregressive, and Toeplitz) were evaluated for each frailty metric and model fit compared using Akaike Information Criterion (AIC) (31, 32). The correlation structure yielding the lowest AIC value was used to fit the final model. Models were fit using Proc Mixed in SAS version 9.4. Subjects lost to follow up were not included in this analysis.




Overall, fifty-eight patients were enrolled for study, of which 31 were trauma patients, 19 were EGS patients, and eight were burn surgery patients. Of the 31 trauma patients, 25 (80.7%) suffered a fall, four (12.9%) a motor vehicle accident, one suffered an ATV accident, and one was struck by someone falling from height. Patients were prospectively followed for the duration of their hospital course and for one year after hospital discharge. One patient was dropped from analysis after it was found that they did not have enrollment frailty metrics. A second patient was excluded from the analysis as they did not have any PCS or MCS measurements. The median TSFI score was 0.18 (median range 0.1–0.67). Eighteen patients were identified as frail by TSFI score. The most common past medical history findings were cardiovascular disease (67.2% of patients), cancer (24.1%), and diabetes (24.1%). Six patients in our cohort had previously sustained a stroke while two patients had a history of Parkinson’s Disease and one patient had a history of dementia noted at enrollment. Demographics data is summarized in Table 1.

Table 1. Demographic characteristics of the study participants (n = 58)

Abbreviations: BMI – Body Mass Index; GCS – Glasgow Coma Scale; ISS – Injury Severity Score; LOS – Length of Stay; Median was used for qualitative variables


When grouping patients based on mortality, 8 patients died and 44 patients lived, no significant differences in frailty scoring, age, and PCS (p = 0.97) or MCS (p = 0.78) at enrollment were observed between patients who ultimately lived and those who died.

Analysis of PCS and MCS Scoring

PCS and MCS scores were evaluated to assess differences at various time points and changes over time. Fifty-six patients were included in the analysis of PCS and MCS scoring as two patients did not have any scores recorded. An overall downtrend in overall mean PCS from 47.59 ± 13.71 to 42.67 ± 13.79 was noted from the time of enrollment up to 12 months following hospital discharge. Significant differences in PCS were noted between frail patients and non-frail patients at enrollment (31.91 ± 10.88 vs. 50.44 ± 13.10, p = 0.03) and at 12 months following hospital discharge (26.51 ± 11.48 vs. 47.43 ± 12.54, p = 0.03; Table 2). In comparison, the overall mean MCS remained relatively stable over the course of the study period. Significant differences were noted in MCS between frail and non-frail patients at enrollment only (38.91 ± 6.16 vs. 42.55 ± 5.27, p = 0.04; Table 2).

Table 2. Comparison of PCS and MCS for All Patients and By Frailty

Abbreviations: MCS – Mental Component Score; PCS – Physical Component Score; TSFI – Trauma Specific Frailty Index; Frailty was defined by TSFI with score greater than 0.27; p <0.05 was considered significant; Frailty was defined by TSFI score greater than 0.27; p < 0.05 was considered significant)


A pairwise comparison of PCS and MCS was then conducted between time points. In direct comparison of time points for PCS, no significant differences were noted between enrollment and the 3 month (p = 0.78), 6 month (p = 0.42), and 9 month (p = 0.34) time points. However, PCS was significantly different at 12 months compared to enrollment (p = 0.03). No other significant differences were noted in further comparisons of time points within the study period. Pairwise comparison of MCS showed significant differences between enrollment and the 6 month (p < 0.01), 9 month (p < 0.01), and 12 month (p < 0.01) time points. Similarly, no significant differences were noted with any other time point comparisons. Pairwise comparisons of PCS and MCS are summarized in Table 3 and Table 4, respectively.

Table 3. Pairwise Comparisons of PCS Between Time Points

Abbreviations: MCS – Mental Component Score; PCS – Physical Component Score; TSFI – Trauma Specific Frailty Index; Frailty was defined by TSFI with score greater than 0.27; p <0.05 was considered significant; Frailty was defined by TSFI score greater than 0.27; p < 0.05 was considered significant)


Table 4. Pairwise Comparisons of MCS Between Time Points

Abbreviations: MCS – Mental Component Score; p < 0.05 was considered significant.

Correlation of PCS and MCS with Frailty

The data was then assessed for associations between PCS and MCS with frailty using Pearson correlation. At enrollment, TSFI significantly correlated with PCS (r = -0.72, p < 0.01), MCS (r = -0.35, p < 0.01), and age (r = 0.34, p = 0.02). The correlation with frailty continued only for PCS and age at 3 months (PCS: r = -0.64, p < 0.01; age: r = 0.40, p = 0.02), 6 months (PCS: r = -0.79, p < 0.01; age: r = 0.34, p = 0.03), 9 months (PCS: r = -0.75, p < 0.01; age: r = 0.35, p = 0.03), and 12 months (PCS: r = -0.79, p < 0.01; age: r = 0.19, p = 0.23) following enrollment. MCS only correlated with age at the 12 month time point (r = 0.32, p = 0.04).
A linear regression model was performed using PCS and MCS at enrollment with age as a covariable and enrollment TSFI was the dependent variable. Age (p < 0.01), PCS (p < 0.01), and MCS (p < 0.01) were all independent predictors of enrollment frailty (Table 5). Univariate and multivariate analysis showed an association between PCS and age with frailty. MCS was not associated with frailty on multivariate analysis. PCS (OR = 0.894 [0.84-0.95], p = 0.0004) and age (OR = 1.113 [1.012-1.223], p = 0.03) independently predicted enrollment frailty. The area under the curve for this model was 0.89.

Table 5. Analysis of Maximum Likelihood Parameter Estimates

Abbreviations: PCS – Physical Component Score; MCS – Mental Component Score; p <0.05 was considered significant.



Our data shows that quality of life measures that assess a patient’s overall perception of their health significantly correlated with frailty in this cohort. In addition to clinician frailty assessment, the information provided by these measures may further inform healthcare providers on the degree of risk burden their patients have for repeat falls or other injury following their initial injury. Others have examined the use and accuracy of patient reported global health measures as a link to clinical disorders. Lapin et al. assessed the accuracy of general health cross-walk tables in a clinical sample of patients with spine disorders using VR-12 for general health evaluation. They noted that VR-12 could be accurately linked within a sample of patients with spine disorders but noted high bias and low precision (2). Further combinations with other global health scores have strengthened that linkage. Schalet et al. found that the combined set of VR-12 and PROMIS global health measures were linkable. They noted that linking worked better between physical and mental health scores using VR-12 item responses than with linkage based on algorithmic scores (18).
Much like frailty assessments, measures of global health typically occur at the time of injury or admission but does not consider the long-term dynamics of patient management or rehabilitation as their course and disposition change over time. As a result, most literature typically utilizes an initial evaluation of either frailty or global health measures to determine outcomes. This study presents a unique approach to patient reported global health assessment as well as its association with frailty. In our cohort, 18 patients were identified as frail as defined by the TSFI. An overall downtrend was observed in PCS scores following admission and up to twelve months after. Additionally, significant differences were observed in PCS scores between patients identified as frail and those who were not. This significant difference was present twelve months after as well. MCS scores, in comparison, remained steady throughout the survey period. Our data additionally noted significant correlations between frailty, PCS, and MCS. This significance continued following enrollment for PCS. These data suggest a relationship between PCS and frailty and that repeated assessment of an individual’s physical component score over time may be crucial to determining risk burden or those more at risk for additional injury following acute injury.
Identifying an individual’s risk for additional injury easily and observing significant changes over time through the use of frequent reassessments is crucial to developing prevention and clinical intervention strategies. Prospective interventions in frail older populations consisting of delirium prevention, nutritional and geriatrician consultation, physical therapy, early ambulation, and pain control have demonstrated significant decreases in 30-day readmission and delirium compared to no intervention in frail patients (33). Increasing the frequency of patient reported global health measures may help identify changes in both frailty and clinical progress earlier and may lead to more rapid intervention.
This study presents several limitations. It is a prospective initial analysis of patient reported global health measures and its association with frailty in acutely injured older patients. It should be noted that this is a single center study performed on a mainly rural, Caucasian population in the Midwest and, as such, these findings may not be generalizable to other populations. While this study provides significant observations to consider in the course of frailty assessment and the study and management of frail patients, the study itself remains underpowered. A higher-powered study over an even longer period of time could tease out associations and outcomes heretofore unidentified. Furthermore, while this study encompasses older patients with acute injury, the diversity of injury types sustained required trauma surgery, emergency general surgery, or burn surgery management. Older patients who sustain a fall will require different management and rehabilitation compared to older patients who sustain large total body surface area burns. Each injury type may affect patient reported global health measures differently over time and, as such, must be considered on an individual basis in future studies. Additionally, the subjectiveness of the questionnaire must be considered when assessing patient frailty and risk burden. While TSFI considers degree of dementia in its frailty assessment, global health measures like the VR-12 are dependent on patient self-assessment and may be unreliable depending on a patient’s medical history and clinical course.
Despite these limitations, this study provides a novel look at global health measures as a metric to consider when assessing frailty in acutely injured older patients. It is further bolstered by its methodology whereupon repeated assessments are performed over a period of time. This methodology may prove useful in more quickly and accurately determining physiologic changes that require intervention.


Conclusions and Future Directions

PCS scores were strongly associated with frailty. Our results suggest that these global measures of health could be utilized in addition to or in lieu of frailty scores when assessing older individuals. Further studies are warranted to confirm this association.

Plain summary

This study provides a novel look at global health measures as a metric to consider when assessing frailty in acutely injured older adults. Global health measures such as the Veterans RAND 12 Item Health Survey (VR-12) are quality of life measures that assess an individual’s overall perception of their health and may serve as a useful adjunct when predicting frailty. Our results show that the physical component score (PCS) produced by VR-12 surveys was strongly associated with frailty. These global measures of health could be utilized in addition to or in lieu of frailty scores when assessing patients in the clinical setting.


Acknowledgements: We thank Ella Born for assisting with follow up phone calls to participants.

Conflict of Interest: The authors have no conflicts.

Sources of funding: This research was supported by the University of Iowa Injury Prevention Research Center and funded in part by grant # R49 CE002108-05 of the National Center for Injury Prevention and Control/CDC. This project was further supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through grant number UL1 TR001860 and UL1 TR002537. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Sponsor’s Role: The sponsor had no role in the design, methods, subject recruitment, data collection, and preparation of the manuscript.

Ethics Approval: This prospective observational study was approved by the University of Iowa Institutional Review Board (IRB # 201611731).

Consent to participate: A written informed consent was obtained from all participants.

Consent for publication: not applicable.

Availability of data and material: Data are available upon request to the corresponding author.

Author Contributions: All authors have made equal contributions to the conception, design, analysis, interpretation, drafting, and revision of the manuscript for this study.



1. Ahmed S, Berzon RA, Revicki DA, Lenderking WR, Moinpour CM, Basch E, et al. The Use of Patient-reported Outcomes (PRO) Within Comparative Effectiveness Research: Implications for Clinical Practice and Health Care Policy. Medical Care, 2012. 50(12): 1060-70.
2. Lapin BR, Kinzy TG, Thompson NR, Krishnaney A, and Katzan IL. Accuracy of Linking VR-12 and PROMIS Global Health Scores in Clinical Practice. Value Health, 2018. 21(10): 1226-33.
3. Bjorner J, Fayers P, and Idler E, Self-rated health. Assessing Quality of Life in Clinical Trials: Methods and Practice 2nd edn (ed Fayers,P. M.;Hays,R. D.). 2005, Oxford: Oxford University.
4. Han PK, Lee M, Reeve BB, Mariotto AB, Wang Z, Hays RD, et al. Development of a prognostic model for six-month mortality in older adults with declining health. J Pain Symptom Manage, 2012. 43(3): 527-39.
5. Brazier J, Roberts J, and Deverill M. The estimation of a preference-based measure of health from the SF-36. J Health Econ, 2002. 21(2): 271-92.
6. Eichler H, Kong S, Gerth W, Mavros P, and Jonsson B. Use of Cost-Effectiveness Analysis in Health-Care Resource Allocation Decision-Making: How Are Cost-Effectiveness Thresholds Expected to Emerge? Value in Health Value in Health, 2004. 7(5): 518-28.
7. Kazis LE, Bierman A, Diehr P, Hays R, Holzman J, Iqbal S, et al. A White Paper report commissioned by CMS with Recommendations for the Medicare Health Outcomes Survey. 2007.
8. Selim AJ, Rogers W, Qian SX, Brazier J, and Kazis LE. A preference-based measure of health: the VR-6D derived from the veterans RAND 12-Item Health Survey. Qual Life Res, 2011. 20(8): 1337-47.
9. Neumann PJ, Rosen AB, and Weinstein MC. Medicare and cost-effectiveness analysis. N Engl J Med, 2005. 353(14): 1516-22.
10. Ware J, Kosinski M, and Gandek B, SF-36 Health Survey: Manual and Interpretation Guide. 2005, Lincoln, RI: Quality Metric Inc.
11. Garratt AM, Ruta DA, Abdalla MI, and Buckingham JK. The SF 36 Health Survey Questionnaire: An Outcome Measure Suitable for Routine Use Within the NHS? British Medical Journal, 1993. 306(6890): 1440.
12. Revicki DA and Kaplan RM. Relationship between psychometric and utility-based approaches to the measurement of health-related quality of life. Quality of life research: An international journal of quality of life aspects of treatment, care and rehabilitation, 1993. 2(6): 477-87.
13. Ware J, Kosinski M, and Keller SD. A 12-Item Short-Form Health Survey: Construction of Scales and Preliminary Tests of Reliability and Validity. Medical Care Philadelphia, 1996. 34(3): 220-33.
14. Kazis LE, Miller DR, Skinner KM, Lee A, Ren XS, Clark JA, et al. Applications of Methodologies of the Veterans Health Study in the VA Healthcare System: Conclusions and Summary. Journal of Ambulatory Care Management, 2006. 29(2): 182-8.
15. Selim AJ, Rogers W, Fleishman JA, Qian SX, Fincke BG, Rothendler JA, et al. Updated U.S. population standard for the Veterans RAND 12-item Health Survey (VR-12). Quality of Life Research, 2009. 18(1): 43-52.
16. Kazis LE, Lee A, Spiro A, Rogers W, Ren XS, Miller DR, et al. Measurement Comparisons of the Medical Outcomes Study and Veterans SF-36 Health Survey. Health Care Financing Review, 2004. 25: 43-58.
17. Kazis LE, Miller DR, Clark JA, Skinner KM, Lee A, Ren XS, et al. Improving the Response Choices on the Veterans SF-36 Health Survey Role Functioning Scales Results From the Veterans Health Study. Journal of Ambulatory Care Management Journal of Ambulatory Care Management, 2004. 27(3): 263-80.
18. Benjamin D, Nan E, Ron D, Lewis E, Karon F, Joshua P, et al. Linking Physical and Mental Health Summary Scores from the Veterans RAND 12-Item Health Survey (VR-12) to the PROMISE Global Health Scale. Journal of General Internal Medicine, 2015. 30(10): 1524-30.
19. Joseph B, Hassan A, and Pandit V. Frailty in trauma: A systematic review of the surgical literature for clinical assessment tools. J Trauma Acute Care Surg, 2016. 81(4): 805.
20. Maxwell CA, Mion LC, Mukherjee K, Dietrich MS, Minnick A, May A, et al. Preinjury physical frailty and cognitive impairment among geriatric trauma patients determine postinjury functional recovery and survival. J Trauma Acute Care Surg, 2016. 80(2): 195-203.
21. Clegg A and Young J. The frailty syndrome. Clin Med Clinical medicine (London, England), 2011. 11(1): 72-5.
22. Lang P, Michel J, and Zekry D. Frailty Syndrome: A Transitional State in a Dynamic Process. Gerontology, 2009. 55(5): 539-49.
23. Morley JE, Vellas B, Abellan van Kan G, Anker SD, Bauer JM, Bernabei R, et al. Frailty Consensus: A Call to Action. Journal of the American Medical Directors Association, 2013. 14(6): 392-7.
24. Curtis E, Romanowski K, Sen S, Hill A, and Cocanour C. Frailty score on admission predicts mortality and discharge disposition in elderly trauma patients over the age of 65 y. J Surg Res, 2018. 230: 13-9.
25. Moschny A, Platen P, Klaaflen-Mielke R, Trampisch U, and Hinrichs T. Barriers to physical activity in older adults in Germany: A cross-sectional study. International Journal of Behavioral Nutrition and Physical Activity, 2011. 8: 121.
26. Rasinaho M, Hirvensalo M, Leinanen R, Lintunen T, and Rantanen T. Motives for and Barriers to Physical Activity Among Older Adults With Mobility Limitations. Journal of Aging and Physical Activity, 2007. 15(1): 90.
27. Hatcher VH, Galet C, Lilienthal M, Skeete DA, and Romanowski KS. Association of Clinical Frailty Scores With Hospital Readmission for Falls After Index Admission for Trauma-Related Injury. JAMA Netw Open, 2019. 2(10): e1912409.
28. Galet C, Zhou Y, Eyck PT, and Romanowski KS. Fall injuries, associated deaths, and 30-day readmission for subsequent falls are increasing in the elderly US population: A query of the WHO mortality database and National Readmission Database from 2010 to 2014. Clin Epidemiol, 2018. 10: 1627-37.
29. Joseph B, Pandit V, Zangbar B, Kulvatunyou N, Tang A, O’Keeffe T, et al. Validating trauma-specific frailty index for geriatric trauma patients: a prospective analysis. J Am Coll Surg, 2014. 219(1): 10-7 e1.
30. Maxwell CA, Patel MB, Suarez-Rodriguez LC, and Miller RS. Frailty and Prognostication in Geriatric Surgery and Trauma. Clin Geriatr Med, 2019. 35(1): 13-26.
31. Akaike H. New Look at Statistical-Model Identification. IEEE Transactions on Automatic Control, 1974. AC19(6): 716-23.
32. Akaike H. Information theory and an extension of the maximum likelihood principle. 2nd International Symposium on Information Theory (Akademia Kiado, Vudapest), 1973: 267-81.
33. Bryant EA, Tulebaev S, Castillo-Angeles M, Moberg E, Senglaub SS, O’Mara L, et al. Frailty Identification and Care Pathway: An Interdisciplinary Approach to Care for Older Trauma Patients. J Am Coll Surg, 2019. 228(6): 852-9 e1.




M. Cesari1,2, R. Calvani3,4, M. Canevelli5,6, I. Aprahamian7,8, P. de Souto Barreto9,10, D. Azzolino1,2, R.A. Fielding11, N. Vanacore5, M. Inzitari12,13, E. Marzetti3,14


1. Department of Clinical Sciences and Community Health, University of Milan, Milan, Italy; 2. Healthy Aging Lab, IRCCS Istituti Clinici Scientifici Maugeri, Milan, Italy;
3. Fondazione Policlinico Universitario «Agostino Gemelli» IRCCS, Rome, Italy; 4. Aging Research Center, Department of Neurobiology, Care Sciences and Society, Karolinska Institutet and Stockholm University, Stockholm, Sweden; 5. National Centre for Disease Prevention and Health Promotion, Italian National Institute of Health, Rome, Italy; 6. Department of Human Neuroscience, Sapienza University of Rome, Rome, Italy; 7. Geriatrics Division, Internal Medicine Department, Jundiaí Medical School, Jundiaí, Brazil; 8. University of Groningen, University Medical Center Groningen, Groningen, The Netherlands; 9. Gérontopôle de Toulouse, Institut du Vieillissement, Centre Hospitalo-Universitaire de Toulouse, Toulouse, France; 10. CERPOP INSERM 1295, University of Toulouse III, INSERM, UPS, Toulouse, France; 11. Nutrition, Exercise Physiology, and Sarcopenia Laboratory, Jean Mayer USDA Human Nutrition Research Center on Aging, Tufts University, Boston, MA, USA; 12. RE-FIT Barcelona research group, Parc Sanitari Pere Virgili and Vall d’Hebron Institute of Research (VHIR), Barcelona, Spain; 13. Department of Health Sciences, Universitat Oberta de Catalunya (UOC), Barcelona, Spain; 14. Università Cattolica del Sacro Cuore, Institute of Internal Medicine and Geriatrics, Rome, Italy.

Corresponding Author: Matteo Cesari, MD, PhD. IRCCS Istituti Clinici Scientifici Maugeri; Via Camaldoli 64, 20138 Milan, Italy. Email: macesari@gmail.com; Twitter: @macesari

J Frailty Aging 2021;in press
Published online May 24, 2021, http://dx.doi.org/10.14283/jfa.2021.23


Key words: Research, methodology, publishing, medicine, science.


From the beginning of 2020, the world has been fighting the SARS-Cov-2 outbreak. The life of each one of us has profoundly changed. Unavoidably, our clinical routine has drastically modified in its priorities and methodologies (1). The COVID-19 pandemic has also raised significant issues in the field of research. The investigators’ responsibility has increased with the need to thoughtfully weigh the risk-benefit ratio for each protocol in an emergency and evolving scenario (2).
Many projects that had started in the pre-COVID-19 era were halted in the past months, mainly because of 1) the restrictions that governments put in place to limit the spread of the Coronavirus, 2) the inadequacy/inappropriateness of the research activities as planned before the pandemic, and/or 3) the participants’ reluctancy at attending the scheduled visits in facilities usually located within the hospital premises. Some investigators managed to redesign their study methods and make them compatible with the new situation. Someone also proposed possible solutions to tackle the most common issues of the “new” research routine (3). Others faced challenging (if not insurmountable) stalls to solve and eventually decided to stop their studies. Indeed, it was not rare the closure of research activities in some countries, especially during the early period of the pandemic.
How should we judge the findings generated by studies that suffered the effects of the COVID-19 pandemic? In principle, the modification of a protocol “on the way” or the premature ending of a study are deviations that, in “normal” times, would be flagged as major methodological flaws. Purists of research methodology may indeed find these deviations hard to accept. But, can we judge today’s research activities without considering the new context of COVID-19 and the global consequences we all have been experiencing? What degree of flexibility should we apply when reviewing an article containing issues clearly due to the pandemic? In this context, it is also essential to consider the increasing interest around pragmatic research for achieving a wise compromise between the potentially decontextualized findings from rigid methodologies and the heterogenous “real life” scenarios where the study results will be applied (4). Perhaps, the pandemic may lead us to introduce sufficient doses of pragmatism and creativity as core elements of future research. For sure, scientific projects will have to include backup plans considering the possibility of postponing/modifying recruitment strategies, maintaining protocol adherence despite environmental changes, and designing alternative strategies for intervention delivery.
In this issue of The Journal of Frailty & Aging, Tavoian and colleagues (5) present the results of a pilot study initiated before the COVID-19 outbreak. The trial was focused on exploring whether stationary bicycle high-intensity interval training was a more efficient standalone exercise strategy to improve cardiovascular and muscular function compared with resistance or aerobic training in older adults. Unfortunately, the restrictions imposed by the emergent pandemic situation led to a premature stop of the program, precluding the achievement of the expected sample size and leaving the researchers with partial results.
The data could appear insufficient to convey a complete and clear message (at least coherent with the original study aims). For example, the unbalanced representation of women in the study sample (11 vs. only 3 men) might already render the findings of difficult interpretation. On the other hand, we might say that something is always better than nothing! Those partial results could still feed the discussion in the scientific community and serve to support further steps in the field (6). The Chinese philosopher Lao Tzu once said: “The journey of a thousand miles begins with one step”.
The study by Tavoian and colleagues was well-presented and conducted rigorously, methodologically speaking. It also had its record on clinicaltrials.gov, having been registered before the initiation of recruitment. The topic was of interest and novel. Last but not least, the trial was framed as a pilot study with the primary aim of collecting preliminary data for future larger-scale activities. Under these premises, it could have been unfair to reject it simply because an external event (i.e., the COVID-19 pandemic) had disrupted the researchers’ original plans. Are we sure that what is (partially) presented by Tavoian and colleagues may not be sufficient to address questions raised by other researchers? After all, when a study is well-conducted, it is always important to publish it, independently of issues that might have affected its conduct. Of course, the expected impact of an unforeseen event disrupting a pilot study is different from that occurring in a phase 3 trial. This implies that a case-by-case approach is crucial but needs to be structured on specific criteria. Otherwise, the risk could be to fall into a dangerous level of subjectivity or to reject everything in search of a utopic perfection.
No study is perfect, and every research has an own value. It is a matter of clearly and fairly presenting the experience, constructively discussing the strengths and weaknesses. At the same time, the readers should be transparently informed of the deviations imposed by the COVID-19 outbreak and provided with all the information necessary to judge the value of findings and discern potentially relevant results from “background noise”. The need for clarity and details is important for discriminating vague justifications (e.g., “due to the pandemic”) from founded decisions based on political, sanitary, and/or scientific choices (2).
It is also noteworthy that the numbers we report in our tables are persons who have trusted in our research and devoted part of their time to it. The restitution of the study results has to be considered an ethical duty of the researcher in front of the participant. Consistently, many investigations benefit from funds allocated by private or public institutions for specific research purposes and potentially taken from other priorities. It is, thus, also a matter of financial clarity and accountability behavior.
With the aim of providing the correct legitimacy to those valuable research activities critically affected by the COVID-19 pandemic, it has been felt that a set of standardizing criteria could be helpful. Table 1 presents aspects that might support clinicians, researchers and policymakers in better judging the quality of the information provided by partial scientific results.
The post-COVID-19 world we live in is putting us in the position of reconsidering many aspects of our (professional) lives. Different priorities are today modifying our decisions. How we conduct research today is not the same as before COVID-19. The population we are studying is not the same either, and its values and needs might be different (7). Some adaptations of research to the new scenario (e.g., more extensive use of technologies, higher involvement of participants in collecting their data) are necessary to continue advancing in research and avoiding that what was initiated before COVID-19 could go lost.

Table 1. Specific criteria to additionally consider when judging if an article is still worth being published despite COVID-19-related deviations from original plans


Conflicts of interest: MC has received honoraria from Nestlè Health Sciences for presentations at scientific meetings and serving as a member of Expert Advisory Boards. MI has received honoraria from Nestlé Health Sciences for serving as an expert advisor. EM has received honoraria from Abbott, Nestlè, Nutricia, and Thermofisher for presentations at scientific meetings. RAF received grants and personal fees from Nestlé. RAF also reports grants from National Institutes of Health (National Institute on Aging), during the conduct of the study; grants, personal fees and other from Axcella Health, other from Inside Tracker, grants and personal fees from Biophytis, grants and personal fees from Astellas, personal fees from Cytokinetics, personal fees from Amazentis, personal fees from Glaxo Smith Kline, personal fees from Juvicell, outside the submitted work. No conflict of interest reported by the other authors.

Acknowledgments: RAF is partially supported by the US Department of Agriculture (USDA), under agreement No. 58-8050-9-004 and by NIH Boston Claude D Pepper Center (OAIC; 1P30AG031679). Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the USDA.




1. Astrone P, Cesari M. Integrated Care and Geriatrics: A Call to Renovation from the COVID-19 Pandemic. J Frailty Aging. 2021;10(2):182-183. doi: 10.14283/jfa.2020.59
2. Fleming TR, Labriola D, Wittes J. Conducting Clinical Research During the COVID-19 Pandemic: Protecting Scientific Integrity. JAMA. 2020;324(1):33-34. doi:10.1001/jama.2020.9286
3. Kroenke K, Bair MJ, Sachs GA. Continuing Research During a Crisis. J Gen Intern Med. 2021;36(4):1086-1088. doi:10.1007/s11606-021-06636-5
4. Ford I, Norrie J. Pragmatic Trials. N Engl J Med. 2016;375(5):454-463. doi:10.1056/NEJMra1510059
5. Tavoian D, Russ DW, Law TD, et al. Effects of three different exercise strategies for optimizing aerobic capacity and skeletal muscle performance in older adults: a pilot study. J Frailty Aging. doi:10.14283/jfa.2021.21
6. Karmakar S, Dhar R, Jee B. Covid-19: research methods must be flexible in a crisis. BMJ. 2020:m2668. doi:10.1136/bmj.m2668
7. Palmer K, Monaco A, Kivipelto M, et al. The potential long-term impact of the COVID-19 outbreak on patients with non-communicable diseases in Europe: consequences for healthy ageing. Aging Clin Exp Res. 2020;32(7):1189-1194. doi:10.1007/s40520-020-01601-4



M. Gagesch1,2, P.O. Chocano-Bedoya2,3, L.A. Abderhalden1,2, G. Freystaetter1,2, A. Sadlon1,2, J.A. Kanis4, R.W. Kressig5, S. Guyonnet6, J.A. P. DaSilva7, D. Felsenberg8, R. Rizzoli9, M. Blauth10, E.J. Orav11, A. Egli2, H.A. Bischoff-Ferrari1,2,12


1. Department of Geriatrics, University Hospital Zurich, Zurich, Switzerland; 2. Centre on Aging and Mobility, University of Zurich, Zurich, Switzerland; 3. Institute of Primary Health Care (BIHAM), University of Bern, Bern, Switzerland; 4. Mary McKillop Institute for Health Research, Australian Catholic University, Melbourne, Australia and Centre for Metabolic Bone Diseases, University of Sheffield Medical School, Sheffield, UK; 5. University Department of Geriatric Medicine FELIX PLATTER, Basel, Switzerland; 6. Gérontopôle, Department of Geriatrics, CHU Toulouse, Toulouse, France; 7. Institute for Clinical and Biomedical Research (i.CBR), Faculty of Medicine, University of Coimbra, Department of Rheumatology, Centro Hospitalar e Universitário de Coimbra, Coimbra, Portugal; 8. Centre for Muscle and Bone Research, Department of Radiology, Charité Universitätsmedizin Berlin, Berlin, Germany (Prof. Felsenberg is deceased); 9. Service of Bone Diseases, Geneva University Hospitals and Faculty of Medicine, Geneva, Switzerland; 10. University of Innsbruck, Innsbruck, Austria; 11. Department of Medicine, Harvard Medical School, Boston, MA, USA; 12. University Clinic for Acute Geriatric Care, City Hospital Waid and Triemli, Zurich, Switzerland

Corresponding Author: Michael Gagesch, MD, Department of Geriatrics, University Hospital Zurich, Ramistrasse 100, 8091 Zurich, Switzerland, Email: michael.gagesch@usz.ch, https://orcid.org/0000-0003-3089-5768

J Frailty Aging 2021;in press
Published online May 3, 2021, http://dx.doi.org/10.14283/jfa.2021.18



BACKGROUND: Frailty is a geriatric syndrome associated with multiple negative health outcomes. However, its prevalence varies by population and instrument used. We investigated frailty and pre-frailty prevalence by 5 instruments in community-dwelling older adults enrolled to a randomized-controlled trial in 5 European countries.
METHODS: Cross-sectional baseline analysis in 2,144 DO-HEALTH participants recruited from Switzerland, Austria, France, Germany, and Portugal with complete data for frailty. Frailty status was assessed by the Physical Frailty Phenotype [PFP], SOF-Frailty Index [SOF-FI], FRAIL-Scale, SHARE-Frailty Instrument [SHARE-FI], and a modified SHARE-FI, and compared by country, age, and gender. Logistic regression was used to determine relevant factors associated with frailty and pre-frailty.
RESULTS: Mean age was 74.9 (±4.4) years, 61.6% were women. Based on the PFP, overall frailty and pre-frailty prevalence was 3.0% and 43.0%. By country, frailty prevalence was highest in Portugal (13.7%) and lowest in Austria (0%), and pre-frailty prevalence was highest in Portugal (57.3%) and lowest in Germany (37.1%). By instrument and overall, frailty and pre-frailty prevalence was highest based on SHARE-FI (7.0% / 43.7%) and lowest based on SOF-FI (1.0% / 25.9%). Frailty associated factors were residing in Coimbra (Portugal) [OR 12.0, CI 5.30-27.21], age above 75 years [OR 2.0, CI 1.17-3.45], and female gender [OR 2.8, CI 1.48-5.44]. The same three factors predicted pre-frailty.
CONCLUSIONS: Among relatively healthy adults age 70 and older enroled to DO-HEALTH, prevalence of frailty and pre-frailty differed significantly by instrument, country, gender, and age. Among instruments, the highest prevalence of frailty and pre-frailty was documented by the SHARE-FI and the lowest by the SOF-FI.

Key words: Frailty, community-dwelling, prevalence, clinical trials, epidemiology.



Frailty is an age-related medical syndrome of reduced functional capacity and one of the most burdensome conditions for the growing segment of the older population, and society as a whole (1). Frailty is often preceded by sarcopenia (2), the age-associated loss of muscle mass and function and both concepts overlap, e.g. in regard to reduced muscle strength, physical function (e.g. gait speed), and unintentional weight loss (3). In addition, frailty and sarcopenia are often promoted by factors such as malnutrition (4, 5) and multimorbidity (6, 7).
Over the last two decades, frailty has been identified as a stronger predictor of acute-care complications, length of stay, and hospital readmission than multimorbidity or chronological age alone (8-11). E.g. in a 2016 systematic review and meta-analysis of 31 prospective studies, Vermeiren et al. found that frailty increases the likelihood of mortality by 1.8-2.3-fold, hospital admissions by 1.2-1.8-fold, incident functional impairment related to basic activities of daily living by 1.6-2.0, and falls and fractures by 1.2-2.8-fold (12).
The development of frailty in older adults should be recognized as a dynamic process including transitions from robust, to pre-frail and to frail and vice versa over time (13). Therefore, early detection of the at-risk group of pre-frail individuals appears as a window of opportunity for the prevention of overt frailty, and perhaps prevention of sarcopenia as well (3, 14, 15).
While the identification of frailty is becoming increasingly recognized as important, the choice of frailty instrument might be guided by medical practice within a specialty, the population, setting, and outcome under investigation (16). Frailty is actively screened for in various clinical settings, including oncology and surgical specialties. However, the lack of consensus for defining frailty still hinders the comparison of frailty prevalence between different populations (1, 17-19).
Over time, the concept of the Physical Frailty Phenotype (PFP), introduced by Fried and colleagues in 2001 (20) and the deficit accumuation approach (i.e. Frailty Index) introduced by Mitnitski and Rockwood (21, 22) have emerged as the most frequently cited ways to operationalize frailty (23). While the Frailty Index approach scores accumulating deficits in age-associated variables covering multipe body systems (24), the five components of the PFP (fatigue, weight loss, slowness, weakness, and low activity level) aggregate to a clinical syndrome (20). Subsequently, many researchers have modified the PFP in order to accommodate variables available from research studies’ data sets (25). Among those, the Study of Osteoporotic Fractures (SOF)-Frailty Index (26), a condensed variation of only 3 items, and the FRAIL-Scale (27), a self-report based 5-item screening tool substituting the low activity criterion with multimorbidity (defined as the presence of five or more chronic diseases) have been frequently cited.
To our knowledge, the prevalence of frailty and pre-frailty in older Europeans according to the PFP has not been studied in a single multi-national dataset with the exception of the Survey of Health Aging and Retirement in Europe (SHARE) database in 2009 and 2019 (28, 29). The reported frailty prevalence by the latter study in SHARE ranged from 3.0% (Switzerland) to 15.6% (Portugal), thus indicating an uneven distribution in the older European population, plus indicating a need for further studies on frailty and its assessment concepts in older Europeans.
DO-HEALTH is a clinical trial in relatively healthy community-dwelling participants age 70 years and older without major health events (cancer or myocardial infarction) in the 5 years prior to enrollment, sufficient mobility to come to the study centers without help, and intact cognitive function, recruited from 5 European countries (30). Examining the prevalence of frailty and pre-frailty in DO-HEALTH at baseline aims to contribute important insights to the participants’ clinical presentation upon enrolment, and also with regard to several primary endpoints of the study, including cardiovascular health and physical function (30).
In view of the variety of instruments derived from the PFP, and to estimate the variability of frailty prevalence based on the instrument used, we investigated the PFP as our primary measure and four alternative derived instruments in DO-HEALTH. However, one has to acknowledge that this is not a probabilisitc sample of the countries investigated: in addition to the DO-HEALTH eligibility criteria listed above, each countries sample consisted of older adults volunteering to participate in a clinical trial, and recruited according to the methods of each study site.
In addition to examine the variability in frailty according to instrument, we examine whether the prevalence of frailty and pre-frailty differs by country, age, and gender.


Materials and methods

This is a cross-sectional study taking advantage of the baseline examination of 2,144 of 2,157 participants of the DO-HEALTH clinical trial recruited from Switzerland (Zurich, Basel, Geneva), Austria (Innsbruck), France (Toulouse), Germany (Berlin), and Portugal (Coimbra), with complete data for frailty. As a multicenter randomized controlled trial (RCT), DO-HEALTH (clinical trials: NCT01745263) was designed to investigate the effect of Vitamin D3 and Omega 3-fatty acids supplementation and a simple home exercise program in a 2x2x2 factorial design over a three year follow-up addressing multiple endpoints associated with aging, including frailty (30).
Inclusion criteria for DO-HEALTH were age 70 years and older, a Mini-Mental Status Examination Score of at least 24 points (31), living in the community and being sufficiently mobile to visit the study center, i.e. being capable to walk 10 meters and getting in and out of a chair without help. In order to capture the broad scope of the trial’s endpoints, the participants were meticulously phenotyped with standardized assessments of multiple organ, cognitive, and physical function at each of the four whole-day clinical study visits (baseline and follow-up at 12, 24, and 36 months). Assessments followed a strict protocol with regular site visits for inter-center quality control, overseen by the primary investigator. Further details of the design, aims and scope, exclusion criteria, and primary endpoints of DO-HEALTH are described elsewhere (30).

Assessments of Physical Function and Comorbidities

For the assessment of physical function, we used the Short Physical Performance Battery (SPPB), a validated assessment tool for lower extremity function in older adults (32). The SPPB includes a gait speed test, a chair-rise test, and a balance test. Its three components each score 0–4, with 4 points indicating the highest level of performance, and add up to an overall score of maximum 12 points. The number of comorbidities was recorded by a self-administered comorbidity questionnaire (33).

Fried Physical Frailty Phenotype

The physical frailty phenotype (PFP) has been validated in community dwelling older adults from numerous countries and in many in- and outpatient settings by various operationalizations (17, 25, 34). We built our PFP model with variables available from the baseline DO-HEALTH dataset as our primary frailty assessment tool. Unintentional weight loss was defined as a self-reported loss of more than 5% of body weight over the prior 12 months. Fatigue was operationalized as a positive answer to the self-reported question: “In the last month, have you had too little energy to do things you wanted to?” from the SHARE study original questionnaire. For weakness, grip-strength was taken in kilopascal (kPa) from the best of three consecutive trials at the dominant hand using a Martin Vigorimeter (KLS Martin Group, Tuttlingen, Germany) (35). Across all countries and ages, we used cut points by the lowest quintile approach by gender as did Fried and colleagues in their landmark study, but for simplification, irrespective of BMI (20). Slowness was defined as a gait speed below 0.67 m/s and 0.7 m/s respectively, according to gender and height as in the original study by Fried et al (20). Low activity level was conceptualized as done earlier by Santos-Eggimann and colleagues analyzing the SHARE dataset (28). An answer of “less than once a week” to the question: “How often do you engage in activities that require a low or moderate level of energy such as gardening, cleaning the car, or doing a walk?” fulfilled the criterion. Pre-frailty was fulfilled with 1-2 positive criteria and frailty with ≥3 positive criteria.

SOF-Frailty Index

The Study of Osteoporotic Fractures-Frailty Index (SOF-FI), introduced by Ensrud et al. is a condensed three-item variant of the Fried PFP including only the variables for weakness, involuntary weight loss, and fatigue (26, 36). It has been validated in community dwelling participants in the MOBILIZE Boston Study (37). In DO-HEALTH, we used data from the chair-rise test of the Short Physical Performance Battery for the SOF-FI weakness criterion.[38] Weight loss was defined by a self-reported loss of more than 5% of body weight over the prior 12 months. Fatigue was defined by a positive answer to the question: “Do you feel full of energy?” from the Geriatric Depression Scale as in the original study (26). Pre-frailty was defined by the presence of one, and frailty by ≥2 positive criteria.


The FRAIL-Scale (Fatigue, Resistance, Ambulation, Illnesses, and Loss of Weight-Scale) consists completely of self-reported items (39). Since its introduction, it has been validated for the prediction of most relevant frailty associated negative outcomes in various settings (40-42). For building the FRAIL-Scale in DO-HEALTH, we used the same question for the fatigue criterion as described above for the PFP. For resistance (weakness) we used the question for difficulties climbing stairs from SHARE-FI (see below). For the ambulation criterion, we used the question: “Because of a health problem, do you have difficulty (expected to last more than 3 months) walking 100 meters”, also from SHARE-FI. Illnesses (i.e. multimorbidity) was defined by the presence of >5 diagnoses). Weight loss was defined by a self-reported loss of more than 5% of body weight over the prior 12 months. On the FRAIL-Scale, pre-frailty was defined by 1-2 positive criteria, and frailty by ≥3 positive criteria.

SHARE-Frailty Instrument (SHARE-FI1)

To match the available variables from DO-HEALTH, we used a marginally modified version of the original SHARE-Frailty Instrument (SHARE-FI), validated in community dwelling older adults and in primary care.[43-45] Weight loss was defined by a self-reported >5% loss of body weight over the prior 12 months. Fatigue was operationalized in the same way as for the PFP, by a positive answer to the question: “In the last month, have you had too little energy to do things you wanted to do?” For indicating low grip strength on Martin Vigorimeter readings, we used the lowest quintile approach as described for the PFP above. Slowness was defined as originally described by Santos-Eggimann et al. by a positive answer to either i) “Because of a health problem, do you have any difficulty walking 100 meters? (Exclude any difficulties that you expect to last less than three months.)”, or ii) “Because of a health problem, do you have any difficulty climbing one flight of stairs without resting? (Exclude any difficulties that you expect to last less than three months.)” (28). Low activity was fulfilled by a response of “less than once a week” (i.e. “one to three times a month” or “hardly ever or never”) to the same question as for the PFP: “How often do you engage in activities that require a low or moderate level of energy such as gardening, cleaning the car, or going for a walk?” Pre-frailty was defined as 1-2 positive criteria and frailty with ≥3 positive criteria. The SHARE-FI1 differs from our implementation of the PFP only in the way that slowness was assessed by self-report only.


For a fully self-report based variant of the SHARE-FI in DO-HEALTH, we substituted grip strength measurement by a positive answer to the question: “Because of a health problem, do you have any difficulty climbing one flight of stairs without resting (exclude any difficulties that you expect to last less than three months)?” This question was usually meant only to depict slowness in the original SHARE-FI operationalization. However, measuring resistance (or weakness) by a single question of difficulty walking one flight of stairs has been validated as a surrogate marker for weakness (46), as it is also used as the weakness criterion in the FRAIL-Scale introduced by Abellan van Kan et al. (see above) (39). For SHARE-FI2, pre-frailty was fulfilled with 1-2 positive criteria and frailty with ≥3 positive criteria.
Supplementary Table 1 provides an additional overview of the single components comprising the described instruments.

Statistical Analysis

Data is reported as means and standard deviations (SD) for continuous variables and frequencies and percentages for categorical variables. We estimated the prevalence of fraily and pre-frailty as the proportion of our study population who could be classified as either frail or pre-frail according to the PFP and the derived instruments and reported as frequencies and percentages. We then compared the prevalences by country, age sub-groups (<75y, ≥75y), and gender using a Chi-square (χ2) test or Fisher’s exact test when a frequency was less than five. We report on the general prevalence by screening tool including participants with a maximum of two missing values, as instruments indicate frailty as having ≥3 out of 5 criteria present (with the exception of the SOF where ≥2 positive items indicate frailty). We constructed a logistic regression model to compare frail and non-frail patients according to the PFP by study site, age group and gender using Zurich as our reference site. A separate model was run with pre-frailty as the outcome. All analyses were performed with SAS v9.4 (SAS Inc., Carey, North Carolina, USA).



Overall, mean age was 74.9 ±4.4 years, and 61.6% were women. Mean Mini-Mental Status Examination score was 28.5 points (SD 1.5). Mean BMI was 26.3kg/m2 (SD 4.3), and 28.5% (n=569) of participants had three or more comorbidities at baseline. Baseline characteristics of all study participants by gender are presented in Table 1. Of the total 2,157 participants in DO-HEALTH, 2,144 (99.4%) had complete data at baseline on at least three of the variables for all frailty measures (Supplementary Table 2). In all, 1,005 participants were included from Switzerland (1 participant was excluded from Zurich), 350 from Germany, 300 from Portugal (1 participant excluded), 289 from France (11 participants excluded), and 200 from Austria.

Table 1. Baseline Characteristics of the DO-HEALTH participants

Values are means and standard deviations unless otherwise noted. BMI (Body Mass Index), MMSE (Mini-Mental Status Examination, used to measure cognitive impairment, range of 0-30 points, where higher scores are better, and scores >24 suggest normal cognitive function), kPa (Kilopascal), SPPB (Short Physical Performance Battery, used to measure lower extremity function, scores range from 0-12 points, where higher scores indicate better functioning), aaccording to a Self-Administered Comorbidity Questionnaire (23).


Prevalence of frailty and pre-frailty in all DO-HEALTH participants by instrument

According to our primary measure PFP, 3.0 % (n=64) of DO-HEALTH participants were frail, and 43.0% (n=922) were pre-frail (i.e. at-risk for frailty). See Figure 1 for the overall prevalence of frailty and pre-frailty based on the PFP and the other four derived instruments. Prevalence of frailty in the total population was highest based on SHARE-FI1 (7.0%), and lowest based on SOF-FI (1.0%). Also, prevalence of pre-frailty in the total population was highest based on SHARE-FI1 (43.7%), and lowest based on SOF-FI (25.9%).

Figure 1. Prevalence of frailty and pre-frailty by frailty instrument among DO-HEALTH participants (n=2,144 with available data out of n=2,157 total participants)


Prevalence of frailty and pre-frailty by country, age, and gender

With regard to country, the prevalence of frailty was highest using the SHARE-FI1 (range 1.5-24.3%), and lowest with the SOF-FI (range 0-3.0%). With regard to country, the prevalence of pre-frailty was highest with the SHARE-FI1 tool for all countries (range 40.3-51.3%), except for Portugal (51.3% vs. 57.3% on the PFP). With regard to country, the prevalence of pre-frailty was lowest with the SOF-FI for all countries (range 23.2-32.0%), except for Austria (27.5% vs. 24.0% on the FRAIL-Scale).
Figure 2a shows all prevalence rates for frailty by instrument and country. Figure 2b shows all prevalence rates for pre-frailty by instrument and country.

Figure 2a. Prevalence of frailty by country and investigated frailty instrument

Figure 2b. Prevalence of pre-frailty by country and investigated frailty instrument


With regard to age group (<75 years vs. ≥75 years), persons 75 years and older were more often frail compared to younger persons, regardless of the frailty instrument used (P <0.05 for all instruments). In both age groups, the prevalence of frailty was highest on the SHARE-FI1 (4.4%; 10.4%, P <0.0001) and lowest on the SOF-FI (0.5%, 1.7%, P = 0.0069). With regard to age group (<75 years vs. ≥75 years), there were significant differences in pre-frailty with a higher proportion of pre-frail participants among those age 75 years and older (P <0.05 for all instruments). In the younger age group, the prevalence of pre-frailty was highest using the SHARE-FI1 (42.0%), and lowest using the SOF-FI (22.9%), while in the older age group, the prevalence of pre-frailty was highest using the PFP (46.2%), and lowest using the SOF-FI (30.1%).
With regard to gender, there were significant differences for frailty by all investigated instruments with women having an overall higher prevalence of frailty compared to men (P <0.01), with the exception of the SOF-FI (P = 0.3505). With regard to pre-frailty, there were significant differences by gender for all investigated instruments with women having an overall higher prevalence of pre-frailty compared to men (P <0.001).
Results for frailty by country (study site) and each subgroup are shown in detail in Table 2. Results for pre-frailty by country (study site) and each subgroup are shown in detail in Supplementary Table 3.

Table 2. Prevalence of Frailty by Instrument in DO-HEALTH at baseline according to country (study site), age, and gender

All values are n (%); of the total 2,157 participants of DO-HEALTH, 2,144 (99.4%) had complete data at baseline on at least three of the variables for all frailty measures (Fried PFP: 42 missing 1 variable, 3 missing 2 variables, 10 missing 3 variables; SOF-FI: 37 missing 1 variable, 10 missing 2 variables; FRAIL-Scale: 34 missing 1 variable, 1 missing 2 variables, 12 missing ≥3 variables; SHARE-FI1: 5 missing 1 variable, 1 missing 2 variables, 12 missing ≥3 variables; SHARE-FI2: 13 missing ≥3 variables) and were included in this analysis. In all, 1,005 participants were from Switzerland (1 participant was excluded from Zurich), 350 from Germany, 300 from Portugal (1 participant excluded), 289 from France (11 participants excluded), and 200 from Austria). Fisher’s exact test was used to compare subgroups by country. Chi-square tests were used to compare frailty between subgroups by age and gender.


In our logistic regression model, we compared frail and non-frail participants according to the PFP by study center (using Zurich as our reference), age and gender. In regard to frailty, we found a 12-fold increased odds for participants from Coimbra (Portugal) [OR 12.00; CI 5.30-27.21] compared to Zurich, a 2-fold increased odds for older participants (≥75 years of age) [OR 2.00; CI 1.17-3.45], and a 2.8-fold increased odds for female participants [OR 2.84; CI 1.48-5.44]. In regard to pre-frailty, we found an almost 2-fold increased odds for participants from Coimbra (Portugal) [OR 1.90; CI 1.43-2.53] compared to Zurich, a 1.2-fold increased odds for older participants (≥75 years of age) [OR 1.20; CI 1.00-1.42], and a 1.45-fold increased odds for female participants [OR 1.45; CI 1.21-1.73] as shown in Supplementary Table 4.



In this large study of 2,144 relatively healthy, community dwelling European adults age 70 and older recruited from 5 European countries, and based on our primary measure PFP instrument, 3% of participants were frail and close to half of all participants were pre-frail. With regard to subgroups, prevalence of frailty and pre-frailty was highest in Portuguese participants, women, and adults age 75 and older. Notably, prevalence of frailty and pre-frailty varied substantially by the instrument used and was lowest by the SOF-FI, and highest by the SHARE-FI1.
Earlier studies in community dwelling older adults from the US and Germany reported frailty prevalence rates ranging from 2.5 to 17% using the PFP definition (26, 47-49). Of note, our reported frailty prevalence in DO-HEALTH with a mean age of 74.9 years was substantially lower than in similar age groups of population-based studies such as SHARE (1.3 vs. 4.4% for Switzerland, 13.8 vs. 28.6% for Portugal, 2.9 vs. 15.6% for France, 0.6 vs. 9.6% for Germany, and 0 vs. 11.6% for Austria) (29). However, the prevalence appeared more comparable to the age group between 65-74 years in the same study hereby reflecting the relatively good health of the DO-HEALH participants. At the same time our results of a higher odds for being frail and pre-frail by higher age and female gender are in line with the prior literature (1).
While our findings do not arise from a representative sample of the population, our results of a higher frailty prevalence in our Portugal center compared to the more central European countries in DO-HEALTH are consistent with the SHARE cohort findings (waves 1 and 6) and supported by a higher rate of physical inactivity in southern European countries, including Portugal (28, 29, 50). This is further supported by a recent study investigating limitations in instrumental activities of daily living (iADL) from SHARE data, highlighting the uneven distribution of burdens in self-care with higher rates in southern European countries, including Portugal (51). Additionally, the observed health disparity with regard to frailty prevalence between Portugal and Austria or Switzerland in DO-HEALTH may be explained by a potential difference in educational level (28)and linked to national economic indicators (52, 53), or specific patient sampling. In fact, DO-HEALTH participants from Portugal had less years of education compared to the four other countries. Further, the 2015 OECD (Organization for Economic Co-operation and Development) report on the attainment of overall goals in health care supports our findings by ranking Switzerland (No. 2), France (No. 6), Austria (No. 10), and Germany (No. 14) relatively high, compared with Portugal (No. 32) (54). Moreover, the observed pattern of a higher frailty prevalence in Portugal is supported by the 2015 OECD data where older adults from southern European countries had worse health states than central European older adults (54).
To the best of our knowledge, we lack population-based data on the prevalence of frailty and pre-frailty for the DO-HEALTH countries comparing different PFP operationalizations with test-based measurements of grip-strength and gait speed (i.e. PFP, SOF-FI) to mainly self-report instruments (i.e. FRAIL-Scale, SHARE-FI). In our analysis, the mainly self-report based instruments indicated a higher frailty prevalence compared with our primary measure PFP that included the test-based measurements of grip-strength and gait speed. This is in contrast to prior research indicating lower levels of frailty by self-report, e.g. in a longitudinal study from Ireland, and thus warrants further investigation (55).
With regard to instruments, the SOF-FI differed most from the PFP in terms of prevalence overall, but also in terms of differences by country, gender, and age. A possible explanation may be that SOF-FI only covers three of the original five PFP components (56, 57). A higher prevalence of frailty and pre-frailty among Portuguese, female, and participants age 75+ was observed, regardless of the frailty instrument used. Therefore, our findings caution its use as a frailty prevalence tool among relatively healthy older adults.
Our study has several strengths. First, we were able to utilize a large RCT baseline data set of 2,144 meticulously phenotyped older adults with standardized assessments of all participants from five European countries. Second, given the rich DO-HEALTH data set, we were able to match five frailty operationalizations with self-report and test-based assessments derived from Fried’s original physical frailty phenotype. Third, despite that fact that DO-HEALTH does not reflect population-based data and selected relatively healthy older adults, our data provides a conservative comparison of frailty and pre-frailty prevalence by five different frailty operationalizations and between five European countries.
Our study also has its limitations. First, we have to acknowledge that this is not a probabilistic sample and participants were recruited as relatively-healthy volunteers from the regions where the study was performed. Thus, suggesting that the actual prevalence of frailty may be subject to self selection and healthy participant bias (e.g. more frail participants may refrain from participating). Further, adaptions were necessary to apply the original Fried PFP to DO-HEALTH, due to the availability of variables from our data set. However, this has been necessary in most other studies following the initial publication by Fried et al (25, 58). A conceptual limitation of our study is the expected low overall prevalence of frailty in the relatively healthy community dwelling older adults enrolled in DO-HEALTH. In addition, DO-HEALTH data does not include information on differences in health care systems or access to primary care between participating centers. However, we still document significant variability between countries, and subgroups.
In summary, our study contributes unique data on the prevalence of pre-frailty and frailty in adults age 70 and older from five European countries enrolled in a large-scale RCT. With regard to the high prevalence of pre-frailty, we add important data indicating that interventions at this early stage possibly modulate the progression to frailty in a substantial share of the older adult population. Thus, our findings may help trigger efforts in the prevention of frailty, not only in Portugal, but also in the central European countries involved in DO-HEALTH. Our results support the fact that fraily prevalence is significantly influenced by the choice of instrument among relatively healthy community dwelling older adults. Further, all instruments but SOF-FI applied in the DO-HEALTH data set, appear to have construct validity with regard to country, gender, and age differences in prevalence of pre-frailty and frailty.
In conclusion, our findings suggest that the choice of frailty instrument among relatively healthy older adults is relevant with regard to frailty prevention and care planning efforts. Given the suggested large health care disparities with regard to frailty among participants from Portugal versus the central European countries in DO-HEALTH, further studies are needed to explore these differences with regard to other aspects of health and at the population-based level.


Funding: This work was supported by the European Commission’s 7th framework programme (FP7, HEALTH.2011.2.2.2-1) for investigator-driven clinical trials for therapeutic interventions in elderly populations (Project No: 278588-2).

Acknowledgements: MG drafted the manuscript and interpeted the data. POCB and LAA analysed and interpreted the data and contributed to the drafting of the manuscript. GF and AS provided critical revision of the manuscript. JAK, RWK, BV, JAPD, DF, RR, MB, EJO, AE and HABF designed the study concept, aquised the data and critically revised the manuscript. We would like to thank all DO-HEALTH investigators, researchers, and study personnel at the seven partner sites throughout Europe for their contribution.

Conflict of Interest: The authors declare no conflict of interest in regard to this work.

Ethical standards: The DO-HEALTH study was approved by ethics and regulatory agencies of all 5 countries and the study protocol has been previously published (30). A data and safety monitoring board oversaw the study.

Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.





1. Hoogendijk, E.O., et al., Frailty: implications for clinical practice and public health. The Lancet, 2019. 394(10206): p. 1365-1375.
2. Landi, F., et al., Sarcopenia as the Biological Substrate of Physical Frailty. Clin Geriatr Med, 2015. 31(3): p. 367-74.
3. Cruz-Jentoft, A.J., et al., Sarcopenia: revised European consensus on definition and diagnosis. Age and Ageing, 2018: p. afy169-afy169.
4. Verlaan, S., et al., High Prevalence of Physical Frailty Among Community-Dwelling Malnourished Older Adults-A Systematic Review and Meta-Analysis. J Am Med Dir Assoc, 2017.
5. Vanpottelbergh G, Vaes B, Adriaensen W, Mathei C, Legrand D, Wallemacq P, Degryse JM: The glomerular filtration rate estimated by new and old equations as a predictor of important oucomes in elderly patients. BMC Medicine 2014, 12(27):doi:10.11186/11741.
6. Rockwood, K. and A. Mitnitski, Frailty Defined by Deficit Accumulation and Geriatric Medicine Defined by Frailty. Clinics in Geriatric Medicine, 2011. 27(1): p. 17-26.
7. Pacifico, J., et al., Prevalence of sarcopenia as a comorbid disease: A systematic review and meta-analysis. Experimental Gerontology, 2020. 131: p. 110801.
8. Fimognari, F.L., et al., The Severity of Acute Illness and Functional Trajectories in Hospitalized Older Medical Patients. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 2017. 72(1): p. 102-108.
9. Lin, H.S., et al., Frailty and post-operative outcomes in older surgical patients: a systematic review. BMC Geriatr, 2016. 16(1): p. 157.
10. Wou, F., et al., The predictive properties of frailty-rating scales in the acute medical unit. Age Ageing, 2013. 42(6): p. 776-81.
11. Apóstolo, J., et al., Predicting risk and outcomes for frail older adults: an umbrella review of frailty screening tools. JBI Database System Rev Implement Rep, 2017. 15(4): p. 1154-1208.
12. Vermeiren, S., et al., Frailty and the Prediction of Negative Health Outcomes: A Meta-Analysis. J Am Med Dir Assoc, 2016. 17(12): p. 1163.e1-1163.e17.
13. Gill, T.M., et al., Transitions between frailty states among community-living older persons. Arch Intern Med, 2006. 166(4): p. 418-23.
14. Subra, J., et al., The integration of frailty into clinical practice: preliminary results from the Gerontopole. J Nutr Health Aging, 2012. 16(8): p. 714-20.
15. Oliveira, J.S., et al., Evidence on Physical Activity and the Prevention of Frailty and Sarcopenia Among Older People: A Systematic Review to Inform the World Health Organization Physical Activity Guidelines. J Phys Act Health, 2020: p. 1-12.
16. Faller, J.W., et al., Instruments for the detection of frailty syndrome in older adults: A systematic review. PLoS One, 2019. 14(4): p. e0216166.
17. Theou, O., et al., What do we know about frailty in the acute care setting? A scoping review. BMC Geriatr, 2018. 18(1): p. 139.
18. Morley, J.E., et al., Frailty consensus: a call to action. J Am Med Dir Assoc, 2013. 14(6): p. 392-7.
19. Rodriguez-Manas, L., et al., Searching for an operational definition of frailty: a Delphi method based consensus statement: the frailty operative definition-consensus conference project. J Gerontol A Biol Sci Med Sci, 2013. 68(1): p. 62-7.
20. Fried, L.P., et al., Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci, 2001. 56(3): p. M146-56.
21. Mitnitski, A.B., A.J. Mogilner, and K. Rockwood, Accumulation of deficits as a proxy measure of aging. ScientificWorldJournal, 2001. 1: p. 323-36.
22. Rockwood, K., A. Mogilner, and A. Mitnitski, Changes with age in the distribution of a frailty index. Mech Ageing Dev, 2004. 125(7): p. 517-9.
23. Buta, B.J., et al., Frailty assessment instruments: Systematic characterization of the uses and contexts of highly-cited instruments. Ageing Res Rev, 2016. 26: p. 53-61.
24. Searle, S.D., et al., A standard procedure for creating a frailty index. BMC Geriatr, 2008. 8: p. 24.
25. Theou, O., et al., Modifications to the frailty phenotype criteria: Systematic review of the current literature and investigation of 262 frailty phenotypes in the Survey of Health, Ageing, and Retirement in Europe. Ageing Res Rev, 2015. 21: p. 78-94.
26. Ensrud, K.E., et al., Comparison of 2 frailty indexes for prediction of falls, disability, fractures, and death in older women. Arch Intern Med, 2008. 168(4): p. 382-9.
27. Abellan van Kan, G., et al., The I.A.N.A Task Force on frailty assessment of older people in clinical practice. J Nutr Health Aging, 2008. 12(1): p. 29-37.
28. Santos-Eggimann, B., et al., Prevalence of frailty in middle-aged and older community-dwelling Europeans living in 10 countries. J Gerontol A Biol Sci Med Sci, 2009. 64(6): p. 675-81.
29. Manfredi, G., et al., Prevalence of frailty status among the European elderly population: Findings from the Survey of Health, Aging and Retirement in Europe. Geriatr Gerontol Int, 2019. 19(8): p. 723-729.
30. Bischoff-Ferrari, H.A., et al., DO-HEALTH: Vitamin D3 – Omega3 – Home exercise – Healthy aging and longevity trial – Design of a multinational clinical trial on healthy aging among European seniors. Contemporary Clinical Trials, 2020: p. 106124.
31. Folstein, M.F., S.E. Folstein, and P.R. McHugh, “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psych Res, 1975. 12.
32. Guralnik, J.M., et al., Lower extremity function and subsequent disability: consistency across studies, predictive models, and value of gait speed alone compared with the short physical performance battery. J Gerontol A Biol Sci Med Sci, 2000. 55(4): p. M221-31.
33. Sangha, O., et al., The Self-Administered Comorbidity Questionnaire: a new method to assess comorbidity for clinical and health services research. Arthritis Rheum, 2003. 49(2): p. 156-63.
34. Bouillon, K., et al., Measures of frailty in population-based studies: an overview. BMC Geriatr, 2013. 13: p. 64.
35. Sipers, W.M., et al., The Martin Vigorimeter Represents a Reliable and More Practical Tool Than the Jamar Dynamometer to Assess Handgrip Strength in the Geriatric Patient. J Am Med Dir Assoc, 2016. 17(5): p. 466.e1-7.
36. Ensrud, K.E., et al., A comparison of frailty indexes for the prediction of falls, disability, fractures, and mortality in older men. J Am Geriatr Soc, 2009. 57(3): p. 492-8.
37. Kiely, D.K., L.A. Cupples, and L.A. Lipsitz, Validation and comparison of two frailty indexes: The MOBILIZE Boston Study. J Am Geriatr Soc, 2009. 57(9): p. 1532-9.
38. Guralnik, J.M., et al., A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol, 1994. 49(2): p. M85-94.
39. Abellan van Kan, G., et al., Frailty: toward a clinical definition. J Am Med Dir Assoc, 2008. 9(2): p. 71-2.
40. Morley, J.E., T.K. Malmstrom, and D.K. Miller, A simple frailty questionnaire (FRAIL) predicts outcomes in middle aged African Americans. J Nutr Health Aging, 2012. 16(7): p. 601-8.
41. Vellas, B., et al., Designing pharmaceutical trials for sarcopenia in frail older adults: EU/US Task Force recommendations. J Nutr Health Aging, 2013. 17(7): p. 612-8.
42. Woo, J., et al., Frailty Screening in the Community Using the FRAIL Scale. Journal of the American Medical Directors Association, 2015. 16(5): p. 412-419.
43. Cuenoud, P., Prevalence of the Frailty Phenotype in community-dwelling population aged 50 and over in the Survey of Health, Ageing and Retirement in Europe, in IEMS Lausanne – Faculté des Hautes Etudes Commerciales et Faculté de Biologie et de Médecine. 2007, Université de Lausanne.
44. Romero-Ortuno, R., The SHARE operationalized frailty phenotype: a comparison of two approaches. Eur Geriatr Med, 2013. 4(4).
45. Pialoux, T., J. Goyard, and B. Lesourd, Screening tools for frailty in primary health care: a systematic review. Geriatr Gerontol Int, 2012. 12(2): p. 189-97.
46. Verghese, J., et al., Self-reported difficulty in climbing up or down stairs in nondisabled elderly. Arch Phys Med Rehabil, 2008. 89(1): p. 100-4.
47. Woods, N.F., et al., Frailty: emergence and consequences in women aged 65 and older in the Women’s Health Initiative Observational Study. J Am Geriatr Soc, 2005. 53(8): p. 1321-30.
48. Cawthon, P.M., et al., Frailty in older men: prevalence, progression, and relationship with mortality. J Am Geriatr Soc, 2007. 55(8): p. 1216-23.
49. Fuchs, J., M. Busch, and C. Scheidt-Nave, [Frailty: Prevalence in the Resident Population of Germany 70 – 79 Years of Age – a Population-Based Approach]. Anasthesiol Intensivmed Notfallmed Schmerzther, 2017. 52(11-12): p. 758-763.
50. Gomes, M., et al., Physical inactivity among older adults across Europe based on the SHARE database. Age Ageing, 2017. 46(1): p. 71-77.
51. Portela, D., et al., Instrumental Activities of Daily Living (iADL) Limitations in Europe: An Assessment of SHARE Data. Int J Environ Res Public Health, 2020. 17(20).
52. Theou, O., et al., Exploring the relationship between national economic indicators and relative fitness and frailty in middle-aged and older Europeans. Age Ageing, 2013. 42(5): p. 614-9.
53. World Health Organization. World report on ageing and health. 2015 30.04.2017]; Available from: http://apps.who.int/iris/bitstream/10665/186463/1/9789240694811_eng.pdf?ua=1.
54. OECD, Health at a Glance 2015. OECD Publishing.
55. Theou, O., et al., Measuring frailty using self-report and test-based health measures. Age Ageing, 2015. 44(3): p. 471-7.
56. Fried, L.P., et al., Nonlinear multisystem physiological dysregulation associated with frailty in older women: implications for etiology and treatment. J Gerontol A Biol Sci Med Sci, 2009. 64(10): p. 1049-57.
57. Bandeen-Roche, K., et al., Phenotype of frailty: characterization in the women’s health and aging studies. J Gerontol A Biol Sci Med Sci, 2006. 61(3): p. 262-6.
58. Saum, K.U., et al., Development and evaluation of a modification of the Fried frailty criteria using population-independent cutpoints. J Am Geriatr Soc, 2012. 60(11): p. 2110-5.



R. Jiwani1,2, J. Wang1,3, C. Li1,3, B. Dennis3,4, D. Patel1,5,6, J. Gelfond7, Q. Liu7, N. Siddiqui3, C. Bess1, S. Monk1, M. Serra2,8,9, S. Espinoza2,8,9


1. School of Nursing, University of Texas Health Science Center at San Antonio, Texas, USA; 2. Geriatric Research, Education & Clinical Center (GRECC), South Texas Veterans Health Care System, San Antonio, Texas, USA; 3. Center on Smart and Connected Health Technologies, School of Nursing, University of Texas Health Science Center at San Antonio, Texas, USA; 4. Graduate School of Biomedical Sciences, University of Texas Health Science Center at San Antonio, Texas, USA; 5. Biobehavioral Research Laboratory, School of Nursing, University of Texas Health Science Center at San Antonio, Texas, USA; 6. Mays Cancer Center, University of Texas Health Science Center at San Antonio, Texas, USA; 7. Department of Epidemiology & Biostatistics, University of Texas Health Science Center at San Antonio, Texas, USA; 8. Sam and Ann Barshop Institute for Longevity and Aging Studies, University of Texas Health Science Center at San Antonio, Texas, USA; 9. Department of Medicine, Division of Geriatrics, Gerontology & Palliative Medicine, University of Texas Health Science Center at San Antonio, Texas, USA

Corresponding Author: Rozmin Jiwani, PhD, RN, 7703 Floyd Curl Drive, San Antonio, Texas, 78229, Phone: 210-450-8498, Fax: 210-567-5822, Jiwani@uthscsa.edu

J Frailty Aging 2021;in press
Published online April 24, 2021, http://dx.doi.org/10.14283/jfa.2021.17



Background: Older adults with Type 2 diabetes (T2D) are more likely to be frail, which increases the risk for disability and mortality.
Objectives: To determine the feasibility of a behavioral lifestyle intervention, enhanced with mobile health technology for self-monitoring of diet and activity, to improve frailty in overweight/obese older adults (≥65 years) diagnosed with T2D.
Design, Setting, and Participants: Single arm, 6-month study of a behavioral lifestyle intervention in 20 overweight/obese (BMI>25) older adults (≥ 65 years) with self-reported T2D diagnosis who owned a smartphone. A Fitbit tracker was provided to all participants for self-monitoring of diet and physical activity. Our primary outcome of feasibility was measured by session attendance, adherence to Fitbit usage to self-monitor diet and physical activity, and study retention. Secondary outcomes included the preliminary efficacy of the intervention on frailty, physical function, quality of life, and T2D-related outcomes.
Results: Eighteen participants completed the study. The mean age was 71.5 (SD ± 5.3) years, 56% were female, and half were Hispanic. At baseline, 13 (72%) were pre-frail, 4 (22%) were frail, and 1 (6%) were non-frail. At follow-up, frailty scores improved significantly from 1.61 ± 1.15 to 0.94 ± 0.94 (p=0.01) and bodyweight improved from 205.66 ± 45.52 lbs. to 198.33 ± 43.6 lbs. (p=<0.001).
Conclusion: This study provides evidence for the feasibility of a behavioral lifestyle intervention in overweight/obese older adults with T2D and preliminary results support its potential efficacy in improving frailty score.

Keywords: Type 2 diabetes, lifestyle intervention, self-monitoring, frailty, personal fitness technology.



Frailty is characterized as a geriatric syndrome of vulnerability and progressive physical decline, which significantly increases risks for falls, disability, and death (1). Frailty prevalence is approximately 10% population over 60 years old and reaches over 25% in those ages 80 years and older (2). Type 2 Diabetes (T2D) and overweight/obesity are highly prevalent in older adults and are significant predictors of both onset and worsening of frailty in older adults (3-5). Frailty prevalence is much higher in older adults diagnosed with T2D (32% to 48%) than in the general older population (5% to 10%), and frail individuals with T2D have higher mortality rates (6). Indeed, frailty is emerging as another category of T2D complications, in addition to the traditional micro-and macrovascular sequelae (7). To date, there are no widely accepted evidence-based interventions to improve frailty (8). Such interventions would be especially useful in older adults with T2D as frailty is highly associated with diabetes.
Lifestyle modification is the first-line treatment for T2D and there is evidence to suggest that weight loss and exercise improve components of frailty, such as gait speed, muscle strength, and physical function (8). The Look AHEAD (Action for Health in Diabetes) randomized clinical trial study demonstrated the efficacy of behavioral intensive lifestyle intervention (ILI) of caloric restriction and physical activity, targeting a 7% weight loss over the control group to reduce adiposity, improve glycemic control (9), reduce risk factors for cardiovascular disease (10), microvascular complications (11), and physical function (12, 13) over the follow-up years (range 8-11 years). In addition, ILI participants had less need for diabetes medications (14) and lowered health care costs (15). The ILI intervention was also associated with lower frailty index (FI) scores throughout follow-up (16). The FI operationalizes frailty as deficit accumulation and calculates the proportion of health deficits present in a person. This differs from the Fried frailty phenotype criteria, which scores frailty based on the presence of five criteria, including unintentional weight loss, exhaustion, low physical activity, poor muscle strength, and slow gait speed. The Look AHEAD trial did not assess frailty by Fried phenotype criteria at study baseline or during the intervention but did have this measure at later follow-up examinations. Therefore, the effect of the ILI intervention on frailty characteristics as measured by the Fried criteria is not known.
Self-monitoring is a cornerstone of behavioral lifestyle interventions. The rise in popularity and accessibility of personal fitness technology, such as wearable mobile health technology, apps, and mobile devices have become an informative and useful option to positively reinforce health, exercise, and nutritional behavior (17). Technology use has been steadily increasing in older adults, as smartphone ownership in this population has increased from 18% in 2013 to 43% in 2017, and the number of older adults who access the internet has risen to 67% (18). The integration of personal fitness technologies into behavioral interventions has been shown to increase physical activity, decrease sedentary behaviors, and improve objective measures (i.e., glycemic control) in older adults (19-23).
This study aimed to determine the feasibility of Look AHEAD behavioral lifestyle intervention, enhanced with mobile health technology for self-monitoring of diet and physical activity, to improve frailty as assessed by Fried criteria. Our primary outcome of feasibility was measured by session attendance, adherence to Fitbit usage to self-monitor diet and physical activity, and study retention. Secondary outcomes included the preliminary efficacy of the intervention on frailty, physical function, quality of life, and T2D-related outcomes. The primary hypothesis was the study is feasible and behavioral lifestyle intervention will improve frailty in community-dwelling overweight/obese older adults (≥ 65 years) diagnosed with T2D.



Study Design

We conducted a single-arm, 6-month study of a behavioral lifestyle intervention program enhanced with mobile health technology. A Fitbit wristband activity tracker was provided to all participants for self-monitoring of diet and physical activity and the companion app was downloaded on their smartphone by the research team during the baseline screening visit. The study was approved by the Institutional Review Board at the University of Texas Health Science Center at San Antonio (UTHSCSA).

Participant recruitment

We recruited participants using a combination of methods supported by the San Antonio Claude D. Pepper Older Americans Independence Center, including a call center, community advisory board, and research volunteer registry. We also utilized our University’s Find-a-Study recruitment website, advertisement in our primary care clinics, local senior centers, local newspapers, local health fairs/community events, outreach to community leaders, and word of mouth. A study flow diagram using the Consolidated Standards of Reporting Trials (CONSORT) model is shown in Figure 1.

Figure 1. CONSORT flow-diagram for a single arm non-randomized study


Sample and setting

We recruited 20, community-dwelling, overweight/obese (BMI ≥ 25 kg/m2), older adults (≥ 65 years) diagnosed with T2D (self-report of provider-diagnosis). We planned that all participants would receive a total of 10 face-to-face group sessions for the behavioral lifestyle intervention over 6 months. However, due to local and institutional COVID-19 restrictions, 3 of our 4 groups (groups 2, 3, and 4) attended 2-3 virtual (WebEx) group educational sessions to complete the study. They were then seen in-person for their end-of-study assessments. See Table 1 for detailed inclusion and exclusion criteria. We conducted all the study-related visits at the Biobehavioral Research Laboratory at UTHSCSA School of Nursing.

Table 1. Inclusion and exclusion criteria


Sample size

For this feasibility study, a sample size of n=20 was used, based on Moore et al. (24) which provide recommendations for planning pilot studies in clinical and translational research. We assumed a 40% attrition rate and enrolled 20 participants.

Study procedures

Potential participants were informed about the study over the phone and, if interested, proceeded with phone screening for eligibility based on inclusion and exclusion criteria (Table 1). Eligible participants were scheduled for an in-person baseline screening visit. T2D was ascertained by participant self-report of provider-diagnosed T2D in response to the question, “Has a medical doctor or other provider told you that you have Type 2 diabetes?” We also performed medication review and baseline lab work (fasting blood glucose and Hemoglobin A1c [HbA1c]) to verify the reliability of the self-reported T2D diagnoses. After informed consent, we collected baseline data. These same assessments were repeated at 6-month at the end of the study visit.

Data collection

Health history and physical assessment: A standardized history and physical assessment were performed and participants’ self-reported medical, surgical, and medication history were recorded. We measured participants’ vital signs (heart rate, blood pressure, temperature) and anthropometric measurements (waist circumference [inches], height [inches], weight [pounds]. Body mass index was calculated as weight (kg)/height (m2).
Frailty assessment: We classified frailty status using the Fried phenotype criteria (1): 1) Self-reported unintentional weight loss of ≥10 pounds in the past year; 2) Self-reported exhaustion; 3) Low energy expenditure using the Minnesota Leisure Time Physical Activity Questionnaire (MLTQ) to assess physical activity (duration and frequency) (25); 4) Weakness measured via grip strength using a handheld dynamometer in the dominant hand; 5) Timed gait at usual speed over a 10-foot walk as previously described (26). Any intentional weight loss that may have occurred due to the lifestyle intervention would not meet the criteria for frailty-related weight loss, as frailty-related weight loss is defined as unintentional. A frailty score was calculated as the number (0 – 5) of frailty characteristics present. Those with ≥ 3 of these 5 characteristics are categorized as frail; those with 1 or 2 are categorized as pre-frail, and those with none are categorized as non-frail (1).
Physical function: We administered the Short Physical Performance Battery (SPPB), a reliable and valid tool to assess lower extremity physical functions. The SPPB consists of 1) standing balance (ability to stand with the feet together in side-by-side, semi-and full-tandem positions for 10 seconds each); 2) a 4-meter walk (gait speed test) to assess the time it takes to complete the walk at a usual pace; 3) time to complete five repeated chair stand without using hands. Each of the 3 performance measures was assigned a score ranging from 0 to 4 and summed to create a score ranging from 0 (worst) to 12 (best). The SPPB is sensitive to change over time (27, 28).
Self-reported measures of health and quality of life: Participants completed the Patient-Reported Outcome Measurements Information System (PROMIS-57 and PROMIS Global Health) questionnaires to assess self-reported health and quality of life (29). The PROMIS-57 assesses seven health domains (anxiety, depression, fatigue, pain interference, physical function, satisfaction with participation in social roles, and sleep disturbance) using eight items for each domain. PROMIS Global Health measures are generic, rather than disease-specific, and are intended to globally reflect individuals’ assessment of health. The questionnaires are ranked on a 5-point Likert scale and include an additional pain intensity scale (0-10 numeric rating). The feasibility of using PROMIS tools has been supported among ethnic minority, predominantly African American, overweight/obese adults with T2D who had greater symptom burden and poorer physical functioning than the general US population (22).
Clinical laboratory measures: Fasting blood glucose and HbA1c were measured at baseline and at 6-months to monitor safety and the impact of adherence to the interventions. Blood was collected from participants via venipuncture in the antecubital space of the preferred arm. Samples were allowed to sit at room temperature for 30 minutes after which they were centrifuged for 15 mins at 1000xg. Analyses were performed by Quest Diagnostics (Dallas, TX).

Study Intervention

Behavioral Lifestyle Intervention: We used the publicly available behavioral lifestyle intervention available from the Look AHEAD and Diabetes Prevention Program Group Lifestyle Balance websites in this study. The PI (RN, PhD) delivered all the sessions. All participants received 10 behavioral lifestyle intervention sessions over 6 months as shown in Table 2. The frequency and design of these sessions (weekly in month 1, biweekly in months 2 and 3, and monthly in months 4 to 6), allowed participants to master new skills gradually, then eventually adopt the behaviors as part of their daily lives (Table 2). Each session lasted 60 to 90 minutes. Missed sessions were replaced with either an individual make-up session or phone consultation depending on the participant’s availability. All 10 group sessions focused on adherence to the behavioral strategies, such as self-monitoring, goal setting, feedback, mindful eating, talkback negative thoughts, social support, problem-solving, relapse prevention, and handling holidays, among others (Table 2). Participants were given a tailored weight loss goal of 5-7% based on calorie and fat intake based on their current weight, and physical activity goals to gradually increase to 175 minutes (about 3 hours) per week by the end of the study following Look AHEAD guidelines (9). In every group session, we reinforced the message to adhere to self-monitoring of diet, physical activity, and follow the behavioral education provided during the session. We recorded participants’ weight in every in-person group session, but we did not require them to check their weight at home.

Table 2. Topics and schedules for group sessions


Mobile health technology

A Fitbit wristband activity tracker was provided to all participants at the baseline screening visit, for self-monitoring of diet and activity, and the companion Fitbit app was downloaded on each participants’ smartphone. Participants received training on how to record their food (portion size, calories, and fat) and physical activity (duration and type of activity) using the Fitbit wristband activity tracker application on their smartphones. Participants were reminded during group sessions to self-monitor their diet and activity. We entered participants’ tailored bodyweight loss and activity goals in their device and encouraged them to record their food and activity daily. We also gave a written/pictorial step-by-step “cheat sheet” of instructions on how to record their food and activity and we reinforced these instructions throughout the study. Participants were asked to wear the device at all times. The research team sent two text messages per month, either for positive feedback or for a gentle reminder to those who were struggling or not recording their diet or physical activity data based on the data gathered from the connected health platform. Adherence to Fitbit usage for physical activity was inferred from step count data. If participants generated step count data for a given day, we determined that the participant had worn the device.

Statistical Analyses

Feasibility was measured via session attendance, adherence to Fitbit usage to self-monitor diet and physical activity (step count data), and retention at 6 months. We examined the preliminary efficacy of the intervention on frailty phenotype (Fried criteria), including physical function using the Short Physical Performance Battery (SPPB), T2D-related outcomes (body weight, waist circumference, fasting blood glucose, and HbA1c), including self-reported health and quality of life using the Patient-Reported Outcome Measurements Information System (PROMIS) questionnaires. We used a paired t-test to perform the efficacy analysis on the Fried frailty score (0 – 5), which was selected based upon Type I and II error performance with similar ordinal variables (30). Other exploratory endpoints of frailty, SBBP, and PROMIS component scores were assessed with paired t-tests. The five frailty components (e.g., grip strength, gait speed, physical activity, exhaustion, unintentional weight loss) were analyzed using McNemar’s test. Factors (age, gender, and ethnicity) may have moderated intervention effects on outcomes (frailty, body weight, and SPPB) using interaction terms (factor x time) within linear mixed-effects models. All testing was two-sided with a significance threshold of <0.05 for the p-values. All analyses were conducted in R (v3.5+, Vienna, Austria) within an accountable data analysis process.



Participant characteristics

Participant characteristics are detailed in Table 3. A total of 20 participants were enrolled with a mean age of 71.5 (± 5.3) years. Mean BMI was in the obese range (mean=33.7 kg/m2 ± 6.3), mean frailty score was 1.61 ± 1.15 (range: 0-4), and mean SBBP score was 9.22 ± 2.13 (range: 7-12) at study baseline. The majority of participants were pre-frail (72%) at the start of the study.

Table 3. Participant characteristics (N=18)


Feasibility outcome

Eighteen (90%) participants completed the study. Two participants dropped out after the first educational session and were no longer interested in participating. Data from the 18 participants who completed the study have been included in all analyses. Of the 18 participants who completed the study, 11 (61.1%) attended all 10 sessions, 6 missed 1 session, and 1 missed 3 sessions. The mean number of sessions attended was 9.5. The median [interquartile range] adherence rate for Fitbit activity tracker device was 95.6% [79.5%, 100%]. The maximum and minimum adherence rates were 100% and 36.4%, respectively. The median [interquartile range] of diet logging rate (days with at least 1 diet item logged/total days in study) was 81.5% [40.7%, 84.1%]. For the physical activity logging rate, these numbers were 8.4% [2.7%, 48.7%].

Effect of the study intervention

Table 4 shows the effect of the study intervention on frailty, physical function, quality of life, and T2D-related outcomes (tailored weight loss, waist circumference, fasting blood glucose, and HbA1c). Frailty total score improved significantly by 0.67 points (95% CI [1.15, 0.18], p=0.01). Frailty sub-scores for the 10-foot timed gait improved significantly at 0.30 points (95% CI [0.61, -0.01], p = 0.05), while physical activity and grip strength did not change significantly. Although the SBBP total score did not significantly change, there was a trend toward improvement by 0.83 (95% CI [0.07, -1.74] p=0.069). SBBP sub-scores for the balance tests improved significantly 0.50 (95% CI [-0.15, -0.85], p=0.008), while gait speed and repeated chair stand scores did not change significantly.

Table 4. Effect of the study intervention on frailty, physical function, quality of life, and T2D-related outcomes


About T2D-related outcomes, on average, participants lost 7.3 lbs. (95% CI [10.73, 3.92], p<0.001) and reduced their BMI by 1.15 kg/m2 (95% CI [1.69, 0.61], p<0.001). Age and ethnicity did not moderate treatment effects (p>0.05). In exploratory analyses, we found that males were more responsive to the intervention, with a decrease in weight (-7.9 lbs., p<0.05) and SBBP scores (2.8, p<0.001) relative to females (p<0.05). The intervention led to a 3.0% reduction in HbA1c, a 2.3% reduction in fasting blood glucose, a 1.0% reduction in waist circumference, but these changes were not statistically significant. PROMIS scores for self-reported health and quality of life were not affected by the study intervention (p-values > 0.05), however, we noticed some improvements in PROMIS Global Physical and Mental health scores (Table 4). Individual participant’s weight for every session is shown in Figure 2, which depicts weight/weight loss normalized by baseline weight. Most of the participants (except for two participants from group one) participated in virtual monthly educational sessions (sessions 8 to 10) due to the local COVID-19 quarantine. Seven out of 18 (38%) participants regained an average of 4.5 lbs. of weight during the virtual sessions, even though their weight had appropriately declined before this point (Figure 2).

Figure 2. Individual participants’ weight/weight loss normalized by baseline weight. The y-axis is weight percentage of baseline and x-axis is week number at the bottom and session number on the top for behavioral lifestyle intervention



To our knowledge, this is the first study to determine the feasibility of a behavioral lifestyle intervention, enhanced with mobile health technology for self-monitoring of diet and activity, for the goal of improving frailty in overweight/obese older adults with T2D. We demonstrated a high retention rate with 90% of participants completing the study, even though we encountered an unexpected obstacle due to the COVID-19 pandemic at the latter part of the study. Additionally, our participants were highly adherent to the use of the mobile health technology device (Fitbit) for self-monitoring of diet and activity. Although median adherence was high, we did observe a wide range with the lowest adherence rate of 36.4%, which indicates difficulty by some participants with self-monitoring and/or use of the technology. The procedures for logging diet and physical activity were similar in that they were both entered through the Fitbit app. Step counts were automatically logged when participants wore the Fitbit. We had a high adherence rate for wearing the Fitbit (95.6%). Although both logging of diet and physical activity was emphasized during group sessions, it is possible that participants could have assumed that since the Fitbit automatically calculates daily step count, it would also automatically detect all other types of physical activity. This could potentially have resulted in a lower logging rate for physical activity. As logging foods is tedious and requires more sophisticated user operations (selecting food items, portions, etc.), the high diet logging rate is encouraging and reaffirms that mobile technology is acceptable in this population. On the other hand, the low logging rate of physical activity suggests that the intervention may need further improvements to increase physical activity, or that there is a barrier to logging any activities that were performed. Similar findings related to a decrease in self-monitoring adherence in older adults have been reported and authors have suggested that more tailored technology instructions and further engineering may optimize user interface features (31). Future studies might consider providing direct exercise training and/or additional counseling, such as motivational interviewing, on the use of the activity tracker device. In a review of the literature of self-monitoring (diet, exercise, self-weighing) in therapeutic weight loss studies, Burke et al. reported that there was a correlation between adherence to self-monitoring and weight loss. However, there was a gradual decline in self-monitoring adherence throughout the study, which worsened when the treatment sessions decreased in frequency (32). This may, in part, explain our participants’ gradual weight increase starting with the monthly group sessions (sessions 8 to 10). However, the weight increase occurred simultaneously with the start of the local COVID-19 pandemic, which also likely impacted progress on weight loss.
The COVID-19 pandemic began during the implementation of this study, and local regulations required shelter-in-place protocols to be implemented throughout the region. Stay-at-home orders directly impacted the design of this study and led to our decision to use a virtual format to continue group sessions. This shift affected data collection for some outcomes (such as body weight measurement for every group session) and reduced in-person contact during the latter part of the study, however, we were able to continue using the group format virtually and participants recorded their self-reported weight (recording weight at home was not required for the study) for that session. The COVID-19 pandemic (started from monthly sessions 8-10) also likely had an impact on the behaviors of self-monitoring that we sought to modify with behavioral lifestyle intervention, even though bodyweight had appropriately declined prior to this point (Figure 2). This may also explain non-significant scores for self-reported health and quality of life PROMIS (e.g., anxiety, depression, participation in social activities) questionnaires score. We explored the impact of the COVID-19 experience and acceptability of the behavioral lifestyle intervention in our recently published focus group study wherein participants noted additional challenges to engaging in healthy behaviors as compared to before COVID-19 (33).
At baseline, most of our participants were characterized as pre-frail (72%) and frail (22%), supporting the high prevalence of frailty in older adults with T2D, as noted in other studies (34). Our pilot findings suggest that the Look AHEAD lifestyle intervention improves frailty characteristics (Table 3), and total frailty score (Table 4) in older adults diagnosed with T2D. In addition, our study participants started with a lower SPPB total score of 9.2 ± 2.1. A lower SPPB (score ≤ 9) is independently associated with falls (35) and frailty (36). We noticed a significant improvement in the SPPB balance score (p=0.008). Although there was a positive trend in SPPB total score it was not statistically significant (p=0.069). Similar improvements in SPPB total score have been reported by a 12-month multimodal (diet and exercise) intervention in older adults in the intervention group (37).
The small sample size and the lack of a control group were limitations of this study that should be mentioned. A small sample size limits generalizability to a broader population, and a lack of a control group limits a direct comparative set of data in individuals receiving the intervention and those who did not, which does limit the inferences that can be made. However, in spite of these limitations, this data provides information regarding the feasibility of delivering a lifestyle intervention in older adults with T2D and provides preliminary efficacy data for the effect of this intervention on frailty in this population. Secondly, the use of Fitbit to collect self-reported diet and physical activity may have led to imprecise reporting and may have allowed patient biases to affect the study’s outcome; however, this limitation is inherent to all studies using self-reported measures. We attempted to minimize imprecision and bias by using validated objective assessments during the follow-up visit. Lastly, the COVID-19 pandemic did interfere with in-person group intervention as previously discussed.


Conclusion and Implications

Our study provides evidence for the feasibility of using a behavioral lifestyle intervention, enhanced with mobile health technology, for the goal of reducing frailty and improving T2D outcomes in community-dwelling overweight/obese older adults. Our results are particularly encouraging as they were observed despite our study being impacted by the COVID-19 pandemic. We demonstrated a high retention rate and observed that personal fitness technology (Fitbit activity tracker) is acceptable and maybe a valuable intervention tool for self-monitoring in this population. Pre-frailty and frailty were common in our sample of older adults with T2D, in line with prior work showing that T2D and overweight/obesity are highly correlated with frailty. Detecting and managing frailty at its initial stages is ideal to improve healthspan and prevent disability and death. Further, screening for frailty and related measures, such as physical function and self-reported measures of health, may be beneficial for clinicians to identify high-risk patients who may benefit from the interventions to prevent or delay frailty. In the future, larger randomized controlled trials will be needed for definitive evidence of this intervention’s efficacy to reduce frailty-in this high-risk population. Future, more robust studies may include a longer intervention phase and follow-up period, and more precise measurements of diet and physical activity, such as actimetry and 24-hour diet recall. In addition, cognitive assessments may also be important to examine the impact of such an intervention on cognitive function, in addition to frailty and physical function.


Acknowledgments: We would like to thank the University of Texas Health Science Center at San Antonio’s School of Nursing for the use of the Center for Simulation Innovation, Office for Nursing Research Biobehavioral Lab, Center on Smart and Connected Health Technologies, and Geriatric Research, Education, and Clinical Center at the South Texas Veterans Health Care System, San Antonio, Texas for their support for this work.

Funding: This work was supported by the RL5 Mentored Research Career Development Award (Trainee: Rozmin Jiwani, Mentors: Sara Espinoza, Jing Wang) through the San Antonio Claude D. Pepper Older Americans Independence Center (P30AG044271) and the United States Department of Education through Title V Grant P031S150048 (Trainees: Edward Monk, Chandler Bess, Mentors: Darpan Patel, Rozmin Jiwani). This study was also supported by research infrastructure from the Geriatric Research, Education, and Clinical Center at the South Texas Veterans Health Care System.

Conflicts of Interest: There were no conflicts of interest declared by the authors for this study.

Ethics declaration: The University of Texas Health Science Center’s Institutional Review Board for the protection of human participants approved all procedures.



1. Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontology A Biol Sci Med Sci. 2001;56(3):M146-56.
2. Collard RM, Boter H, Schoevers RA, Oude Voshaar RC. Prevalence of frailty in community-dwelling older persons: a systematic review. J Am Geriatr Soc. 2012;60(8):1487-92.
3. Assar ME, Laosa O, Rodríguez Mañas L. Diabetes and frailty. Curr Opin Clin Nutr Metab Care. 2019;22(1):52-7.
4. Espinoza SE, Jung I, Hazuda H. Frailty transitions in the San Antonio Longitudinal Study of Aging. J Am Geriatr Soc. 2012;60(4):652-60.
5. Li G, Prior JC, Leslie WD, Thabane L, Papaioannou A, Josse RG, et al. Frailty and Risk of Fractures in Patients With Type 2 Diabetes. Diabetes Care. 2019;42(4):507-13.
6. Morley JE, Malmstrom TK, Rodriguez-Manas L, Sinclair AJ. Frailty, sarcopenia and diabetes. JAMDA. 2014;15(12):853-9.
7. Ulley J, Abdelhafiz AH. Frailty predicts adverse outcomes in older people with diabetes. Practitioner. 2017;261(1800):17-20.
8. Espinoza SE, Jiwani R, Wang J, Wang CP. Review of Interventions for the Frailty Syndrome and the Role of Metformin as a Potential Pharmacologic Agent for Frailty Prevention. Clinical Therapeutics. 2019;41(3):376-86.
9. Pi-Sunyer X, Blackburn G, Brancati FL, Bray GA, Bright R, Clark JM, et al. Reduction in weight and cardiovascular disease risk factors in individuals with type 2 diabetes: one-year results of the look AHEAD trial. Diabetes Care. 2007;30(6):1374-83.
10. Wing RR. Long-term effects of a lifestyle intervention on weight and cardiovascular risk factors in individuals with type 2 diabetes mellitus: four-year results of the Look AHEAD trial. Arch Intern Med. 2010;170(17):1566-75.
11. Look Ahead Research Group. Effect of a long-term behavioural weight loss intervention on nephropathy in overweight or obese adults with type 2 diabetes: a secondary analysis of the Look AHEAD randomised clinical trial. Lancet Diabetes Endocrinol. 2014;2(10):801-9.
12. Rejeski WJ, Bray GA, Chen SH, Clark JM, Evans M, Hill JO, et al. Aging and physical function in type 2 diabetes: 8 years of an intensive lifestyle intervention. J Gerontology A Biol Sci Med Sci. 2015;70(3):345-53.
13. Rejeski WJ, Ip EH, Bertoni AG, Bray GA, Evans G, Gregg EW, et al. Lifestyle change and mobility in obese adults with type 2 diabetes. N Engl J Med. 2012;366(13):1209-17.
14. Redmon JB, Bertoni AG, Connelly S, Feeney PA, Glasser SP, Glick H, et al. Effect of the look AHEAD study intervention on medication use and related cost to treat cardiovascular disease risk factors in individuals with type 2 diabetes. Diabetes Care. 2010;33(6):1153-8.
15. Espeland MA, Glick HA, Bertoni A, Brancati FL, Bray GA, Clark JM, et al. Impact of an intensive lifestyle intervention on use and cost of medical services among overweight and obese adults with type 2 diabetes: the action for health in diabetes. Diabetes Care. 2014;37(9):2548-56.
16. Simpson FR, Pajewski NM, Nicklas B, Kritchevsky S, Bertoni A, Ingram F, et al. Impact of Multidomain Lifestyle Intervention on Frailty Through the Lens of Deficit Accumulation in Adults with Type 2 Diabetes Mellitus. J Gerontology A Biol Sci Med Sci. 2019.
17. Agarwal S, LeFevre AE, Lee J, L’Engle K, Mehl G, Sinha C, et al. Guidelines for reporting of health interventions using mobile phones: mobile health (mHealth) evidence reporting and assessment (mERA) checklist. BMJ (Clinical research ed). 2016;352:i1174.
18. Anderson M, Perrin A. Technology Adoption Climbs Among Older Adults. Washington, DC: Pew Research Center. 2017. https://www.pewresearch.org/internet/2017/05/17/technology-use-among-seniors/
19. Elavsky S, Knapova L, Klocek A, Smahel D. Mobile Health Interventions for Physical Activity, Sedentary Behavior, and Sleep in Adults Aged 50 Years and Older: A Systematic Literature Review. J Aging Phys Act. 2019;27(4):565-93.
20. Espinoza SE, Orsak B, Wang CP, MacCarthy D, Kellogg D, Powers B, et al. An Individualized Low-Intensity Walking Clinic Leads to Improvement in Frailty Characteristics in Older Veterans. J Frailty Aging. 2019;8(4):205-9.
21. Jiwani R, Wang J, Berndt A, Ramaswamy P, Mathew Joseph N, Du Y, et al. Changes in Patient-Reported Outcome Measures With a Technology-Supported Behavioral Lifestyle Intervention Among Patients With Type 2 Diabetes: Pilot Randomized Controlled Clinical Trial. JMIR Diabetes. 2020;5(3):e19268.
22. Wang J, Cai C, Padhye N, Orlander P, Zare M. A Behavioral Lifestyle Intervention Enhanced With Multiple-Behavior Self-Monitoring Using Mobile and Connected Tools for Underserved Individuals With Type 2 Diabetes and Comorbid Overweight or Obesity: Pilot Comparative Effectiveness Trial. JMIR mHealth and uHealth. 2018;6(4):e92.
23. Jiwani R, Wang C-P, Orsak B, MacCarthy D, Kellogg D, Powers B, et al. A geriatrics walking clinic improves hemoglobin A1c and time gait in veterans with type 2 diabetes. Geriatric Nursing. 2020.
24. Moore CG, Carter RE, Nietert PJ, Stewart PW. Recommendations for planning pilot studies in clinical and translational research. Clin Transl Sci. 2011;4(5):332-7.
25. Folsom AR, Jacobs DR, Jr., Caspersen CJ, Gomez-Marin O, Knudsen J. Test-retest reliability of the Minnesota Leisure Time Physical Activity Questionnaire. J Chronic Dis. 1986;39(7):505-11.
26. Espinoza SE, Hazuda HP. Frailty in older Mexican-American and European-American adults: is there an ethnic disparity? J Am Geriatr Soc. 2008;56(9):1744-9.
27. Guralnik JM, Ferrucci L, Simonsick EM, Salive ME, Wallace RB. Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability. N Engl J Med. 1995;332(9):556-61.
28. Guralnik JM, Simonsick EM, Ferrucci L, Glynn RJ, Berkman LF, Blazer DG, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontology A Biol Sci Med Sci. 1994;49(2):M85-94.
29. Cella D, Riley W, Stone A, Rothrock N, Reeve B, Yount S, et al. The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005-2008. J Clin Epidemiol. 2010;63(11):1179-94.
30. Meek G, Ozgur C, Dunning K. Comparison of the t vs. Wilcoxon Signed-Rank Test for Likert Scale Data and Small Samples. J. Mod. Appl. Stat. Methods. 2007;6(1).
31. Rye Hanton C, Kwon Y-J, Aung T, Whittington J, High RR, Goulding EH, et al. Mobile Phone-Based Measures of Activity, Step Count, and Gait Speed: Results From a Study of Older Ambulatory Adults in a Naturalistic Setting. JMIR mHealth and uHealth. 2017;5(10):e104.
32. Burke LE, Wang J, Sevick MA. Self-monitoring in weight loss: a systematic review of the literature. J Am Diet Assoc. 2011;111(1):92-102.
33. Jiwani R, Dennis B, Bess C, Monk S, Meyer K, Wang J, et al. Assessing acceptability and patient experience of a behavioral lifestyle intervention using fitbit technology in older adults to manage type 2 diabetes amid COVID-19 pandemic: A focus group study. Geriatric Nursing. 2021;42(1):57-64.
34. Pritchard JM, Kennedy CC, Karampatos S, Ioannidis G, Misiaszek B, Marr S, et al. Measuring frailty in clinical practice: a comparison of physical frailty assessment methods in a geriatric out-patient clinic. BMC Geriatr. 2017;17(1):264.
35. Lauretani F, Ticinesi A, Gionti L, Prati B, Nouvenne A, Tana C, et al. Short-Physical Performance Battery (SPPB) score is associated with falls in older outpatients. Aging Clin Exp Res. 2019;31(10):1435-42.
36. Bandinelli S, Lauretani F, Boscherini V, Gandi F, Pozzi M, Corsi AM, et al. A randomized, controlled trial of disability prevention in frail older patients screened in primary care: the FRASI study. Design and baseline evaluation. Aging Clin Exp Res. 2006;18(5):359-66.
37. Rodriguez-Mañas L, Laosa O, Vellas B, Paolisso G, Topinkova E, Oliva-Moreno J, et al. Effectiveness of a multimodal intervention in functionally impaired older people with type 2 diabetes mellitus. J Cachexia Sarcopenia Muscle. 2019;10(4):721-33.




I. Ader1, L. Pénicaud1, S. Andrieu2, J.R. Beard3, N. Davezac4, C. Dray5, N. Fazilleau6, P. Gourdy5, S. Guyonnet7, R. Liblau6, A. Parini4, P. Payoux8, C. Rampon5, I. Raymond-Letron1, Y.Rolland2,7, P. de Souto Barreto2,7, P. Valet5, N. Vergnolle9, F. Sierra10, B. Vellas2,7, L. Casteilla1

1. RESTORE, UMR 1301-Inserm 5070 Etablissement Français du Sang-Occitanie (EFS), Inserm 1031, University of Toulouse III, National Veterinary School of Toulouse (ENVT), CNRS, Toulouse, France; 2. Inserm UMR 1027, Toulouse, France; University of Toulouse III, Toulouse, France; Department of Epidemiology and Public Health, CHU Toulouse, Toulouse, France; 3. Centre of Excellence in Population Ageing Research, University of New South Wales, Sydney, Australia; 4. Centre de Recherches sur la Cognition Animale (CRCA), Centre de Biologie Intégrative (CBI), CNRS, University of Toulouse III, Toulouse, France ; 5. Institut des Maladies Métaboliques et Cardiovasculaires, Inserm/Université Paul Sabatier UMR 1048 – I2MC 1 avenue Jean Poulhès BP 84225 31432 Toulouse Cedex 4 – France; 6. Infinity – Institut Toulousain des Maladies Infectieuses et Inflammatoires, CNRS U5051, INSERM U1291, University of Toulouse III, Toulouse, France; 7. Gérontopôle, Department of Geriatrics, CHU Toulouse, Toulouse, France; 8. ToNIC, Toulouse NeuroImaging Center, Université de Toulouse III, Inserm, UPS, France; 9. IRSD, Université de Toulouse, INSERM, INRA, ENVT, Université de Toulouse III, U1220, CHU Purpan, CS60039, 31024, Toulouse, France; 10. Institut du Vieillissement, Université de Toulouse, CHU Toulouse France

Corresponding Author: Louis Casteilla, RESTORE, UMR 1301-Inserm 5070 Etablissement Français du Sang-Occitanie (EFS), Inserm 1031, University of Toulouse III, National Veterinary School of Toulouse (ENVT), CNRS, Toulouse, France; louis.casteilla@inserm.fr
J Frailty Aging 2021;in press
Published online April 24, 2021, http://dx.doi.org/10.14283/jfa.2021.15



The find solutions for optimizing healthy aging and increase health span is one of the main challenges for our society. A novel healthcare model based on integration and a shift on research and care towards the maintenance of optimal functional levels are now seen as priorities by the WHO. To address this issue, an integrative global strategy mixing longitudinal and experimental cohorts with an innovative transverse understanding of physiological functioning is missing. While the current approach to the biology of aging is mainly focused on parenchymal cells, we propose that age-related loss of function is largely determined by three elements which constitute the general ground supporting the different specific parenchyma: i.e. the stroma, the immune system and metabolism. Such strategy that is implemented in INSPIRE projects can strongly help to find a composite biomarker capable of predicting changes in capacity across the life course with thresholds signalling frailty and care dependence.

Key words: Biomarkers, healthy aging, frailty, dependence.




The result of the balance between biological damage accumulation and compensatory mechanisms (1). As time passes, the compensatory mechanisms become less and less effective, which leads to more and more damage, a gradual decrease in physiological reserves and, phenotypically, this manifests as aging.

Healthy aging

Healthy aging is defined by The World Health Organization (WHO) as the process of developing and maintaining the functional ability that enables well-being in older age, including not just absence of disease, but also happiness, satisfaction and fulfillment (2). This requires to consider health span as final outcome instead of life span that corresponds to the duration of life whatever the quality of this life.


When an individual displays a maximal intrinsic capacity and resilience, she/he is considered as robust.


Frailty is a progressive age-related clinical condition characterized by a deterioration of physiological capacity leading to an increased vulnerability of the individual (3) and a higher risk of having poor health as well as a faster entry into care dependence (4). Frailty status occurs when the mobilization of the physiological reserve is not able to overcome the challenge (see adaptive capacity and resilience).

Intrinsic capacity (IC)

IC is the composite of 5 major physiological and mental capacities of an individual that can be assessed in a day-to-day environment (5, 6). Taking into account data from large cohort studies (7, 8), the WHO has suggested 5 key domains of capacity to maintain autonomy: sensorial (vision, audition), locomotor, cognitive, psychological and underlying physiological/cellular processes (9). These domains influence each other and are in turn influenced by environmental determinants.

Functional ability

Comprises those health-related attributes that enable people to be able to do what they want to do. This ability comprises not just the intrinsic capacity of the individual, but also the interactions between the individual and their environment.
Adaptive capacity, reserve and resilience: For a function, the ability to handle stress depends on the ability to mobilize adaptive capacity. For an individual, reserves are her/his maximal adaptive capacity. When adaptive capacity is more related to a physiological perspective, the term resilience is more generic and is coming from psychological field. At any time when the reserve exceeds the required adaptation capacity necessary to face a challenge, the individual is resilient. Frailty takes place as soon as these reserves become limited and resilience is diminished, resulting in a high sensitivity to any challenge that could lead to dependency (10).


Geroscience aims to identify the underlying molecular causes of chronic diseases and aging-related conditions. It focuses on hallmarks of the aging process, irrespective of the tissue, since most of these hallmarks are molecular and not specific to a given cell type. For example, a large literature is emerging that indicates the importance of cellular senescence in defining the aging phenotype (11).


Irreversible form of long-term cell-cycle arrest, caused by excessive intracellular or extracellular stress or damage. Senescent cells secrete a panel of molecules deleterious for surrounding cells.


The challenges of healthy aging and the search for biomarkers

While life expectancy has increased in recent decades, healthy life expectancy has not increased to the same extent, which means that people are living more years with functional losses generating crucial socioeconomical issues (12). As proposed by the WHO, a novel healthcare model must be privileged focused on heathy aging (glossary). This new paradigm requires an integrative and multi-disciplinary view based on the functional ability of individual to be able to do what they have reason to value. This corresponds to an interplay between the intrinsic capacity (IC, glossary) of each individual with its own environment. Intrinsic capacity can be described as the interactions between 5 domains (cognition, psychosocial, locomotion, sensory, vitality) that permits each individual to be autonomous. Consistent with this view, recent analysis suggests that quantifying an individual IC is a more useful predictor of future care dependence that the number of clinical morbidities an individual may be experiencing (9). Altogether this requires a shift of current research and healthcare models generally reactive disease-and organ centered towards the maintenance of optimal functional levels defining preventive and function-centered view. In this perspective, identifying frail persons (glossary) and especially when their IC declines and their resilience is limited (pre-frail state, glossary), as well as monitoring their individual evolution in order to propose solutions to maintain or recover IC and to prevent dependency is crucial. Pre-frailty was identified as a potentially reversible condition. Having this function-centered approach in mind, aging can be understood as the result of a subtle and progressive dysregulation of different balances that change overtime even at infra-clinic level rather than the result of a sudden appearance of one molecular or cellular defect. There is great individual diversity in the rate of this deterioration. This makes healthy aging research particularly tricky as well as in need of new multidisciplinary and global approaches to cover this unmet clinical need (13).
While chronological age (civil age, date of birth) is an easy-to-determine number and only reflects the time a person has spent on Earth, no precise definition is yet available for biological age. The basic idea behind biological aging is that aging occurs as damage to various cells and thus tissues in the body, which accumulate gradually with time and decrease the capacity to handle stress. It is the result of the interactions between intrinsic (eg. genetic, epigenetic, physiological…) and extrinsic (eg. lifestyle, infections…) factors that vary individually. In consequence, biological age differs from individual to individual for the same chronological age and thus the kinetics and slope of the biological aging process are unique to each of us (14). Very recently, the Snyder’s team has started to decipher certain elements that could explain biological age (15). They defined different types of biological aging patterns in different individuals, termed “ageotypes”. They discovered that people tend to fall into one of four biomolecular pathways associated with kidney, liver, metabolic and immune dysregulations. Some people fall squarely into one category, but others may meet the criteria of the four, depending on how their biological systems resists against aging. Unfortunately this work lacks long-term follow-up to assess whether these molecular phenotypes were ultimately associated with physiological failures.
Altogether, these reports imply that prediction, prevention and care must be personalized and adjusted to the individual’s biological, not just chronological age. This would be in line with the vision of 4P medicine (personalized, preventive, predictive and participatory) as proposed by Hood and colleagues (16-18). Even though biomarkers related to aging are at present widely used in medicine and a first list of 258 putative candidate biomarkers of aging was proposed by the TAME workgroup (19), identifying relevant biomarkers parameters in the context of biological aging and more specifically to the prefrail and frail state represents a new challenge, simply because those biomarkers do not fit within the disease-centred approaches of the current medicine.


Biomarkers for healthy aging: Contribution of the Occitanie Toulouse INSPIRE initiative

It is now widely recognized that the decline of health and the appearance of aging phenotypes would be the result of simultaneous deregulations of multiple physiological systems and their interactions (20). Thus, it is likely that only the use of multiple or composite biomarkers is relevant to assess these global body disturbances. In the field of healthy aging, ideal biomarkers would be to monitor the evolution towards frailty in order to intervene as soon as possible. It would be a single composite measurement used throughout the lifetime of individuals that encompasses and combines many physiological parameters. Such biomarkers should be capable of predicting changes in capacity across the life course with thresholds signalling the evolution from one state to another. Through deconvolution into their individual components, it should suggest the most appropriate pathway of care (prevention or/and treatment) for a given individual. Furthermore, the use of these biomarkers must be able to not just assess overt expressions of capacity but the resilience of an individual’s capacities in the face of additional stressors particularly for pre-frail biomarkers. The assessment of resilience is an important and challenging issue because this requires identifying the best challenge suitable to reveal the maximal adaptive capacity of the organism as a whole. Furthermore, in the best-case scenario, ideal biomarkers should be able to inform the clinician about the best-adapted care pathway for the individual (Figure 1). As usual, ideal biomarkers should also be robust, specific, as non-invasive as possible and cost-effective. Obviously, such biomarkers are much closer to fantasy than to reality, but such a definition can help the community to direct progress towards this tremendous challenge (21).

Figure 1. Markers for 4 P Healthy aging Medicine

Figure 1: Clinical markers (such as those derived from ICOPE) should be assessed at first to determine the category (robust, pre-frail, frail and dependent) of individuals of various chronological ages and what WHO function(s) is (are) altered. This should help to facilitate the search for relevant biological markers taking into account both WHO functions, hallmarks of aging as well as markers based on the SIM-INSPIRE strategy. By using artificial intelligence, we believe that relevant mixed-markers of biological ages will be made available.


A large list of potential biomarkers is already available and discussed in several reviews, but these biomarkers are most often unique, studied and validated as such for specific pathologies or derived from preclinical studies that have taken changes in disease specific physiology or lifespan, rather than health expectancy. A great step forward to identify transverse elements has been taken with the development of geroscience that considers biological aging as the main risk factors of chronic diseases (22). To go beyond these concepts and in the framework of WHO perspective, we propose to implement this analysis with a physiological transverse analysis. While the current approach to biology of aging is mainly focused on parenchymal cells that are specific to each organ consistent with the current organ/disease centered view, we propose that age-related loss of function is largely determined by dysfunctions of key generic components of tissue repair processes associated with tissue renewal and particularly the defective support of parenchyma cell environment (23, 24). As such, we propose to focus much of the investigations on the stroma, the immune system and metabolism (SIM), three elements considered as the general ground that constitutes the soil supporting the emergence and the functioning of the different specific parenchyma. Inflammation and immunity both represent a warning signal and the housekeeping guardians of tissue integrity. Mesenchymal stem/stroma cells (MSC) are supportive cells of the stroma that create and maintain the functional macro-architecture of tissues, in close connexion with vascular cells. Finally, metabolism is a central component that controls any cell decision and fate. Under challenges and chronic injuries, it is well known that these three inter-related components drive cell turnover and repair outcomes. However, malfunctioning on any of these components leads to i) chronic inflammation and a low-grade immune attrition (25), generation of fibrotic cells with extracellular matrix accumulation and oxidative damage within tissue (26) and ii) accumulation of ectopic adipocytes associated with loss of metabolic flexibility and organ dysfunctions (27). These features are classically described as key elements of the aging process and age-related organ dysfunction. The INSPIRE project proposes to set-up an interdisciplinary approach that gathers clinical and scientific community working on shared bio-resources (28, 29), a fine characterization of these three inter-related components and two types of key parenchyma cells: neurons and muscle cells, corresponding to the WHO-defined functions of locomotion and cognition.


A large panel of bio-resources facilitating the discovery of biomarkers

INSPIRE shared bio-resources

Identifying relevant biomarkers to detect and/or predict human frailty is challenging according to the complexity of the thematic and the human lifespan. These limits can be circumvented by the creation of a panel of biological models and samples that can cover the entire chain from the bench to the bedside and vice-versa. As starting point and at population level, the first personalized assessment to detect frailty to anticipate care dependency can be based on a non-invasive self-evaluation that can be easily deployed to a broad population in its specific environment through a digital application (30). When a problem is flagged, a team of health workers is alerted, and the individual taken care of in a more specific and detailed way guided by further assessment (Figure 1). Beside such e-cohort serving as a pool of participants to further investigations, different cohort can be implemented, including humans, dogs, mice and fish (Fig 2). In longitudinal human cohort, the longitudinal follow-up of the same individuals with a short delay between recurrent evaluations will ensure a reliability of the data and a fine description of any change in IC associated with any relevant life events (28, 30). A computational approach will permit to define predictive and prognostic biomarkers. As many experiments are not easily done in human being due both to experimental procedures, ethics considerations as well as the quite long life-time, complementary animal cohorts should help the quest of putative biomarkers.

INSPIRE aims to build and manage various cohorts in different species including human, rodents, fishes and dogs.


As the simplest but relevant model for studying biological aging, Nothobranchius Furzeri (African or Turquoise killifish) the vertebrate with the shortest known lifespan can be used (31). This animal develops very rapid aging of all its organs (4 to 6 months), with the recapitulation of all the aging hallmarks found in older-adults patients, both at the molecular and functional level (cognitive alterations, reduction of locomotion, sarcopenia onset…) (32, 33) and its health (meaning behavior) can be finely evaluated. It represents a unique animal model to quickly test and validate new hypotheses within the framework of the INSPIRE project. A murine cohort as small mammals and gold-standard physiopathological models has been designed as a mirror to the human cohort and fully described in (34). Finally, a dog cohort from adulthood to elderly represents an unvaluable tool because dog is a mammal displaying rapid aging compared to humans but with a very similar physiology including aging features and pathologies, as well as an evolution towards biological aging with IC reduction. Furthermore, it is possible to evaluate the biological and physical parameters of dogs in a manner almost identical to that of humans (physical and clinical examinations, blood and urine analyzes, medical imaging, functional tests, etc.). In addition, dogs share with their owners their environment (air, water, food, exposure to potential contaminants, …) and their lifestyle (active or sedentary) (35, 36).
In all of these animal cohorts, individuals has to be widely phenotyped and generate a large biobanking in order not only to identify putative biomarkers that can be tested and validated in human cohort but also to better investigate those mechanisms identified in the human cohorts in terms of changes in metabolism, immune and regenerative capacities.

The challenges to identify resilience markers

Prefrail state can be determined only if the individual is submitted to a defined challenge or stress which will led to a return, or not, to a “normal” state if resilience is present, or not. Thus, to evaluate functional/physiological resilience as part of a battery of potential biomarkers of aging, we propose to carefully evaluate responses to stressors, either “lifelong challenges” or well-controlled acute stressors. In fact, different physical and psychological life events and injuries can be considered as real-life challenges, including for example both acute or chronic infections (HIV, flu, COVID-19…) but also other life events such as hip fracture or surviving the death of a spouse or a relative, during a divorce or the loss of the job. Since the follow-up of people enrolled in the INSPIRE human cohort is frequent, as soon as such challenge occurs, the stress effect on functions and biological parameters will be informed during the following evaluation. Thus, INSPIRE design can reveal physiological adaptations specific to biological aging in order to unmask biomarkers following a situation of stress.
For controlled stressors, we propose to challenge immune, inflammatory and metabolic homeostasis by an exercise corresponding to the evaluation of muscle tone (isometric measurements) coupled with intense physical exercise. Blood sampling before and just after this VO2 max challenge will allow to a better appraising of physiological reserve, as well as to test the acute response of metabolic, immune and inflammatory blood biomarkers after a maximal effort. To in vivo assess the immune system, we will monitor the magnitude and quality of the antigen-specific immune response as well as the bystander inflammatory response following administration of a vaccine included in standard care (flu, pneumococcal or zoster vaccination). Serum and peripheral blood mononuclear cells will be prospectively collected prior to, and at different time points following the vaccine shots. We will use several ‘omics’ technologies to monitor population and individual-level changes such as cell-subset phenotyping, cytokine response assays, and gene expression profiles, which will allow the longitudinal tracking of multiple immune features. These investigations can be supplemented by ex vivo challenges of precise immune cell types using stimuli targeting either innate immune cells or lymphocyte subsets. With these investigations, we aim at assessing not only the immune reserves but also immune cell global dysfunction, which may contribute by itself to premature aging, as elegantly documented in mice (37). To assess the role of supportive mesenchymal stroma/stem cells (MSC) in repair processes, the definition of a relevant physiological challenge and the way to investigate it is very tricky. The evaluation of the kinetics and quality of wound healing processes after a standardized small incision required for tissue sampling could represent a relevant test. In addition, the ability of MSC isolated from biopsies to participate in repair processes can be also indirectly evaluated through their fine characterization and their repair potential. The quality of the supportive mesenchymal compartment will be determined after an in vitro challenge induced by various inducers mimicking native molecules released after tissue injury. This corresponds to the determination of MSC multipotency including the sensitivity of these cells to differentiate towards myofibroblast as well as their pleiotropic induced paracrine activities. This will be achieved via robotized phenotyping and standardized assays using Advanced Medicinal Therapeutical Product quality control facility (www.ecellfrance.fr).

Implementation and perspectives

An important aspect is the rapid sharing of bio-resources and data to a large multi-disciplinary community. Pending sufficient data from the living longitudinal human cohort, cross-sectional studies will be carried out to develop first lists of putative biomarker profiles with the corresponding relevant thresholds for their uses in clinical practice. Since the healthy aging field is inherently very complex, the irruption of artificial intelligence in precision medicine should be of a great help to build expert systems suitable to deploy 4P Medicine approaches (38, 39). This should largely change the field of biomarkers, moving from a single or a limited profile of biomarkers with associated relevant thresholds towards a profile of multi-modal parameters including multi-Data driven algorithms.
A complete biomarker kit to address the issue of healthy aging should contain biomarkers of pre-frailty and frailty capable of predicting individual health trajectory and the risk of care dependency. Initial biomarker sets will be tested as composite biomarkers on the INSPIRE longitudinal cohort but also will reveal putative targets for innovative therapeutic strategies that could be investigated and tested in animal models. From this first round of development and evaluation, sequential rounds will be conducted to refine and adapt these initial sets with definite validation or invalidation, inclusion of new biomarkers and a better definition of the relevant thresholds (Figure 3). Thus, the first Icope version is particularly appropriate for a large screen of individuals in primary care, but it is primarily adapted to individuals of more than 60 years old. So, if the goal is to be able to predict the risk for subsequent abnormal declines in capacity or frailty at an early age, the present evaluation should be adapted for young people prior to overt losses of capacity, it is imperative that such future measurements include parameters of resilience to challenges and stresses. For the future we can reasonably envision other types of approaches, such as artificial intelligence based image analysis methodologies. Indeed, facial recognition and locomotion gait analyses represent promising bio-visual biomarker tools, because of their accuracy and ease of implementation at the whole population level, using different digital applications.

Figure 3. INSPIRE strategy

INSPIRE will used the shared bio-resources (see figure 2) and mainly focalize on two WHO functions namely cognition (C) and mobility (M).


The innovative paradigm of mixing the clinical perspective provided by the WHO innovative vision of tissue homeostasis with the geroscience perspective and modern digital tools would lead to provide a composite score to determine the biological age of each individual. Artificial intelligence will allow to achieve this objective and play a transformative role in diagnosis and therapeutic strategy. Indeed, data analysis in complex datasets comprising multiple clinical and biological variables repeated over time is extremely difficult for humans. The development of artificial intelligence in the medical field, through machine learning methodologies adapted to the need to take into account temporal data, makes it possible to envisage tools that may in the near future help the clinician to identify relevant biomarkers and to make outcome predictions. One of the keys to success in achieving this objective consists in the development of data sharing with standardized high quality data sets. This is made possible by a major scientific and financial investment by both research institutions and local authorities in INSPIRE project. It is the beginning of a long effort absolutely required to face one of the major societal challenges of our time.


Authors contribution: All authors of the paper “The INSPIRE research initiative: a program for GeroScience and healthy aging research going from animal models to humans and the healthcare system” participated to the writing of the manuscript. LC, LP and IA supervised the writing and the final version of the manuscript.
Acknowledgments: The Inspire Program is supported by grants from the Region Occitanie/Pyrénées-Méditerranée (Reference number: 1901175), the European Regional Development Fund (ERDF) (Project number: MP0022856).

Conflict of interest: All authors of the paper “The INSPIRE research initiative: a program for GeroScience and healthy aging research going from animal models to humans and the healthcare system” declare no Conflicts of Interest related to this manuscript.

Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.



1. Thuault, S., Reflections on aging research from within the National Institute on Aging. Nature aging, 2021: p. 14-18.
2. Beard, J.R., et al., The World report on ageing and health: a policy framework for healthy ageing. Lancet, 2016. 387(10033): p. 2145-2154.
3. Clegg, A., et al., Frailty in elderly people. Lancet, 2013. 381(9868): p. 752-62.
4. Dent, E., et al., Management of frailty: opportunities, challenges, and future directions. Lancet, 2019. 394(10206): p. 1376-1386.
5. WHO, World report on ageing and health. Global strategy and action plan on ageing and health. J Frailty Aging, 2015.
6. Cesari, M., et al., Evidence for the Domains Supporting the Construct of Intrinsic Capacity. J Gerontol A Biol Sci Med Sci, 2018. 73(12): p. 1653-1660.
7. At, J., et al., Frailty and the prediction of dependence and mortality in low- and middle-income countries: a 10/66 population-based cohort study. BMC Med, 2015. 13: p. 138.
8. Van der Vorst, A., et al., Correction: Limitations in Activities of Daily Living in Community-Dwelling People Aged 75 and Over: A Systematic Literature Review of Risk and Protective Factors. PLoS One, 2017. 12(1): p. e0170849.
9. Beard, J.R., et al., The structure and predictive value of intrinsic capacity in a longitudinal study of ageing. BMJ Open, 2019. 9(11): p. e026119.
10. Ferrucci, L., et al., Measuring biological aging in humans: A quest. Aging Cell, 2020. 19(2): p. e13080.
11. Khosla, S., et al., The role of cellular senescence in ageing and endocrine disease. Nat Rev Endocrinol, 2020. 16(5): p. 263-275.
12. Tabue-Teguo, M., et al., Frailty in elderly: a brief review. Geriatr Psychol Neuropsychiatr Vieil, 2017. 15(2): p. 127-137.
13. Kirkwood, T.B., A systematic look at an old problem. Nature, 2008. 451(7179): p. 644-7.
14. Belsky, D.W., et al., Quantification of biological aging in young adults. Proc Natl Acad Sci U S A, 2015. 112(30): p. E4104-10.
15. Ahadi, S., et al., Personal aging markers and ageotypes revealed by deep longitudinal profiling. Nat Med, 2020. 26(1): p. 83-90.
16. Weston, A.D. and L. Hood, Systems biology, proteomics, and the future of health care: toward predictive, preventative, and personalized medicine. J Proteome Res, 2004. 3(2): p. 179-96.
17. Hood, L., et al., Systems biology and new technologies enable predictive and preventative medicine. Science, 2004. 306(5696): p. 640-3.
18. Hood, L., R. Balling, and C. Auffray, Revolutionizing medicine in the 21st century through systems approaches. Biotechnol J, 2012. 7(8): p. 992-1001.
19. Justice, J.N., et al., A framework for selection of blood-based biomarkers for geroscience-guided clinical trials: report from the TAME Biomarkers Workgroup. Geroscience, 2018. 40(5-6): p. 419-436.
20. Li, Q., et al., Homeostatic dysregulation proceeds in parallel in multiple physiological systems. Aging Cell, 2015. 14(6): p. 1103-12.
21. Guerville, F., et al., Revisiting the Hallmarks of Aging to Identify Markers of Biological Age. J Prev Alzheimers Dis, 2020. 7(1): p. 56-64.
22. Sierra, F., The Emergence of Geroscience as an Interdisciplinary Approach to the Enhancement of Health Span and Life Span. Cold Spring Harb Perspect Med, 2016. 6(4): p. a025163.
23. Lopez-Otin, C., et al., Metabolic Control of Longevity. Cell, 2016. 166(4): p. 802-821.
24. Lopez-Otin, C., et al., The hallmarks of aging. Cell, 2013. 153(6): p. 1194-217.
25. Keenan, C.R. and R.S. Allan, Epigenomic drivers of immune dysfunction in aging. Aging Cell, 2019. 18(1): p. e12878.
26. Lu, L., et al., Cardiac fibrosis in the ageing heart: Contributors and mechanisms. Clin Exp Pharmacol Physiol, 2017. 44 Suppl 1: p. 55-63.
27. Palmer, A.K. and J.L. Kirkland, Aging and adipose tissue: potential interventions for diabetes and regenerative medicine. Exp Gerontol, 2016. 86: p. 97-105.
28. P. De Souto Barreto, S.G., I. Ader, S. Andrieu, L. Casteilla,N. Davezac, C. Dray, N. Fazilleau, P. Gourdy, R. Liblau, A. Parini, P. Payoux,L. Pénicaud, C. Rampon, Y. Rolland, P. Valet, N. Vergnolle, B. Vellas, The Inspire Research Initiative: A Program for Geroscience and Healthy Aging Research going from Animal Models to Humans and the Healthcare System. The Journal of Frailty & Aging, 2020.
29. S Guyonnet, Y.R., C Takeda, P-J Ousset, I Ader, N.Davezac, C Dray, N Fazilleau, P Gourdy, R Liblau, A Parini, P Payoux, L Pénicaud, C Rampon, P Valet, N Vergnolle, S Andrieu,P de Souto Barreto, L Casteilla, B Vellas., The INSPIRE Bio-resource Research Platform for Healthy Aging and Geroscience: Focus on the Human Translational Research Cohort (The INSPIRE-T Cohort). The Journal of Frailty & Aging, 2020.
30. N. Tavassoli, A.P., C. Berbon, J. de kerimel, C. Lafont, P. De souto Barreto, S. Guyonnet, C. Takeda, I. Carrie, D. Angioni, F. Paris, C. Mathieu, P.J. Ousset, L. Balardy, T. Voisin, S. Sourdet, J. Delrieu, V. Bezombes,S. Andrieu, F. Nouhashemi, Y. Rolland, M.E. Soto, J. Beard, Y. Sumi, I. Araujo carvalho, B. Vellas, Framework implementation of the inspire icope-care program in collaboration with the world health organization (who) in the occitania region. The Journal of Frailty & Aging, 2020.
31. Singh, P.P., et al., The Genetics of Aging: A Vertebrate Perspective. Cell, 2019. 177(1): p. 200-220.
32. Valenzano, D.R., et al., The African Turquoise Killifish Genome Provides Insights into Evolution and Genetic Architecture of Lifespan. Cell, 2015. 163(6): p. 1539-54.
33. Poeschla, M. and D.R. Valenzano, The turquoise killifish: a genetically tractable model for the study of aging. J Exp Biol, 2020. 223(Pt Suppl 1).
34. Y Santin, S.L., I Ader, S Andrieu, N Blanchard, A Carrière, L Casteilla, B Cousin, N Davezac, P De Souto Barreto, C Dray, N Fazilleau, D Gonzalez-Dunia, P Gourdy, S Guyonnet, N Jabrane-Ferrat, O Kunduzova, F Lezoualc’h, R Liblau, L O. Martinez, C Moro, P Payoux, L Pénicaud, V Planat-Bénard, C Rampon, Y Rolland, Joost-P Schanstra, F Sierra, P Valet, A Varin, N Vergnolle, B Vellas, J Viña, B P. Guiard, A Parini, Towards a large-scale assessment of relationship between biological and chronological aging: The INSPIRE Mouse Cohort. The Journal of Frailty & Aging, 2020.
35. Creevy, K.E., et al., The Companion Dog as a Model for the Longevity Dividend. Cold Spring Harb Perspect Med, 2016. 6(1): p. a026633.
36. Kaeberlein, M., K.E. Creevy, and D.E. Promislow, The dog aging project: translational geroscience in companion animals. Mamm Genome, 2016. 27(7-8): p. 279-88.
37. Desdin-Mico, G., et al., T cells with dysfunctional mitochondria induce multimorbidity and premature senescence. Science, 2020. 368(6497): p. 1371-1376.
38. Goecks, J., et al., How Machine Learning Will Transform Biomedicine. Cell, 2020. 181(1): p. 92-101.
39. Zhavoronkov, A., et al., Artificial intelligence for aging and longevity research: Recent advances and perspectives. Ageing Res Rev, 2019. 49: p. 49-66.



C.P. Launay1,2,3, L. Cooper-Brown2,3, V. Ivensky2,4, O. Beauchet1,2,4,5,6,7

1. Department of Medicine, Division of Geriatric Medicine, Sir Mortimer B. Davis – Jewish General Hospital and Lady Davis Institute for Medical Research, McGill University, Montreal, Quebec, Canada; 2. Centre of Excellence on Longevity of McGill Integrated University Health and Social Services Network, Quebec, Canada; 3. Faculty of Medicine, McGill University, Montreal, Quebec, Canada; 4. Faculty of Medicine, University of Montreal, Montreal, Quebec, Canada; 5. Departments of Medicine and Geriatrics, University of Montreal, Montreal, Quebec, Canada; 6. Research Center of the Geriatric University institute of Montreal, Montreal, Quebec, Canada; 7. Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore.

Corresponding Author: Cyrille Launay, MD, PhD; Department of Medicine, Division of Geriatric Medicine, Sir Mortimer B. Davis – Jewish General Hospital, McGill University, 3755 chemin de la Côte-Sainte-Catherine, Montréal, QC H3T 1E2, Canada; E-mail: cyrille.launay@mcgill.ca; Phone: (+1) 514-340-8222, #23837; Fax: (+1) 514-340-7547

J Frailty Aging 2021;in press
Published online March 18, 2021, http://dx.doi.org/10.14283/jfa.2021.9


Key words: Aged, epidemiology, community dwelling, frailty, COVID-19.

Dear Editor,

The COVID-19 pandemic had severe consequences for older adults. First, COVID-19 was associated with more severe medical complications and an increased mortality rate in older compared to younger adults (1). Second, home confinement, an intervention that reduces the spread of COVID-19, was associated with adverse consequences for the older community-dwelling population (2). It broke down social networks and the continuum of primary care, resulting in medication or food delivery issues, psychological fallout and increasing frailty risks (3). Frailty assessment provides insight into the degree of older community dwellers’ health status vulnerability, social isolation and adverse health event risks, and it should be assessed before interventions are proposed (3). We designed a short assessment tool known as “Evaluation SOcio-GERiatrique” (ESOGER) for Montreal’s homebound community-dwelling older adults (3). In a phone call, ESOGER briefly assessed frailty and social isolation and provided recommendations, facilitating contact with health or social care providers who initiate appropriate health and social care plans (3). This study aims to examine the longitudinal effects of ESOGER on frailty and social isolation in Montreal’s homebound community-dwelling older adults.
Adopting a pre-post intervention, single arm, prospective and longitudinal design, this experimental study enrolled 119 community-dwelling older adults (70.0% female) for a mean follow-up period of 23.2±13.6 days. Selection criteria were being age 70 and over, being homebound, living in Montreal, not being involved in an experimental study and agreeing to participate. The assessments (i.e., baseline and follow-up) were performed using ESOGER. ESOGER assessed frailty using the 6-item brief geriatric assessment (BGA), with scores ranging from 0 (no frailty) to 14 (severe frailty) (4), as well as social isolation through accessibility to essential services (i.e., medication and food delivery, home care) and contact with individuals (family, neighbours, friends, healthcare or social professionals) over the phone or in person. When services and/or social contact were absent, ESOGER provided recommendations addressing social isolation and/or care disruption. Evaluated outcomes were: 1) the difference between baseline and follow-up BGA scores (/14) calculated as (follow-up score – baseline score) / ((follow-up score + baseline score) /2); 2) new-onset moderate to severe frailty staging defined by scores ≥5/14 at follow-up, but not baseline, assessment; and 3) unsuccessful or successful implementation of ESOGER recommendations, using two proxies: a) new-onset social isolation and care disruption (i.e., no social isolation and no care disruption at baseline assessment with social isolation and care disruption at follow-up assessment) when ESOGER recommendations were not implemented; or b) addressed social isolation and care disruption (i.e., social isolation and care disruption at baseline assessment with no social isolation and no care disruption at follow-up assessment) when ESOGER recommendations were implemented. Linear and logistic regressions were used to examine the association between frailty (mean BGA score and moderate-to-severe frailty stage used as dependent variables in separate models) and ESOGER recommendations (used as independent variables) adjusted to participants’ baseline characteristics (Age≥ 85, sex, polypharmacy, length of follow-up). P-values less than 0.05 were considered statistically significant. All statistics were performed using SPSS (version 24.0; SPSS, Inc., Chicago, IL). The protocol received Jewish General Hospital (McGill University, Quebec, Canada) Research Ethics Committee approval.
Baseline characteristics of participants are shown in the Table 2. In our study, social isolation decreased significantly from baseline to follow-up assessments (62.2% versus 38.7% with P≤0.001), whereas frailty increased significantly (BGA score 4.1±3.2 and moderate-to-severe frailty 47.9% versus BGA score 5.1±3.5 and moderate-to-severe frailty 58.0% with P≤0.001). As shown in Table 1, the implementation of ESOGER recommendations was not associated with significant variations in frailty (P>0.166), whereas the absence of the implementation of ESOGER recommendations was associated with a significant increase in frailty (P≤0.038).

Table 1. Multiple linear and regression models showing the association between frailty (mean BGA score and moderate-to-severe frailty stage used as separated dependent variables) and ESOGER recommendations (used as independent variables), adjusted for participants’ baseline characteristics (n=119)

ESOGER: “Evaluation SOcio-GERiatrique”; ß: Coefficient or regression beta; CI: Confidence interval; *: Difference between baseline and follow-up score ranged from 0 (no frailty) to 14 (severe frailty) and calculated from the formula: ((follow-up score – baseline score) / ((follow-up score + baseline score) /2); †: Brief geriatric assessment score ≥5/14; significant P-values (i.e., ≤0.05) in bold.

Table 2. Baseline characteristics of participants (n=119)

BGA: brief geriatric assessment; SD: standard deviation; CI: Confidence interval; *: no formal (i.e., health and/or social professional) or informal (i.e., family and/or friends) help; †: inability to give the current year and/or month; ‡: Number of different medications taken daily ≥5; §: Score 6-item BGA >6 with score ranging from 0 (no frailty) to 14 (severe frailty)


Home confinement was associated with increased frailty in the studied sample of older community dwellers, and this change in frailty was dependent on ESOGER recommendation implementation. There was a significant increase in frailty when ESOGER recommendations were not implemented, whereas no change was observed when they were implemented. The negative impact of social isolation on older adults’ health condition has previously been reported (5,6). This effect could explain the increased frailty observed among participants in this study for whom ESOGER recommendations were not implemented. The absence of significant changes to frailty when ESOGER recommendations were implemented shows that telemedicine may be an effective approach to sustaining the continuum of care in vulnerable homebound patients during crises like COVID-19. The main limitation of our study is the small sample size of participants, and thus studies recruiting greater number of participants are needed to confirm the results of our pilot study. In addition, further developing such interventions may help to remotely stabilize patients and to avoid seeing them in consultation.


Conflicts of Interest: None declared by the authors.



1. Du RH, Liang LR, Yang CQ, Wang W, Cao TZ, Li M, Guo GY, Du J, Zheng CL, Zhu Q, Hu M, Li XY, Peng P, Shi HZ. Predictors of mortality for patients with COVID-19 pneumonia caused by SARS-CoV-2: a prospective cohort study. Eur Respir J. 2020;55(5):2000524.
2. https://blogs.bmj.com/bmj/2020/06/15/covid-19-will-be-followed-by-a-deconditioning-pandemic/ (July 07, 2020 date last accessed)
3. Beauchet O, Cooper-Brown L, Ivensky V, Launay CP. Telemedicine for housebound older persons during the Covid-19 pandemic. Maturitas 2020 [Epub ahead of print]
4. Launay CP, de Decker L, Kabeshova A, Annweiler C, Beauchet O. Screening for older emergency department inpatients at risk of prolonged hospital stay: the brief geriatric assessment tool. PLoS One. 2014;9(10): e110135.
5. Brooke J, Jackson D. Older people and COVID-19: Isolation, risk and ageism. J Clin Nurs. 2020;29(13‑14): 2044‑2046
6. Kf B, Jl H, Ta D. Preventing Frailty Progression during the COVID-19 Pandemic. J Frailty Aging. 2020;9(3):130‑131



Q. Roquebert1,4, J. Sicsic1, B. Santos-Eggimann2, N. Sirven1, T. Rapp1,3


1. LIRAES (EA4470), Université de Paris; Paris, France; 2. Centre universitaire de médecine générale et santé publique (Unisanté), Université de Lausanne, Switzerland; 3. LIEPP, Sciences Po, France; 4. Université de Strasbourg, Université de Lorraine, CNRS, BETA, 67000 Strasbourg, France
Corresponding author: Jonathan Sicsic (PhD), LIRAES, 45 rue des Saints-Pères, Université de Paris 75006 Paris, Jonathan.sicsic@u-paris.fr. +33(0)177219361

J Frailty Aging 2021;10(3)272-280
Published online March 15, 2021, http://dx.doi.org/10.14283/jfa.2021.7



This systematic literature review documents the link between frailty or sarcopenia, conceptualized as dimensions of physical health, and the use of long-term care services by older individuals. Long-term care services include formal and informal care provided at home as well as in institutions. A systematic review was performed according to PRISMA requirements using the following databases: PubMed-Medline, Embase, CINAHL, Web of Science, and Academic Search Premier. We included all quantitative studies published in English between January 2000 and December 2018 focusing on individuals aged 50 or more, using a relevant measurement of sarcopenia or physical frailty and a long-term care related outcome. A quality assessment was carried out using the questionnaire established by the Good Practice Task Force Report of the International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Five subsets of long-term care outcome were considered: 1/ nursing home placement (NHP), 2/ nursing home short stay (NHSS), 3/ formal personal care (FPC), 4/ formal home help (FHH), 5/ informal care (IC). Out of 1943 studies, 17 were finally included in the review. With some studies covering several LTC outcomes, frailty and / or sarcopenia were associated with increased LTC use in 17 out of 26 cases (NHP: 5/6, NHSS: 3/4, FPC: 5/7, FHH: 1/4, IC: 3/5) The association was not consistent in 5 cases (NHP: 1/6, NHSS: 1/4, FPC: 2/7, FHH: 0/4, IC: 1/5) and the association was either not significant or the results inconclusive in the remaining 9 cases. Overall, while results on sarcopenia are scarce, evidence support a positive association between frailty and LTC use. The evidence is stronger for the association of physical frailty with nursing home placement / short stay as well as on FPC. There is less (more heterogeneous) evidence regarding the correlation between physical frailty and FHH or IC use. Results need to be confirmed by more advanced statistical methods or design based on longitudinal data.

Key words: Frailty, sarcopenia, long-term care, systematic review, nursing home, informal care, formal care.



In a context of aging populations, the demand for long-term care (LTC) is increasing rapidly. Referring to the relatively large range of services needed in the long run to individuals because of their functional limitations, LTC can be provided in the community by relatives (informal care), professional nurses or care workers (formal care), or in specific institutions such as in nursing homes. LTC spending represents an increasing share of Western countries’ Gross Domestic Product (GDP) and most of these expenditures are funded by public mechanism schemes involving taxes and/or social insurance (1, 2). Projections suggest that LTC public expenditures should increase by 69% at least (in a healthy aging scenario) reaching 2.7 % of GDP in the European Union in 2070 (3).
Beyond disability status, increasing attention is being paid to the pre-disability condition as conceptualized by frailty and sarcopenia. While frailty is defined as a vulnerable health status resulting from the reduction of individuals’ physiological reserve, sarcopenia indicates the loss of muscle mass and muscle strength (4). In practice, frailty and sarcopenia are closely related, and it may be difficult to empirically distinguish these concepts. Indeed, frailty and sarcopenia often overlap; most frail older people exhibit sarcopenia, and some older people with sarcopenia are also frail (5). Frailty is relatively easy to measure in the clinical practice. Some readily accessible screening tests also exist for sarcopenia, such as the SARC-F (6). Yet precise measurement of sarcopenia involves more complex instruments such as dual x-ray absorptiometry scan to measure appendicular lean mass (ALM) and classify individuals according to pre-defined cut-points (7, 8). Overall, both are prevalent in old age and associated with adverse health events. Efforts to adapt healthcare and social systems to aging populations might benefit from taking into account the effects of frailty and sarcopenia on health and long-term care services utilization.
A growing literature has been exploring the relationship between frailty or sarcopenia and LTC use since the early 2000’s. However, it is difficult to have a clear perspective of the overall findings and the quality of evidence stemming from this literature. Indeed, publications used various measures of frailty, explored sometimes different LTC outcomes (formal care use, informal care use, nursing home use etc.), applied different statistical modelling techniques, sampling methods etc. Therefore, there is need for a systematic review to assess the quality of the methods, classify findings according to different outcomes, and provide an overall perspective of the relationship between frailty, sarcopenia and LTC use. To the best of our knowledge, such an analysis of the literature has not been done yet.
Three prior reviews of the literature have provided evidence that frailty is associated with a higher risk of hospitalization (9, 10) and nursing home placement (10, 11). These studies have found moderate evidence of the association between frailty and hospitalization or institutionalization. Results on sarcopenia are more limited: it was positively correlated with hospitalization in one study (12) but no study examined its link with nursing home entry.
The present review fills a gap in the literature by focusing on the link between frailty or sarcopenia and LTC, not only including nursing home admission but also formal and informal care provided in the community. Moreover, we evaluated the quality of evidence and summarized the current knowledge on frailty and sarcopenia as correlates of LTC use in populations aged 50 years and over. This article addresses the following question: are frailty and sarcopenia, conceptualized as dimensions of physical health, associated with the use of long-term care services by older individuals? Our systematic review is the first to focus on various LTC outcomes, and to provide an evaluation of both the results of the literature and the quality of the published evidence. The following aspects of LTC utilization are investigated: first, informal care, referring to the unpaid care provided by relatives; second, formal care, referring to the paid care provided by professionals, either nurses or personal care workers; finally, we also consider LTC provided in nursing homes.



The protocol for our systematic review was prepared and registered on the International Prospective Register of Systematic Reviews (PROSPERO; Ref CRD42020137212). The systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), and with close adherence to the Cochrane Handbook for Systematic Reviews of Interventions. The resulting PRISMA checklist (see Table A.1, Appendix A.1, in Supplemental Materials), and the flowchart (see Figure 1) depict the stages involved in the selection process. The following subsections briefly discuss the methodology.

Figure 1
Flowchart of screening phases


Data Sources and Search Strategy

The search was conducted from April 2019 to June 2019 on the following databases:
PubMed-Medline, Embase, CINAHL (Academic Search Premier/EconLit), Web of Science. A preliminary search was conducted in order to develop a search strategy. The final search strategy is described in Appendix A.2. We employed some methodological components of the Effective Practice and Organisation of Care (EPOC) group’s search strategy, combined with selected MeSH terms and free text terms related to the PICOS (i.e. population, intervention, comparator, outcome and study design) elements. Studies published in English between January 2000 and December 2018 were collected. Also, the reference lists of the retrieved articles were examined to look for further relevant literature (through a ‘pearling procedure’).

Study Selection

Two researchers independently screened the titles and abstracts of papers published in English between 2000-2018. The selection of relevant papers was completed according to predefined eligibility criteria: i) the population of interest was composed of people aged 50 or more; ii) sarcopenia or physical frailty were explicitly measured by an identified tool (phenotype, index, scale, indicator) measuring a physical or biomedical component (thus excluding papers focusing only on psychological or social conceptions of frailty); iii) the comparators were expected to be populations with different levels of frailty or sarcopenia, including the comparison of a frail (respectively sarcopenic) population and a non-frail (respectively non sarcopenic) population; iv) the outcome should be a long-term care outcome (formal care, informal care, nursing home admission); v) we focused on quantitative research papers and excluded case reports, qualitative works or other studies without design (discussion, commentary, workshop report, systematic review).
A pilot phase was conducted to check that these criteria were precise enough. In the selection process, the information regarding the paper (authorship, institutions, journal titles and year of publication) were not blinded: even though inclusion decisions could be affected by these parameters, blind assessment has been shown to be costly and have a limited value in the study selection (13). Differences in screening results were solved by dialogue between the five members of the research team involved in the preparation of the manuscript.

Study design

We searched for cross-sectional, cohort, and pre-post studies, using a validated measure of physical frailty or sarcopenia. There were no restrictions in terms of study setting, or characteristics of sample population (except the age condition).

Typology of LTC outcomes

The main outcomes were categorized as utilization of any type of LTC services, at the intensive (i.e., probability of use) or extensive (i.e., volume of use, volume of costs) margins. We considered five different categories: 1/ Nursing home placement (NHP), regrouping any long or permanent stay in a nursing home or LTC institution, 2/ Nursing home short stay (NHSS), 3/ Formal personal care (FPC), regrouping home care services explicitly stated (e.g., ‘home care’, ‘nursing care’), 4/ Formal home help (FHH), consisting in any type of non-care help (e.g., help with meals of household duties), and 5/ Informal care (IC), provided by spouse, children or relatives.

Data Extraction, Analysis and Synthesis

Two independent reviewers (QR, JS) screened each study to identify the abstract, title, keywords, and concepts reflecting both the study’s contribution and research context. Databases were searched for a combination of three groups of keywords, referring to the variables of interest (frailty and sarcopenia), the outcomes of interest (long-term care, included formal care provided by nurses and care workers, informal care and nursing homes), and a third group complementing the request with a condition relating to utilization (use, utilization, visits, consumption). The complete search strategy is described in Appendix A.2.
Once the search strategy was completed, the full text of the relevant studies was retrieved and assessed by the two review authors with respect to the inclusion/exclusion criteria. When consensus was difficult to reach, a third review author (TR) assessed the study. The data were extracted by one reviewer (QR or JS) and checked by a second reviewer (JS or QR). The raw data from each manuscript included authors’ names, year of publication, title, journal, country and regions (if specified), study setting, study design, study population, participant demographics (age, share of women), details on the physical frailty measure used, type of LTC outcome use, raw and adjusted estimates, and risk of bias assessment details. The collected data were organized manually and tabulated using standardized forms. Given the heterogeneity of studies in terms of type of outcome and reporting (i.e., odds ratio, risk ratio, hazard ratio, marginal effects, etc.) we were unable to pool the results and conduct a meta-analysis. Therefore, we provide a comparative summary of findings for the main outcomes, using measures of the association in the same way as they were reported.

Risk of bias and quality assessment

The risk of bias was assessed using the Report from the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) Task Force which provides a questionnaire to assess the relevance and credibility of observational studies (14). The credibility of each study was assessed by scoring each study across six quality domains: Design (8 items), Data (4 items), Analysis (3 items), Reporting (7 items), Interpretation (8 items), and Conflict of Interest (2 items). Two independent reviewers assessed each study (QR and JS). When a difference persisted between them, a third reviewer (TR) assessed the study and came to an agreement. Details regarding the grid and the coding strategy for each item are provided in Appendix A.3: Table A.3.1 assesses the credibility with respect to the study design domain, Table A.3.2 with respect to the data, Table A.3.3 with respect to the analysis, Table A.3.4 with respect to reporting, Table A.3.5 with respect to interpretation.
The Task Force Report (14) identified two specific criteria, where failure to address these criteria would indicate a “fatal flaw” and require particular caution in interpreting the results. One criterion concerned the Data: “was exposure outcome valid?” and the second concerned the Analysis: “was there a thorough assessment and control of confounding?”. Because the first criterion was already included as a study selection criterion, we only considered the second criteria as a potential fatal flaw. In case no multivariate regression analysis was conducted (adjusting for potential confounders), we classified the study as providing “not enough evidence” to conclude on the relationship being investigated.



Results of screening phases

The literature search yielded 2,534 titles in the preliminary phase. Figure 1 depicts the flow chart associated to the screening procedure. Removal of duplicates resulted in 1,933 potentially relevant papers. Then, at the screening stage, 1,920 studies were removed after applying the exclusion criteria. The pearling procedure allowed us to identify 4 more relevant papers. In the end, our systematic review comprised 17 studies on the relationship between frailty and /or sarcopenia and LTC use (16 studies using frailty, and 1 study using sarcopenia as main independent variable). Main outcomes, descriptions, and statistics (when available) are presented in Table 1. A complete summary of findings is provided in Appendix A.4, including information on the frailty/sarcopenia measurements, outcomes considered and statistical uncertainty around point estimates.

Table 1
Summary of study design and outcomes

a. Time frame only for longitudinal / cohort studies; *: Significant at the 5% level; b. OR calculated by authors based on raw coefficients (log OR); Abbreviations. HR: hazard ratio; OR: odds ratio; RR: risk ratio; ME: marginal effect; raw coefficient. NA: not available. ALM: appendicular lean mass; Note: details regarding measurements and statistical uncertainty around point estimates are provided in Appendix A.3.


As regard settings, 12 of the 17 reviewed studies took place in the general (community-dwelling) population, (15–26). Two studies considered residents of housing facilities or long-stay home care clients (27, 28) and two studies focused on the specific population of adults with intellectual disabilities (29) or cognitive impairments (30). One study focused on patients with diagnostic of Alzheimer disease (AD) (31). Interest in the association between physical frailty, sarcopenia, and LTC use seems to have arisen since 2008, with an increasing trend since. Most of the research was conducted in North America (n=4), Australia (n=2), and central Europe (n=11) (see Table A.5.1 and Figure A.5.1 for more details on the geographical distribution of studies).

Quality assessment

Overall, the mean relative score achievement (quality) was 84% (min= 67%, max= 100%) for the subdomain design, 97% (min= 67%, max= 100%) for the subdomain data, 28% (min= 0%, max= 67%) for the subdomain analyses, 84% (min= 57%, max= 100%) for the subdomain reporting, and 82% (min= 75%, max= 100%) for the subdomain interpretation. Appendix A.3 provides details on the scoring of each paper according to each domain. In the quality assessment, an important issue lies in the way potential confounders are dealt with: few studies used of an appropriate design (Design, Question 5, 2 studies coded 1 out of 17) and they were scarcely discussed (Interpretation, Question 4, 5 studies coded 1 out of 17). Overall, no study had all criteria 100% fulfilled. There were no large discrepancies in terms of quality according to the class of LTC outcome, though studies investigating FHH had lower scores on average, and studies focusing on LTC placements (short or permanent) tended to have higher quality (see Figures A.3.1 in Appendix A.3).
A “fatal flaw” (assessed by the absence of control or assessment of potential confounders) was diagnosed for four studies (19, 25, 26, 29). Two of the studies focused on formal home care (19, 29), among which one study additionally focused on home care (19), and two studies focused on informal care (25, 26).

Association between frailty, sarcopenia and LTC use

Frailty and LTC use

Overall, 10 studies reported a positive association between frailty and LTC use (15–18, 23, 24, 27–29, 31) (Table 2). Conversely, only two studies showed no significant association between frailty and LTC use (21, 22). In three studies, there was not enough evidence to conclude on the relationship between frailty and LTC use (e.g., because no multivariate regression was run) (19, 25, 26). Among the 10 studies reporting positive associations, three found mixed results (17, 24, 31), which varied according to the frailty measure used (see Table 1 for a summary of results according to each frailty measure used).

Table 2
Summary of results per class of outcome

Legend: NS nonsignificant; (+) Outcome increased significantly; NEE: not enough evidence (e.g., no adjusted multivariate regression conducted); (*) inconsistent association (the association varies according to the frailty measure considered); In each study, the LTC outcomes were groups into one of the five categories: 1/ Nursing home placement (NHP), 2/ Nursing home short stay (NHSS), 3/ Formal personal care (FPC), 4/ Formal home care (FHC), 5/ Informal care (IC). Several outcomes per study are possible.


Sarcopenia and LTC use

Only one study focused on the relationship between sarcopenia and LTC use [20]. Using a 3-classes characterization of sarcopenia defined as sarcopenia I (i.e., low appendicular lean mass / ALM alone), sarcopenia II (low ALM with weakness), or sarcopenia III (low ALM with weakness and poor gait speed), the authors found a positive association between sarcopenia (either I, II, or III) and NHP. Moreover, the relationship was stronger when sarcopenia severity was higher (the effect size was higher for sarcopenia III compared to sarcopenia I or sarcopenia II).

Heterogeneity of results according to the LTC outcome

The number of studies having focused on each type of LTC outcome was distributed as follows (one study may have analysed multiple LTC outcomes): six studies on nursing home placement (NHP) (15, 20, 22, 27, 28, 31), four studies on nursing home short stay (NHSS) (16, 17, 21, 23), seven studies on formal personal care (FPC) (16–19, 24, 25, 29), four studies on formal home help (FHH) (16, 19, 25, 26), and five studies on informal care (IC) (17, 24–26, 30). Among the six studies investigating the association between frailty and NHP, 5 studies found a significant positive association, including one with mixed results (i.e., depending on frailty measurement) (see Table 1). Regarding NHSS, three studies confirmed a positive significant association, including one with mixed results. Regarding FPC, five studies confirmed a positive significant association, including two with mixed results. The evidence was less compelling for the two other LTC outcomes. Regarding FHH, only one study found a positive significant association; it was not possible to conclude for the other three studies because no multivariate regression models were estimated. Last, concerning IC, three studies found a positive significant association with frailty, including one with mixed results.



The purpose of this systematic review was to assess the association between physical frailty or sarcopenia and LTC use. Our paper is the first to systematically review the results of these studies and assess their strength and weaknesses using a validated quality assessment grid (14). We found that most of the reviewed studies reported a positive association of physical frailty and LTC outcomes. Moreover, there is evidence showing an association between frailty and LTC use for each class of LTC outcome. Specifically, there seems to be already reasonable evidence to conclude on a positive relationship of physical frailty with nursing home placement / short stay as well as on FPC. However, there is less (more heterogeneous) evidence regarding the association between physical frailty and FHH or IC use. Besides, only one study investigated the association between sarcopenia and LTC use, and found a positive association for NHP. Further studies should thus focus on the particular LTC use outcomes with insufficient evidence, and on sarcopenia.
We found important heterogeneity across studies in terms of i) type of frailty indicator used, ii) statistical analysis, and iii) reporting of results. This heterogeneity made it difficult to compare or transpose the results from one study to another. Overall, 12 different physical frailty measures and three sarcopenia indicators were used (altogether in one study). Almost all studies used multidimensional frailty indexes (most of which based on self-reported data), with a majority using either the original 5 items Fried frailty index (with categorisation into robust, pre-frail, frail) (16, 18, 21–24, 30), the Tilburg frailty indicator (17, 22, 26) or composite frailty indexes counting the number (or percentage) of deficits accumulated (27, 28, 31). In one study, sarcopenia was defined using clinical assessment of appendicular lean mass, muscle strength, and gait speed (20).
Most studies used multivariate regression analyses controlling for several of the following covariates: age, gender, morbidity, disability, and some basic sociodemographic characteristics such as education, living arrangements, or income. However, none of the studies included exactly the same number and types of controls, making the comparing of results difficult. Moreover, in only few instances the rationale for including specific controls was provided. One study performed sensitivity analysis with respect to the model specification. We conclude that future studies should carefully explain and justify their model, and perform sensitivity analyses regarding model specification (inclusion or exclusion of covariates). Despite the fact that a majority of studies relied on cohort data, very few used statistical methods designed to reduce unobserved confounding using longitudinal or panel data models (e.g. fixed effects or correlated random effects models). Specifically, only two studies analysed LTC outcome in relation with frailty transitions (19, 24), among which only one used fixed effect model to control for time invariant confounders (24). Future studies investigating the association between frailty and LTC use should consider using more advanced designs and more robust statistical methods such as the one suggested.
Finally, the studies were highly heterogeneous in terms of reporting of results. Studies using binary outcomes reported the results either in terms of raw estimates, odds ratio, or marginal effects. Studies using continuous outcomes (e.g., costs or time between events) reported either hazard ratio or risk ratio. We thus recommend that future studies report results that can be compared across settings, such as reporting average marginal effects, which can be computed using linear or non-linear (e.g., logit, probit) models.
In conclusion, our article is the first systematic literature review to focus on the relationship between frailty and sarcopenia, operationalized using existing validated measures, on various LTC outcomes among people worldwide. Overall, there is evidence suggesting that frailty is a key correlate of LTC use. But there is still a lack of evidence on the relationship between sarcopenia and LTC use, due an insufficient number of studies that investigated that issue. However, the strength of evidence is heterogeneous and varies according to the type of LTC outcome considered. Overall, the quality of studies was judged weak to moderate. Very few studies were considered as providing strong evidence, in particular because of weaknesses in statistical analyses performed, although many studies relied on cohort data. Future studies should address potential confounding using longitudinal models allowing integrating the effect of within-individual heterogeneity and assessing the effect of transitions into frailty (or sarcopenia) on LTC use. Moreover, the studies should perform sensitivity analyses by varying the control variables or using stratified analyses. In particular, it is of importance to know if the effect of frailty varies according to predefined socio-economic categories (32). This information would be useful for the definition of appropriate LTC policies aiming at ensuring equity of access.
With regards to the implications, our systematic literature review supports three main orientations for aging policies. First, eligibility criteria for public allowances should include frailty measures because evidence shows that regardless of the measure used, frailty is associated with increased LTC use. Second, the challenges associated with frailty among older people are global, since correlations between frailty measures and LTC use are similar across countries. This calls for global aging initiatives, to learn from each country’s experience. Third, frailty prevention in older populations is a policy priority. While many countries (France, Germany, Netherlands etc.) implemented disability reforms, future policies should also target frail populations, and enhance innovations to detect frailty in the 65-75 age group.


Funding and acknowledgements: The research leading to these results has received support from the Innovative Medicines Initiative Joint Undertaking under grant agreement n° 115621, resources of which are composed of financial contribution from the European Union’ Seventh Framework Programme (FP7 / 2007-2013) and European Federation of Pharmaceutical Industries and Associations (EFPIA) companies’ in kind contribution.
Conflicts of Interest: None of the authors report any conflict of interest whatsoever.



1. Costa-Font, J., Courbage, C., Swartz, K.: Financing Long-Term Care: ex-ante, ex-post or both? Health Econ. 2015;24, 45–57. https://doi.org/10.1002/hec.3152
2. OCDE: Health at a Glance 2019 : OECD indicators.
3. European Commission, Directorate-General for Economic and Financial Affairs, Economic Policy Committee of the European Communities: The 2018 ageing report: underlying assumptions and projection methodologies. 2017
4. Marcell, T.J.: Sarcopenia: causes, consequences, and preventions. J. Gerontol. A. Biol. Sci. Med. Sci. 2003;58, M911-916
5. Cesari, M., Landi, F., Vellas, B., Bernabei, R., Marzetti, E.: Sarcopenia and Physical Frailty: Two Sides of the Same Coin. Front. Aging Neurosci. 2014;6. https://doi.org/10.3389/fnagi.2014.00192
6. Malmstrom, T.K., Morley, J.E.: SARC-F: a simple questionnaire to rapidly diagnose sarcopenia. J. Am. Med. Dir. Assoc. 2013;14, 531–532. https://doi.org/10.1016/j.jamda.2013.05.018
7. Studenski, S.A., Peters, K.W., Alley, D.E., Cawthon, P.M., McLean, R.R., Harris, T.B., Ferrucci, L., Guralnik, J.M., Fragala, M.S., Kenny, A.M., Kiel, D.P., Kritchevsky, S.B., Shardell, M.D., Dam, T.-T.L., Vassileva, M.T.: The FNIH sarcopenia project: rationale, study description, conference recommendations, and final estimates. J. Gerontol. A. Biol. Sci. Med. Sci. 2014;69, 547–558. https://doi.org/10.1093/gerona/glu010
8. Cawthon, P.M., Peters, K.W., Shardell, M.D., McLean, R.R., Dam, T.-T.L., Kenny, A.M., Fragala, M.S., Harris, T.B., Kiel, D.P., Guralnik, J.M., Ferrucci, L., Kritchevsky, S.B., Vassileva, M.T., Studenski, S.A., Alley, D.E.: Cutpoints for low appendicular lean mass that identify older adults with clinically significant weakness. J. Gerontol. A. Biol. Sci. Med. Sci. 2014;69, 567–575. https://doi.org/10.1093/gerona/glu023
9. Kojima, G.: Frailty as a predictor of hospitalisation among community-dwelling older people: a systematic review and meta-analysis. J. Epidemiol. Community Health. 2016;70, 722–729. https://doi.org/10.1136/jech-2015-206978
10. Wang, S.-Y., Shamliyan, T.A., Talley, K.M.C., Ramakrishnan, R., Kane, R.L.: Not just specific diseases: systematic review of the association of geriatric syndromes with hospitalization or nursing home admission. Arch. Gerontol. Geriatr. 2013;57, 16–26. https://doi.org/10.1016/j.archger.2013.03.007
11. Kojima, G.: Frailty as a Predictor of Nursing Home Placement Among Community-Dwelling Older Adults: A Systematic Review and Meta-analysis. J. Geriatr. Phys. Ther. 2018;2001. 41, 42–48. https://doi.org/10.1519/JPT.0000000000000097
12. Cawthon, P.M., Fox, K.M., Gandra, S.R., Delmonico, M.J., Chiou, C.-F., Anthony, M.S., Sewall, A., Goodpaster, B., Satterfield, S., Cummings, S.R., Harris, T.B.: Do Muscle Mass, Muscle Density, Strength, and Physical Function Similarly Influence Risk of Hospitalization in Older Adults? J. Am. Geriatr. Soc. 2009;57, 1411–1419. https://doi.org/10.1111/j.1532-5415.2009.02366.x
13. Centre for Reviews and Dissemination: Systematic Reviews. CRD’s guidance for undertaking reviews in health care. the Centre for Reviews and Dissemination, York 2009
14. Berger, M.L., Martin, B.C., Husereau, D., Worley, K., Allen, J.D., Yang, W., Quon, N.C., Mullins, C.D., Kahler, K.H., Crown, W.: A Questionnaire to Assess the Relevance and Credibility of Observational Studies to Inform Health Care Decision Making: An ISPOR-AMCP-NPC Good Practice Task Force Report. Value Health. 2014;17, 143–156. https://doi.org/10.1016/j.jval.2013.12.011
15. Rothman, M.D., Leo-Summers, L., Gill, T.M.: Prognostic Significance of Potential Frailty Criteria. J. Am. Geriatr. Soc. 2008;56, 2211–2216. https://doi.org/10.1111/j.1532-5415.2008.02008.x
16. Rochat, S., Cumming, R.G., Blyth, F., Creasey, H., Handelsman, D., Le Couteur, D.G., Naganathan, V., Sambrook, P.N., Seibel, M.J., Waite, L.: Frailty and use of health and community services by community-dwelling older men: the Concord Health and Ageing in Men Project. AGE AGEING. 2010;39, 228–233. https://doi.org/10.1093/ageing/afp257
17. Gobbens, R.J.J., van Assen, M.A.L.M., Luijkx, K.G., Schols, J.M.G.A.: The Predictive Validity of the Tilburg Frailty Indicator: Disability, Health Care Utilization, and Quality of Life in a Population at Risk. The Gerontologist. 2012;52, 619–631. https://doi.org/10.1093/geront/gnr135
18. Hoeck, S., Francois, G., Geerts, J., Van der Heyden, J., Vandewoude, M., Van Hal, G.: Health-care and home-care utilization among frail elderly persons in Belgium. Eur. J. Public Health. 2012;22, 671–677. https://doi.org/10.1093/eurpub/ckr133
19. Bentur N., Sternberg S.A., Shuldiner J.: Frailty transitions in community-dwelling older people. Isr. Med. Assoc. J. 2016;18, 449–453
20. Hirani, V., Blyth, F., Naganathan, V., Le Couteur, D.G., Seibel, M.J., Waite, L.M., Handelsman, D.J., Cumming, R.G.: Sarcopenia Is Associated With Incident Disability, Institutionalization, and Mortality in Community-Dwelling Older Men: The Concord Health and Ageing in Men Project. J. Am. Med. Dir. Assoc. 2015;16, 607–613. https://doi.org/10.1016/j.jamda.2015.02.006
21. Bock, J.-O., König, H.-H., Brenner, H., Haefeli, W.E., Quinzler, R., Matschinger, H., Saum, K.-U., Schöttker, B., Heider, D.: Associations of frailty with health care costs – results of the ESTHER cohort study. BMC Health Serv. Res. 2016;16, 128. https://doi.org/10.1186/s12913-016-1360-3
22. Gonzalez-Colaço Harmand, M., Meillon, C., Bergua, V., Tabue Teguo, M., Dartigues, J.-F., Avila-Funes, J.A., Amieva, H.: Comparing the predictive value of three definitions of frailty: Results from the Three-City study. Arch. Gerontol. Geriatr. 2017;72, 153–163. https://doi.org/10.1016/j.archger.2017.06.005
23. Ferrante, L.E., Pisani, M.A., Murphy, T.E., Gahbauer, E.A., Leo-Summers, L.S., Gill, T.M.: The Association of Frailty With Post-ICU Disability, Nursing Home Admission, and Mortality A Longitudinal Study. Chest. 2018;153, 1378–1386. https://doi.org/10.1016/j.chest.2018.03.007
24. Hajek, A., Bock, J.-O., Saum, K.-U., Matschinger, H., Brenner, H., Holleczek, B., Haefeli, W.E., Heider, D., Koenig, H.-H.: Frailty and healthcare costs-longitudinal results of a prospective cohort study. Age Ageing. 2018;47, 233–241. https://doi.org/10.1093/ageing/afx157
25. Lambotte, D., De Donder, L., Van Regenmortel, S., Fret, B., Dury, S., Smetcoren, A.-S., Dierckx, E., De Witte, N., Verte, D., Kardol, M.J.M., Consortium, D.-S.: Frailty differences in older adults’ use of informal and formal care. Arch. Gerontol. Geriatr. 2018;79, 69–77. https://doi.org/10.1016/j.archger.2018.05.018
26. Verver, D., Merten, H., Robben, P., Wagner, C.: Care and support for older adults in The Netherlands living independently. Health Soc. Care Community. 2018;26, e404–e414. https://doi.org/10.1111/hsc.12539
27. Hogan, D.B., Freiheit, E.A., Strain, L.A., Patten, S.B., Schmaltz, H.N., Rolfson, D., Maxwell, C.J.: Comparing frailty measures in their ability to predict adverse outcome among older residents of assisted living. BMC Geriatr. 2012;12. https://doi.org/10.1186/1471-2318-12-56
28. Campitelli, M.A., Bronskill, S.E., Hogan, D.B., Diong, C., Amuah, J.E., Gill, S., Seitz, D., Thavorn, K., Wodchis, W.P., Maxwell, C.J.: The prevalence and health consequences of frailty in a population-based older home care cohort: a comparison of different measures. BMC Geriatr. 2016;16. https://doi.org/10.1186/s12877-016-0309-z
29. Schoufour, J.D., Evenhuis, H.M., Echteld, M.A.: The impact of frailty on care intensity in older people with intellectual disabilities. Res. Dev. Disabil. 2014;35, 3455–3461. https://doi.org/10.1016/j.ridd.2014.08.006
30. Butler, A., Gallagher, D., Gillespie, P., Crosby, L., Ryan, D., Lacey, L., Coen, R., O’Shea, E., Lawlor, B.: Frailty: a costly phenomenon in caring for elders with cognitive impairment. Int. J. Geriatr. Psychiatry. 2016;31, 161–168. https://doi.org/10.1002/gps.4306
31. Kelaiditi, E., Andrieu, S., Cantet, C., Vellas, B., Cesari, M.: Frailty Index and Incident Mortality, Hospitalization, and Institutionalization in Alzheimer’s Disease: Data From the ICTUS Study. J. Gerontol. Ser. A. 2016;71, 543–548. https://doi.org/10.1093/gerona/glv137
32. Sicsic, J., Ravesteijn, B., Rapp, T.: Are frail elderly people in Europe high-need subjects? First evidence from the SPRINTT data. Health Policy. 2020;124, 865–872. https://doi.org/10.1016/j.healthpol.2020.05.009
33. Berger, M.L., Martin, B.C., Husereau, D., Worley, K., Allen, J.D., Yang, W., Quon, N.C., Mullins, C.D., Kahler, K.H., Crown, W.: A Questionnaire to Assess the Relevance and Credibility of Observational Studies to Inform Health Care Decision Making: An ISPOR-AMCP-NPC Good Practice Task Force Report. Value Health. 2014;17, 143–156. https://doi.org/10.1016/j.jval.2013.12.011



S.J. Wood1, J.S. Bell1,2,3, D.J. Magliano2,4, L. Fanning5, M. Cesari3,6, C.S. Keen1, J. Ilomäki1,2


1. Centre for Medicine Use and Safety, Faculty of Pharmacy and Pharmaceutical Sciences, Monash University, Melbourne, Australia; 2. Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia; 3. National Health and Medical Research Council (NHMRC) Centre of Research Excellence in Frailty and Healthy Ageing, Adelaide, Australia; 4. Baker Heart and Diabetes Institute, 75 Commercial Road, Melbourne, Australia; 5. Eastern Health Clinical School, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia; 6. IRCCS Istituti Clinici Scientifici Maugeri, University of Milan, Milan, Italy

Corresponding Author: Stephen Wood, Centre for Medicine Use and Safety, Faculty of Pharmacy and Pharmaceutical Sciences, Monash University Melbourne, Australia, 3052, Tel: +61 423301741, E-mail: Stephen.Wood@monash.edu
J Frailty Aging 2021;in press
Published online February 29, 2021, http://dx.doi.org/10.14283/jfa.2021.6



Background: The risks of intensive blood glucose lowering may outweigh the benefits in vulnerable older people.
Objectives: Our primary aim was to determine whether age, frailty, or dementia predict discharge treatment types for patients with type 2 diabetes (T2D) and related complications. Secondly, we aimed to determine the association between prior hypoglycemia and discharge treatment types.
Design, Setting and Participants: We conducted a cohort study involving 3,067 patients aged 65-99 years with T2D and related complications, discharged from Melbourne’s Eastern Health Hospital Network between 2012 and 2016.
Measurements: Multinomial logistic regression was used to estimate odds ratios (ORs) with 95% confidence intervals (CI) for the association between age, frailty, dementia and hypoglycemia, and being prescribed insulin-only, non-insulin glucose-lowering drugs (GLDs) or combined insulin and non-insulin GLDs compared to no GLD. International Classification of Diseases-10 codes were used to identify dementia status and prior hypoglycemia; frailty was quantified using the Hospital Frailty Risk Score.
Results: Insulin-only, non-insulin GLDs, combined insulin and non-insulin GLDs, and no GLDs were prescribed to 19%, 39%, 20%, and 23% of patients, respectively. Patients >80 years were less likely than patients aged 65-80 to be prescribed any of the GLD therapies, (eg. non-insulin GLDs [OR 0.67; 95%CI 0.55-0.82]), compared to no GLD. Similarly, high vs. low frailty scores were associated with not being prescribed any of the three GLD therapies, (eg. non-insulin GLDs [OR 0.63; 95%CI 0.45-0.87]). However, dementia was not associated with discharge prescribing of GLD therapies. Patients with a hypoglycemia-related admission were more likely than those not hospitalized with hypoglycemia to receive insulin-only (OR 4.28; 95%CI 2.89-6.31).
Clinicians consider age and frailty when tailoring diabetes treatment regimens for patients discharged from hospital with T2D and related complications. There is scope to optimize prescribing for patients with dementia and for those admitted with hypoglycemia.

Key words: Type 2 diabetes, frailty, dementia.



The benefits of intensive glycemic control for preventing microvascular outcomes in middle and older age people with Type 2 diabetes (T2D) have been demonstrated in the ACCORD, ADVANCE, and VADT trials (1-3). However, intensive treatment is associated with an increased risk of hypoglycemia and does not improve survival or the incidence of macrovascular outcomes in people with limited life expectancies (1-3). The risks of intensive treatment may outweigh the benefits in frail older people (4). The guidelines of the American Diabetes Association (ADA) recommend less stringent glycemic targets of <8% and <8.5% (64 mmol/mol and 69 mmol/mol) for older individuals with complex and very complex health status (5). Similarly, Australian guidelines advise less intensive and individualized treatment for these patient groups (6). Nevertheless, UK data suggest that those who are frail and have dementia are treated with similar glucose-lowering drugs (GLDs) and with the aim to achieve similar glycemic targets as robust older people without dementia (7).
Frailty is an important complication of diabetes (8, 9), and is characterized by vulnerability to stressors and a reduced ability to maintain homeostasis (10). Frailty increases the risk of adverse drug events, including falls, disability and death (11). There are reciprocal relationships between hypoglycemia, dementia, and frailty (12). There have been calls for frailty status to guide treatment selection (13), with frail people with diabetes at 71% higher adjusted risk of all-cause hospitalization and twice the risk of mortality than non-frail people (14). Furthermore, older people with diabetes who develop dementia have three times the risk of hypoglycemia compared to those who do not develop dementia (15). The ACCORD-MIND study reported that cognitive decline over 20 months was associated with a higher risk of hypoglycemia regardless of treatment intensity (15).
Hospitalization represents an opportunity for clinicians to adjust T2D treatment regimens, although it is unclear to what extent hospital clinicians consider age, frailty and dementia in prescribing decisions. There is also a paucity of information about GLDs prescribed for older people who are frail and/or live with dementia, who may have different goals of care and treatment benefits and risks (5, 6). The primary aim of this study was to determine whether age, frailty, or dementia predict discharge treatment types for patients with T2D and related complications. Our secondary aim was to determine the association between prior hypoglycemia and discharge treatment types.



Data source, study design, and study population

The study was conducted at Eastern Health, a large metropolitan public hospital network in Melbourne, with three acute and four subacute hospitals (1,423 beds) (16). Eastern Health services a catchment area of 750,000 people and recorded 1,175,249 patient episodes between July 2015 and June 2016 (16). Eastern Health implemented an Electronic Medical Record (EMR) with electronic prescribing (e-prescribing) in 2011 (17). EMR discharge prescriptions record all medications intended for use by a patient after being discharged from the hospital (17). Demographic information and discharge diagnoses were extracted by the health service’s Decision Support Unit, which relies upon the standard practice of Clinical Coders within the Health information Unit (17). Diagnoses were recorded using International Classification of Diseases-10 (ICD-10) codes with up to 40 diagnoses per patient. Discharge medications were identified from the EMR using Anatomical Therapeutic Chemical (ATC) classification codes (18).
We conducted a cohort study of 3,067 adults aged between 65 and 99 years with T2D who were discharged from one of the Eastern Health hospital locations in Melbourne, Australia, between 2012 and 2016 with a principal diagnosis of T2D with a diabetes related complication.

Measures and definitions

Our study population included all patients with a principal diagnosis of T2D, identified using ICD-10 code E11, and an ICD-10 code (E11-E14) for a diabetes-related complication recorded at hospital discharge (index hospitalization) (18). Medications for T2D were broadly classified as insulins (ATC code A10A) or non-insulin GLDs (A10B). ATC codes used to identify GLDs classes are provided in Appendix A. A modified version of the Diabetes Complications Severity Index (DCSI) [19], which converts ICD-10 codes into a 13-level metric to quantify effects of diabetes on seven body systems, was used as an indicator of T2D severity. Although this version of the DCSI does not require laboratory data, validation studies have shown that its capacity to predict diabetes severity is comparable to other versions which do (19, 20). The DCSI is also likely to be indicative of diabetes duration as it has been shown that for every additional year of diabetes duration in people over 60 years, the adjusted odds of microvascular disease increases by 6% (p<0.001) (21).
We utilized a validated Hospital Frailty Risk Score, which categorizes people into three frailty categories based on the sum of weighted scores identified from International Classification of Diseases (ICD-10) codes (22). Gilbert et. al (2018) derived this score using 109 ICD-10 codes at least twice as prevalent in frail versus non-frail patients weighted according to how strongly they predict frailty (22). Codes used to derive the Hospital Frailty Risk Score (HFRS) reflect conditions linked to frailty (for example, volume depletion, cognitive impairment, and falls) or conditions overrepresented in frail populations such as lung disease, heart conditions and elective cataracts. Cut-point scores of <5, 5-15, and >15, as published by Gilbert et. al. indicated low, moderate, and high degrees of frailty, respectively. ICD-10 codes used to identify dementia and hypoglycemia are given in Appendix B.

Statistical analysis

Baseline characteristics were presented as means with standard deviations (SDs), medians with interquartile ranges (IQRs) or as frequencies and percentages. Predictors of treatment initiation were estimated using multinomial logistic regression. Variables were included in the final model if the unadjusted p-value associated with the odds ratio (OR) was <0.25. We included age (65-80 and >80), frailty (low, moderate or high) and dementia in our regression model and estimated adjusted odds ratios (ORs) with 95% confidence intervals (CI), adjusted for sex, index year, DCSI score, congestive cardiac failure (CCF), myocardial infarction (MI), renal disease, transient ischemic attack (TIA) or stroke, and hypoglycemia, (ICD-10 codes for comorbidities given in Appendix B). Variance Inflation Factors (VIF) with a cut-off of 2 were used to assess collinearity between the variables in the model. Statistical differences were evaluated using Pearson’s chi-squared test and ANOVA for categorical and continuous variables, respectively. We excluded the Charlson Comorbidity Index (CCI) from our adjusted model because it was collinear with several comorbidities in our model, though it is included in Table 1 for completeness. Comorbidities and concomitant medications were not included in the same model because concomitant medications were conceptualized as intermediate variables in the causal pathway between the comorbidity and the diabetes treatment regimen.
All analyses were conducted using the statistical software package SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). This study was approved by the Eastern Health and Monash University Human Research Ethics Committees (study number LR41/2017).



Cohort Characteristics

Of the 3,067 people hospitalized with T2D, 19% were prescribed insulin-only, 39% non-insulin GLDs, 20% insulin and non-insulin combinations and 23% no GLDs (Table 1). Slightly less than half of the cohort were female (48%), and the mean age of the cohort was 78.6 years (SD 7.8). Patients not prescribed GLDs were older (81.0, SD 8.1) than those prescribed non-insulin GLDs (78.3, SD 7.7), insulin only (78.2, SD 7.3), or combination therapy (76.5, SD 7.3). Based on ICD-10 codes, 9% of the cohort had a dementia diagnosis, and 11% had been hospitalized with hypoglycemia.
Frailty scores were non-normally distributed, therefore medians and interquartile ranges (IQR) were reported. The median frailty score for the study population was 5.8 (IQR 2.5-10.2), with median frailty scores being higher amongst those who were not prescribed GLDs (6.9, IQR 3.0-11.5) and lower amongst those prescribed combinations (5.3, IQR 2.3-9.3), (Table 1). The Pearson correlation coefficient between age and frailty scores was 0.23 (p<0.0001). Figure 1a) shows that 69.7% of patients prescribed insulin-only therapy had a DCSI score >1, p<0.0001. Figure 1b) indicates that 21.6% and 16.0% of the insulin-only and combination therapy groups had a documented prior hypoglycemia during their index hospitalization. In contrast, 5.7% and 6.7% of individuals receiving no GLD and non-insulin hypoglycemic agents had a documented episode of hypoglycemia, p<0.0001. Patients in the combination group were least likely (9.5%), to have a HFRS >15, p<0.0001 and to have dementia (4.3%), p=0.0002, (Figure 1c). Those with HFRS >15 were most likely (12.5%) to have had an episode of hypoglycemia, but this was not significantly higher than the other groups, p=0.16 (Figure 1d).

Figure 1a. Proportion of patients in each treatment group with baseline DCSI scores ≤1 or >1

DCSI Diabetes Complications Severity Index; GLD Glucose Lowering Drug; p<0.0001 (Pearson’s chi-squared test)

Figure 1b. Proportions of patients in each treatment group with a diagnosis of hypoglycemia recorded during index hospitalization

GLD Glucose Lowering Drug; p<0.0001 (Pearson’s chi-squared test)

Figure 1c. Proportions of patients in each treatment group within each of the three frailty categories or with a diagnosis of dementia at baseline

HFRS Hospital Frailty Risk Score; GLD Glucose Lowering Drug; p<0.0001 for HFRS categories, p=0.0002 for dementia (Pearson’s chi-squared test)

Figure 1d. Proportion of patients in frailty categories with hypoglycemia diagnosis at index discharge

HFRS Hospital Frailty Risk Score; p=0.16 (Pearson’s chi-squared test)


Predictors of Prescribed Anti-Hyperglycemic Therapy

People aged >80 versus those aged 65-80 were less likely to be prescribed insulin only (OR 0.54 95%CI 0.42-0.69), non-insulin GLDs only (OR 0.67 95%CI 0.55-0.82) or combinations of the two (OR 0.37 95%CI 0.29-0.47), compared to no GLDs (Table 2, Figure 2a). People with high frailty scores, compared to low scores, were less likely to be prescribed insulin only (OR 0.62 95%CI 0.42-0.91), non-insulin GLDs (OR 0.63 95%CI 0.45-0.87), or combinations of the two (OR 0.65 95%CI 0.43-0.96), compared to no GLDs (Table 2, Figure 2b).
People with dementia were less likely to be prescribed non-insulin GLDs (OR 0.73 95%CI 0.53-1.01) or insulin and non-insulin GLD combinations (OR 0.72 95%CI 0.47-1.10) compared to no GLDs, although these results were non-statistically significant (Table 2). People hospitalized with hypoglycemia, were more likely to receive insulin only (OR 4.28 95%CI 2.89-6.31) or combinations of insulin and non-insulin GLDs, (OR 3.15 95%CI 2.11-4.69), compared to no GLDs.

Table 1. Demographic and clinical characteristics of people over 65 years with Type 2 diabetes and a primary diagnosis of a diabetes related complication, by prescribed discharge medication

SD Standard deviation; IQR Inter Quartile Range; DCSI Diabetes Complications Severity Index; CCI Charlson Comorbidity Index. Data are presented as n(%). Non-insulin therapy included: metformin, sulfonylureas, acarbose, thiazolidinediones, dipeptidyl peptidase-4 inhibitors (DPP-4Is), glucagon-like peptide-1 agonists (GLP-1As), sodium-glucose cotransporter-2 inhibitors (SGLT-2Is), and fixed-dose combinations (FDC); *P-values were calculated using the Pearson’s chi-squared test and ANOVA for categorical and continuous variables, respectively.

Table 2. Odds ratios for being prescribed Glucose Lowering Drugs (GLDs) versus No GLD at discharge amongst people with Type 2 diabetes and a primary diagnosis of a diabetes-related complication

DCSI: Diabetes Complications Severity Index; bold indicates a statistically significant result.


Types of T2D Therapy Prescribed

The most commonly prescribed insulin types within the group receiving insulin-only therapy were mixed (64.7%), fast-acting (30.0%) and long-acting (29.8%), with most individuals being prescribed either 1 (71.6%) or 2 (28.1%) different insulin products (Appendix C). Within the group receiving combination therapy, mixed (51.4%), long-acting (41.4%), and fast-acting (17.1%) insulins were most likely to be prescribed. All individuals in this group were prescribed either one (83.5%) or two (16.5%) types of insulin.
People in the non-insulin GLD group were most likely to be prescribed either metformin (69.9%) or a sulfonylurea (57.8%), with the majority being issued with either 1 (59.6%) or 2 (34.3%) non-insulin GLDs (Appendix C). Metformin (74.0%) and sulfonylureas (47.1%) were also the most commonly prescribed non-insulin GLDs in the combination group, and people in this group were most likely to receive either 1 (69.7%) or 2 (28.6%) non-insulin GLDs.

Figure 2a. Forest plot of type of antihyperglycemic therapy prescribed for people over 65 years, hospitalised with Type 2 diabetes and a related complication, by age group

OR Odds Ratio; CI Confidence Interval; GLD Glucose Lowering Drug.

Figure 2b. Forest plot of type of antihyperglycemic therapy prescribed for people over 65 years, hospitalised with Type 2 diabetes and a related complication, by frailty score

OR Odds Ratio; CI Confidence Interval; GLD Glucose Lowering Drug; HFRS Hospital Frailty Risk Score.



This was the first study to investigate how age, frailty, and dementia predict hospital discharge prescribing for people with T2D. Older age and frailty predicted less intense treatment of T2D, people 80 and older were 63% less likely than those aged between 65-80 years to receive combinations of insulin and non-insulin GLDs, compared to no GLDs. Moreover, frail people were 35% less likely than robust people to be discharged on a combination of insulin and non-insulin GLDs versus no GLDs.
Our findings suggest clinicians consider age and frailty by tailoring diabetes treatment regimens. This is encouraging because frail older individuals are more vulnerable to adverse events, such as hypoglycemia and mortality (23). In addition, weight loss and sarcopenia associated with frailty (12) may be exacerbated by changes in the natural history of T2D, which shifts from a progressive to a regressive course in individuals who are frail (24). Older age is a well-known risk factor for hypoglycemia, and our findings demonstrate adherence to national and international prescribing guidelines, which advise that individuals with shorter life expectancy derive limited benefits from stringent glycemic targets (5, 6). Older people with T2D are also less likely to recognize early signs of hypoglycemia due to reduced awareness of hypoglycemic symptoms and slower reaction times than younger counterparts (25). Severe hypoglycemia can cause sudden cardiovascular death, and episodes of mild hypoglycemia can cause falls, fractures, cognitive impairment, seizures, coma, cardiovascular events, and arrhythmias (23). National estimates in the US indicate that insulin users >80 years are hospitalized for hypoglycemia or insulin-related errors at five times the rate of insulin users aged 45-64 years (26). Reasons postulated for this increase include reduced food intake and administration of the wrong insulin product (26).
People with dementia tended to be less likely to be discharged on insulin and non-insulin GLD combinations compared to no GLDs. Although not statistically significant, this result suggests possible increasing awareness of the need to align treatment with goals of care (27). It may also reflect prescribers’ awareness that individuals with dementia have a reduced capacity to manage complex regimens, particularly those involving insulin, due to difficulties in remembering dosage directions, to take doses on time or to take with food. Insulin and oral hypoglycemic agents such as sulfonylureas are considered high-risk medications and are associated with preventable hospitalizations, including among residents of nursing homes and long-term care facilities.
People hospitalized with hypoglycemia were over three times as likely to be prescribed insulin and non-insulin GLD combinations and over four times as likely to be prescribed insulin only compared to no GLDs. While we were not able to assess the clinical appropriateness of T2D regimens for individual patients, this suggests a possible opportunity for treatment de-intensification in ‘at risk’ population groups. It is also possible that there is scope for regimen simplification, as 28.5% of individuals prescribed insulin only and 16.5% prescribed combination treatment used at least two insulin products. It has been shown that simplification of multiple insulin regimens to basal insulin glargine only, reduced duration of hypoglycemia by 65% after eight months (28).

Strengths and limitations

Our study analyzed five years of discharge prescribing data from a large public hospital network in Melbourne. To our knowledge, this is the first study to investigate the impact of age, frailty, and dementia status as predictors of T2D discharge treatment intensity. One limitation of this study is that the Hospital Frailty Risk Score was validated for individuals >75 years, whereas we included individuals ≥65 years. The Hospital Frailty Risk Score was calculated using ICD-10 codes including dementia and, therefore, it is possible that there was overlap between dementia and frailty. Prescribing patterns may have evolved since 2016, particularly with the introduction of sodium-glucose cotransporter-2 inhibitors (SGLT-2Is). We considered age, dementia, and frailty status as categorical rather than continuous variables. However, age, frailty and dementia severity are continuous and there is no evidence for specific cut-points to define prescribing appropriateness in relation to these parameters. Lack of data on diabetes duration is a limitation. However, we presented the diabetes treatment according to less and more severe diabetes complications, which are related to diabetes duration (21). We did not have data on pre-admission treatment. However, we have presented the proportion of patients with documented prior hypoglycemia in each of the treatment groups. We hypothesized that prior hypoglycemia would prompt clinicians to modify treatment. Analyzing discharge prescribing is consistent with the treatment decision design in which cohorts are anchored at the point when treatment decisions are made (29). This is because medication regimens are typically evaluated during a hospital episode (29). Additionally, given that the sample comprised Australians who had been hospitalized, the results are not necessarily generalizable to all older patients with T2D across all clinical settings. We were not able to analyze data on HbA1C levels and ethnicity. Finally, a common limitation with the use of prescribing data, is that we do not know whether prescribed medications are actually taken by patients as directed.



Frail older people hospitalized with T2D and diabetes-related complications are less likely to be prescribed insulin-only GLDs, non-insulin GLDs or a combination of both, compared to no GLDs. Increasing age is also associated with receiving less intensive GLD regimens. Conversely, people hospitalized with hypoglycemia are considerably more likely to be discharged with a medication regimen which includes insulin. Clinicians appear to consider age and frailty when prescribing for people with T2D, but there is further opportunity for treatment de-intensification in ‘at risk’ groups.


Acknowledgements: S.W. is supported through an Australian Government Research Training Program Scholarship. J.S.B. is supported by a NHMRC Boosting Dementia Research Leadership Fellowship [grant number #1140298].

Conflicts of Interest: J.S.B. has received grant income paid to his employer from NHMRC, Australian Government Department of Health, Victorian Government Department of Health and Human Services, Dementia Australia Research Foundation, Yulgilbar Foundation, GSK Independent Medical Education and several aged care provider organizations. M.C reports personal fees from Nestlé, outside the submitted work. J.I. reports grants from AstraZeneca, Amgen, Dementia Australia Research Foundation, National Health and Medical Research Council, and National Breast Cancer Foundation, outside the submitted work. S.W., D.J.M., L.F., and C.K. have no competing interests to declare.

Ethical standards: This study was approved by the Eastern Health and Monash University Human Research Ethics Committees (study number LR41/2017).



1. The Action to Control Cardiovascular Risk in Diabetes Study Group, Effects of Intensive Glucose Lowering in Type 2 Diabetes. The New England Journal of Medicine, 2008. 358(24): p. 2545-2559.
2. Heller, S.R., A Summary of the ADVANCE Trial. Diabetes Care, 2009. 32(supplement 2): p. S357.
3. Duckworth, W., et al., Glucose Control and Vascular Complications in Veterans with Type 2 Diabetes. New England Journal of Medicine, 2009. 360(2): p. 129-139.
4. Abdelhafiz, A.H. and Sinclair, A.J., Deintensification of Hypoglycaemic Medications-use of a Systematic Review Approach to Highlight Safety Concerns in Older People with Type 2 Diabetes. Journal of Diabetes Complications, 2018. 32(4): p. 444-450.
5. American Diabetes Association, Older Adults: Standards of Medical Care in Diabetes—2018. Diabetes Care, 2018. 41(Supplement 1): p. S119.
6. RACGP. General Practice Management of Type 2 Diabetes 2016-2018. 2016.
7. Hambling, C.E., et al., Older People with Type 2 Diabetes, Including those with Chronic Kidney Disease or Dementia, are Commonly Overtreated with Sulfonylurea or Insulin Therapies. Diabetic Medicine, 2017. 34(9): p. 1219-1227.
8. Strain, W.D., et al., Type 2 Diabetes Mellitus in Older People: A Brief Statement of Key Principles of Modern Day Management Including the Assessment of Frailty. A National Collaborative Stakeholder Initiative. Diabetic Medicine, 2018. 35(7): p. 838-845.
9. Maddaloni, E., L. D’Onofrio, and P. Pozzilli, Frailty and Geography: Should These Two Factors Be Added to the ABCDE Contemporary Guide to Diabetes Therapy? Diabetes/Metabolism Research and Reviews, 2016. 32(2): p. 169-175.
10. Sinclair, A.J., Dunning, T., and Rodriguez-Mañas, L., Diabetes in Older People: New Insights and Remaining Challenges. The Lancet Diabetes & Endocrinology, 2015. 3(4): p. 275-285.
11. Liau, S.J., et al., Medication Management in Frail Older People: Consensus Principles for Clinical Practice, Research, and Education. Journal of the American Medical Directors Association, 2020. 22(1): p43-49.
12. Abdelhafiz, A.H., McNicholas, E., and Sinclair, A.J., Hypoglycemia, Frailty and Dementia in Older People with Diabetes: Reciprocal Relations and Clinical Implications. Journal of Diabetes Complications, 2016. 30(8): p. 1548-1554.
13. Orlandini, L. and Cesari, M., Should We Care More About Frailty When We Treat Diabetes? The Journal of Frailty & Aging, 2020. 9(2): p. 66-67.
14. Ferri-Guerra, J., et al., The Association of Frailty with Hospitalizations and Mortality among Community Dwelling Older Adults with Diabetes. The Journal of Frailty & Aging, 2020. 9(2): p. 94-100.
15. Punthakee, Z., et al., Poor Cognitive Function and Risk of Severe Hypoglycemia in Type 2 Diabetes. Diabetes Care, 2012. 35(4): p. 787.
16. Fanning, L., et al., Validity of Electronic Hospital Discharge Prescription Records as a Source of Medication Data for Pharmacoepidemiological Research. Therapeutic Advances in Drug Safety, 2018. 9(8): p. 425-438.
17. Fanning, L., et al., The Representativeness of Direct Oral Anticoagulant Clinical Trials to Hospitalized Patients with Atrial Fibrillation. European Journal of Clinical Pharmacology, 2017. 73(11):1427-1436.
18. Dugan, J. and Shubrook, J., International Classification of Diseases, 10th Revision, Coding for Diabetes. Clinical Diabetes. 2017. 35(4): p. 232-238.
19. Chang, H.-Y., et al., Validating the adapted Diabetes Complications Severity Index in claims data. The American Journal of Managed Care, 2012. 18(11): p. 721-726.
20. Glasheen, W.P., Renda, A., and Y. Dong, Diabetes Complications Severity Index (DCSI)-Update and ICD-10 translation. Journal of Diabetes Complications, 2017. 31(6): p. 1007-1013.
21. Nanayakkara, N., et al., Age, Age at Diagnosis and Diabetes Duration are all Associated with Vascular Complications in Type 2 Diabetes. Journal of Diabetes and its Complications, 2018. 32(3): p. 279-290.22.
22. Gilbert, T., et al., Development and Validation of a Hospital Frailty Risk Score Focusing on Older People in Acute Care Settngs using Electronic Hospital Records: an Observational Study. The Lancet, 2018. 391(10132): p. 1775-1782.
23. Sinclair, A.J., et al., Evaluating Glucose-Lowering Treatment in Older People with Diabetes: Lessons from the IMPERIUM Trial. Diabetes, Obesity & Metabolism, 2020.24.
24. Abdelhafiz, A.H., Koay, L., and Sinclair, A.J. The Effect of Frailty should be Considered in the Management Plan of Older People with Type 2 diabetes. Future Science OA, 2016. 2(1): p. FSO102-FSO102.25.
25. McCall, A.L., Insulin Therapy and Hypoglycemia. Endocrinology and Metabolism Clinics of North America, 2012. 41(1): p. 57-87.
26. Geller, A.I., et al., National Estimates of Insulin-Related Hypoglycemia and Errors Leading to Emergency Department Visits and Hospitalizations. JAMA Internal Medicine, 2014. 174(5): p. 678-86.
27. Reeve, E., J.S. Bell, and Hilmer, S.N., Barriers to Optimising Prescribing and Deprescribing in Older Adults with Dementia: A Narrative Review. Current Clinical Pharmacology, 2015. 10(3): p. 168-177.
28. Munshi, M.N., et al., Simplification of Insulin Regimen in Older Adults and Risk of Hypoglycemia. JAMA Internal Medicine, 2016. 176(7): p. 1023-1025.
29. Brookhart, M.A., Counterpoint: The Treatment Decision Design. American Journal of Epidemiology, 2015. 182(10): p. 840-845.



N.K. LeBrasseur1, R. de Cabo2, R. Fielding3, L. Ferrucci4, L. Rodriguez-Manas5, J. Viña6, B. Vellas7


1. Robert and Arlene Kogod Center on Aging and Department of Physical Medicine and Rehabilitation, Mayo Clinic, Rochester, MN, USA; 2. Translational Gerontology Branch, National Institute on Aging, Baltimore, MD, USA; 3. Tufts University, Boston, MA, USA; 4. Intramural Research Program, National Institute on Aging, Baltimore, MD, USA; 5. Servicio de Geriatría, Hospital Universitario de Getafe, Toledo, Spain; 6. Freshage Research Group, Department of Physiology, Faculty of Medicine, University of Valencia, CIBERFES-ISCIII, INCLIVA, 46010 Valencia, Spain; 7. Gerontopole, INSERM U1027, Alzheimer’s Disease Research and Clinical Center, Toulouse University Hospital, Toulouse, France
Corresponding author: Nathan K. LeBrasseur, Robert and Arlene Kogod Center on Aging and Department of Physical Medicine and Rehabilitation, Mayo Clinic, Rochester, MN, USA, lebrasseur.nathan@mayo.edu

Task Force members: Samuel Agus (Paris); Sandrine Andrieu (Toulouse, France); Mylène Aubertin-Leheudre (Montréal, Canada); Amos Baruch (South San Francisco, USA); Shalender Bhasin (Boston, USA); Louis Casteilla (Toulouse, France); Peggy Cawthon (San Francisco, USA); Matteo Cesari (Milan, Italy); Manu Chakravarthy (Cambridge, USA); Alfonso J. Cruz Jentoft (Madrid, Spain); Carla Delannoy (Vevey, Switzerland); Philipe De Souto Barreto (Toulouse, France) ; Waly Dioh (Paris, France); Françoise Forette (Paris, USA); Sophie Guyonnet (Toulouse); Joshua Hare (Miami) ; Darren Hwee (South San Francisco); Kala Kaspar (Vevey); Valérie Legrand (Nanterre, France); Roland Liblau (Toulouse, France); Yvette Luiking (Utrecht, The Netherland) ; Bradley Morgan (South San Francisco, USA) ; Eric Morgen (Richmond, USA); John Morley (St Louis, USA) ; Angelo Parini (Toulouse, USA); Suzette Pereira (Columbus, USA); Alfredo Ramirez (Cologne, USA); Jean-Yves Reginster (Liege, Belgium); Yves Rolland (Toulouse, France); Ricardo Rueda (Columbus, USA); Jorge Ruiz (Miami, USA); Peter Schüler (Langen, Germany); Alan Sinclair (London, United Kingdom); Nicolas Thevenet (Nanterre, France); Janneke Van Wijngaarden (Utrecht, The Netherlands); Jeremy Walston (Baltimore, USA); Debra L. Waters (Dunedin, New Zealand)

J Frailty Aging 2021;10(3)196-201
Published online February 23, 2021, http://dx.doi.org/10.14283/jfa.2021.5



The International Conference on Frailty and Sarcopenia Research Task Force met in March 2020, in the shadow of the COVID-19 pandemic, to discuss strategies for advancing the interdisciplinary field of geroscience. Geroscience explores biological mechanisms of aging as targets for intervention that may delay the physiological consequences of aging, maintain function, and prevent frailty and disability. Priorities for clinical practice and research include identifying and validating a range of biomarkers of the hallmarks of aging. Potential biomarkers discussed included markers of mitochondrial dysfunction, proteostasis, stem cell dysfunction, nutrient sensing, genomic instability, telomere dysfunction, cellular senescence, and epigenetic changes. The FRAILOMICS initiative is exploring many of these through various omics studies. Translating this knowledge into new therapies is being addressed by the U.S. National Institute on Aging Translational Gerontology Branch. Research gaps identified by the Task Force include the need for improved cellular and animal models as well as more reliable and sensitive measures.

Key words: Aging, frailty, resilience, hallmarks of aging, translational research.



The International Conference on Frailty and Sarcopenia Research (ICFSR) Task Force met in Toulouse, France on March 10, 2020 to discuss geroscience. The timing could not have been more prescient: On the following day, March 11, the World Health Organization (WHO) declared the COVID-19 outbreak a pandemic (1). The disease, caused by a virus known as SARS-CoV2, had at that point already claimed the lives of more than 4,000 people in 114 countries, with the risk of morbidity and mortality especially elevated in older people and those with underlying medical conditions (2).
Why older adults are particularly vulnerable to this novel virus remains poorly understood. Age-related physiological changes including immune senescence, the high incidence of chronic illnesses, and frailty may decrease resilience and increase susceptibility to cardiovascular, pulmonary, and infectious diseases (3). Older adults may also present atypically, further complicating diagnosis and treatment (4).
Targeting the biology of aging to prevent and treat aging-associated diseases and geriatric syndromes as a group represents a fundamentally different approach to extending human health. Historically, drug development efforts have centered on evidence-based risk factors identified through epidemiological studies and specific molecular alterations that contribute to singular diseases. This approach has had marked success but has also revealed that interventions for a specific disease, whether it be heart disease, cancer, Alzheimer’s Disease, pneumonia, or COVID-19, have a limited impact on the emergence of the multitude of other age-related conditions. The promise and potential payoff of interventions for aging and compressing morbidity is high; however, a more in-depth understanding of the underlying biology is needed. In response, the interdisciplinary field of geroscience has emerged to explore biological mechanisms of aging and determine how these mechanisms lead to the vast collection of age-related chronic diseases and geriatric syndromes, including sarcopenia and frailty (5–8). Geroscience approaches are clearly and urgently needed as well to better understand the susceptibility of older adults to acute challenges, such as COVID-19, and develop novel treatments for the most vulnerable individuals, including frail older adults (3).
Specialties represented in the ICFSR Task Force play a critical role in advancing geroscience because they are expert in 1) the discovery and quantification of the biological mechanisms of aging; 2) the study of aging and aging-related diseases in preclinical models, and; 3) the design and execution of clinical trials of testing interventions (exercise, dietary modifications, drugs, and combinations thereof) to optimize the health and function or resilience of multiple physiological systems. Indeed, a multidisciplinary approach is necessary to expedite the translation of newly discovered therapies to clinical application.


What is healthy aging?

The WHO introduced the concept that healthy aging and disease prevention hinges on preventing declines in intrinsic capacity – a composite of physical and mental capacities that peak in early adulthood and tend to decline in later years. WHO developed a model for integrated care of older people (ICOPE) that focuses on maintaining intrinsic capacity through the adoption of a healthy lifestyle (9).
As individuals age, they transition between robustness to frailty, defined as increased vulnerability to endogenous and exogenous stressors and a decline in physiological reserve and function across multiple organ systems. Thus, frailty and intrinsic capacity represent distinct but related concepts, both with similar physiological underpinnings (10). Physiological mechanisms of resilience and reserve further impact the capacity of an individual to overcome adverse events (11). Geroscience resides at the intersection of these concepts (12).


Biological hallmarks of aging

Geroscience assumes that aging itself, defined as the accumulation of diverse forms of molecular and cellular damage and repair, ultimately drives the increased risk of chronic diseases among older people (13). López-Otin and colleagues proposed nine distinct yet interrelated forms, or “hallmarks”, of aging: genomic instability, telomere attrition, epigenetic alterations, loss of proteostasis, deregulated nutrient sensing, mitochondrial dysfunction, cellular senescence, stem cell exhaustion, and altered intercellular communication (14). These hallmarks have been incorporated into an emerging view of geroscience that resulted in the creation of a trans-National Institutes of Health (NIH) Geroscience Interest Group, GSIG (15).
Together, the biological mechanisms progressively result in loss of cellular homeostasis, dysregulation across multiple physiological systems and, ultimately, disease, disability, and death. Both the accumulation and the repair of aging hallmarks are strongly influenced by behavior, the environment, and genetics, resulting in substantial variation among individuals. This has led to the important concept that biological age differs from chronological age. Biological age, in contrast to chronological age, is difficult to quantify, which has led to inconsistent definitions in the literature. Herein, approaches to measure aging hallmarks are presented as a means to better define biological age of cells, tissues, and, ultimately, organisms.
One of the longest longitudinal studies of normative human aging, the Baltimore Longitudinal Study of Aging (BLSA), has been running for more than 60 years. The BLSA collected multidomain clinical and functional data from participants with increasing frequency as they aged: every 4 years for those under age 60 increasing to every year for those over age 80. Using these data, the investigative team recently proposed a roadmap to build a phenotypic metric of aging, which by systematically characterizing the continuum of aging in an individual, could advance understanding of the kinetics of aging, as well as discovery and development of effective interventions. The framework encompasses four domains: body composition, energetics, homeostatic mechanisms, and neurodegeneration/neuroplasticity (16).
Scientists in the NIA Intramural Research Program are also conducting a Study of Longitudinal Aging in Mice (SLAM) to better understand whether discoveries in mice can be translated to humans and which aging phenotypes share or do not share common traits in order to fine tune interventions to translatable outcomes. They are conducting the study in two strains of mice of both sexes, selecting the C57BL6/J and the UM-HET3 mice to better recapitulate both the genetic homogeneity of most aging studies and the heterogeneity found in humans respectively. SLAM investigators are conducting a broad range of clinically relevant assessments over time and across multiple domains. To assess frailty, they will apply two newly developed tools: the mouse clinical frailty index and mouse frailty phenotype assessment (17). For example, at 3-month intervals, they assess gait speed, fasting blood glucose, energetic cost on a metabolic treadmill, and frailty. They also perform imaging tests such as magnetic resonance imaging (MRI), nuclear magnetic resonance (NMR), and micro-computed tomography (micro-CT) to obtain organ images and spectra, body composition, and bone mass changes with age.


Implementing biomarkers of aging in clinical research

Metrics of aging span multiple domains and include biological hallmarks, organ impairments (e.g., muscle weakness), functional limitations (e.g., slow gait speed), and disease and deficit accumulation (e.g., frailty). Undoubtedly, as people age, the onset and progression of decline within each domain differs; and understanding whether biomarkers of these different domain are mechanistically connected and exploring temporal relationship between domains is critical to translate this science into effective interventions. Based on the geroscience premise that aging itself is the driver of the majority of chronic diseases and geriatric syndromes, quantifiable indicators or “biomarkers” of biological age would be of significant utility to target individuals who are experiencing accelerated aging as well tracking the effectiveness of interventions aimed at slowing down the aging process.
For clinical research, biomarkers of aging could enable identification of persons appropriate for and, potentially, most responsive to interventions targeting the biology of aging. In such trials, biomarkers may be used to verify target engagement and also serve as informative surrogate endpoints that may change well before clinical outcomes (e.g., frailty measures (18)). In clinical practice, biomarkers may help providers discern between chronological and biological age and, in turn, serve as determinants of risk and guide clinical decision making (e.g., medical versus surgical management of a condition). Implementing laboratory biomarkers of aging in clinical research and practice, however, will first require demonstrating that they can be reliably measured in blood or other accessible tissues and reflect clinical manifestations of aging. Recently, a candidate panel of senescence biomarkers was developed based on the secretome of senescent human endothelial cells, fibroblasts, preadipocytes, epithelial cells, and myoblasts in vitro. In older adults undergoing surgery, the senescence biomarkers were shown to correlate with chronological age and biological age, as defined by the frailty index, and to predict adverse events such as surgical complications and rehospitalizations (19).
Table 1 lists some potential biomarkers of the hallmarks of aging as well as the biological matrix in which they could be assessed.

Table 1
Potential Biomarkers of the Hallmarks of Aging (courtesy of Tamara Tchkonia and Nathan LeBrasseur)


Omics-based laboratory biomarkers have also been investigated in the FRAILOMIC initiative (20,21), an international consortium funded by the European Commission that aims to develop omics-based clinical instruments to assess frailty and predict the risk of frailty and subsequent disability. The consortium is analyzing data from four European cohorts: InCHIATIi (Tuscany, Italy), AMI (Gironde, France), the Three-City (3-C)Study (Bordeaux, France) (3C), and Toledo Study for Healthy Aging (TSHA, Toledo, Spain). The wide range of potential biomarkers investigated in the exploratory phase of this initiative is shown in Table 2.

Table 2
Exploratory phase biomarkers (courtesy of Prof. L. Rodriguez Mañas)


FRAILomic studies thus far have shown that biomarkers of frailty change according to clinical characteristics of participants, suggesting the existence of different clinical phenotypes of frailty. For example, omics biomarkers may be associated with disability, sarcopenia or other organ-specific diseases, and vary by sex, ethnicity, and race. Lab biomarkers appear to be modestly associated with classical biomarkers such as biomarkers of inflammation, hormonal changes, and glucose dysregulation (23), particularly among individuals with disability.
For example, in one study of FRAILomic participants, serum levels of the soluble receptor for advanced glycation end-products (sRAGE) was shown to be an independent predictor of mortality in frail individuals, suggesting that sRAGE levels may be useful for prognostic assessment and treatment stratification (24). Another study demonstrated that frail participants had higher plasma levels of 3-methylhistidine (3-MH) and higher ratios of 3-MH to creatine and 3-MH to estimated glomerular filtration rate, suggesting that these markers may be useful in identifying individuals at risk of frailty Finally, in this same regard, fat-soluble vitamins and carotenoids are biomarkers of frailty status (robust, pre-frail, frail) (26) but do not predict the risk of becoming frail (27)
Future studies of biomarkers of aging and their cross-sectional and longitudinal relationships with parameters of function (e.g., physical, cognitive, cardiovascular, pulmonary, renal metabolic, immune, and sensory) and resilience (e.g., to infection and consequences of SARS-CoV2) affected by advancing age are warranted. Longitudinal studies promise to provide greater insights into rates of biological aging and, potentially, in the context of clinical trials, the extent to which the molecular and cellular effects of aging can be attenuated and/or reversed. As novel interventions targeting the biology of aging emerge, biomarkers of aging will facilitate their testing and development.


Translation – developing agents that target fundamental aging processes

The Translational Gerontology Branch (TGB) at the National Institute on Aging (NIA) is part of the NIA’s intramural research program. Research conducted in TGB labs ranges from drug discovery using a variety of in vivo and in vitro models to clinical and longitudinal studies. For example, TGB researchers are studying the cellular and molecular mechanisms underlying aging, diseases of aging, and longevity, including the hallmarks mentioned earlier. Clinical studies have explored domains of the aging phenotype such as changes in body composition, energy imbalance, homeostatic dysregulation, and neurodegeneration and the impact of those changes on disease susceptibility, reduced functional reserve, impaired stress response and healing capacity, impaired physical function, disability, and dementia. The NIA also supported establishment of the Translational Geroscience Network (TGN) to develop, implement, and test standard operating procedures for translational early phase trials of agents that target fundamental aging processes and to select, optimize, and validate ancillary measures of fundamental aging processes for use across all trials (R33 AG061456). TGN provides statistical and data management support and has established a biobanking and repository network.



Healthy aging involves both delaying the physiological consequences of aging and maintaining functioning as aging progresses. Interventions thus need to focus on preventing frailty and disability. To develop effective and feasible interventions for healthy aging, whether drugs, exercise, diet, or combinations thereof, will require a deeper understanding of the mechanisms of aging as well as identifying biomarkers that track with biological, not simply chronological age, and predict when an individual is approaching a tipping point and nearing a threshold of irreversible decline.
The pathway to these biomarkers is through the interdisciplinary field of geroscience, which seeks to define the biological mechanisms of aging that give rise to age-related diseases and disorders and to identify targets that may be amenable to different kinds of interventions (15). Developing these biomarkers will require improved cellular and animal models and the capacity to translate discoveries from those models into humans. Also required will be reliable and sensitive measures to assess the hallmarks of aging, for example, assessments of mitochondrial dysfunction, and the impact of interventions on these biological mechanisms and, in turn, the health and functioning of older adults.
The interdisciplinarity of geroscience will be essential to define the complex interactions of the multiple biological, physiological, and behavioral pathways that contribute to age-related declines in health. Interdisciplinarity will also ensure that advances in geroscience are applied to other biomedical disciplines such as neuroscience and cardiology.


Acknowledgements: The authors thank Lisa J. Bain for assistance in the preparation of this manuscript. Dr. LeBrasseur would like to acknowledge the support of the National Institutes of Health, National Institute on Aging grants AG060907, AG055529, and AG061456. The authors are grateful to Dr. Tamara Tchkonia at Mayo Clinic for her thoughtful guidance on contents of Table 1.
Conflicts of interest: NKL- none, DLW-none, MAL-none.
Ethical Standards: None.
Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.



1. Cucinotta D, Vanelli M. WHO Declares COVID-19 a Pandemic. Acta Biomed. 2020 19;91(1):157–60.
2. Morley JE, Vellas B. Editorial: COVID-19 and Older Adults. J Nutr Health Aging. 2020;24(4):364–5.
3. Sierra F. Geroscience and the Coronavirus Pandemic: The Whack-a-Mole Approach is not Enough. Journal of the American Geriatrics Society [Internet]. [cited 2020 May 7];n/a(n/a). Available from: https://onlinelibrary.wiley.com/doi/abs/10.1111/jgs.16489
4. Nikolich-Zugich J, Knox KS, Rios CT, Natt B, Bhattacharya D, Fain MJ. SARS-CoV-2 and COVID-19 in older adults: what we may expect regarding pathogenesis, immune responses, and outcomes. Geroscience. 2020 Apr 10;
5. Kaeberlein M. Translational geroscience: A new paradigm for 21st century medicine. Translational Medicine of Aging. 2017 Oct 1;1:1–4.
6. Kirk B, Al Saedi A, Duque G. Osteosarcopenia: A case of geroscience. Aging Med (Milton). 2019 Sep;2(3):147–56.
7. Kennedy BK, Berger SL, Brunet A, Campisi J, Cuervo AM, Epel ES, et al. Geroscience: linking aging to chronic disease. Cell. 2014 Nov 6;159(4):709–13.
8. Sierra F. Editorial: Geroscience and the Role of Aging in the Etiology and Management of Alzheimer’s Disease. J Prev Alzheimers Dis. 2020;7(1):2–3.
9. World Health Organization. WHO Clinical Consortium on Healthy Ageing. Report of consortium meeting 1-2 December 2016. Geneva, Switzerland: World Health Organization; 2017. Report No.: WHO/FWC/LAC/17.2.
10. Belloni G, Cesari M. Frailty and Intrinsic Capacity: Two Distinct but Related Constructs. Front Med [Internet]. 2019 [cited 2020 May 17];6. Available from: https://www.frontiersin.org/articles/10.3389/fmed.2019.00133/full
11. Ferrucci L, Gonzalez-Freire M, Fabbri E, Simonsick E, Tanaka T, Moore Z, et al. Measuring biological aging in humans: A quest. Aging Cell. 2020;19(2):e13080.
12. Guerville F, De Souto Barreto P, Ader I, Andrieu S, Casteilla L, Dray C, et al. Revisiting the Hallmarks of Aging to Identify Markers of Biological Age. J Prev Alzheimers Dis. 2020;7(1):56–64.
13. Beard JR. Editorial: Linking Geroscience and Integrated Care to Reinforce Prevention. J Prev Alzheimers Dis. 2020;7(2):68–9.
14. López-Otín C, Blasco MA, Partridge L, Serrano M, Kroemer G. The hallmarks of aging. Cell. 2013 Jun 6;153(6):1194–217.
15. Sierra F, Kohanski R. Geroscience and the trans-NIH Geroscience Interest Group, GSIG. Geroscience. 2017;39(1):1–5.
16. Kuo P-L, Schrack JA, Shardell MD, Levine M, Moore AZ, An Y, et al. A roadmap to build a phenotypic metric of ageing: insights from the Baltimore Longitudinal Study of Aging. J Intern Med. 2020 Apr;287(4):373–94.
17. Kane AE, Huizer-Pajkos A, Mach J, Mitchell SJ, de Cabo R, Le Couteur DG, et al. A Comparison of Two Mouse Frailty Assessment Tools. J Gerontol A Biol Sci Med Sci. 2017 Jul 1;72(7):904–9.
18. Rodriguez-Mañas L, Araujo de Carvalho I, Bhasin S, Bischoff-Ferrari HA, Cesari M, Evans W, et al. ICFSR Task Force Perspective on Biomarkers for Sarcopenia and Frailty. J Frailty Aging. 2020;9(1):4–8.
19. Schafer MJ, Zhang X, Kumar A, Atkinson EJ, Zhu Y, Jachim S, et al. The senescence-associated secretome as an indicator of age and medical risk. JCI Insight. 2020 Jun 18;5(12).
20. Rodríguez-Mañas L. Use of Biomarkers. J Frailty Aging. 2015;4(3):125–8.
21. Erusalimsky JD, Grillari J, Grune T, Jansen-Duerr P, Lippi G, Sinclair AJ, et al. In Search of ’Omics’-Based Biomarkers to Predict Risk of Frailty and Its Consequences in Older Individuals: The FRAILOMIC Initiative. Gerontology. 2016;62(2):182–90.
22. Yan R, Zhang Y, Li Y, Xia L, Guo Y, Zhou Q. Structural basis for the recognition of SARS-CoV-2 by full-length human ACE2. Science. 2020 27;367(6485):1444–8.
23. Saedi AA, Feehan J, Phu S, Duque G. Current and emerging biomarkers of frailty in the elderly. Clin Interv Aging. 2019 Feb 19;14:389–98.
24. Butcher L, Carnicero JA, Gomez Cabrero D, Dartigues J-F, Pérès K, Garcia-Garcia FJ, et al. Increased levels of soluble Receptor for Advanced Glycation End-products (RAGE) are associated with a higher risk of mortality in frail older adults. Age Ageing. 2019 01;48(5):696–702.
25. Kochlik B, Stuetz W, Pérès K, Féart C, Tegner J, Rodriguez-Mañas L, et al. Associations of Plasma 3-Methylhistidine with Frailty Status in French Cohorts of the FRAILOMIC Initiative. J Clin Med. 2019 Jul 10;8(7).
26. Kochlik B, Stuetz W, Pérès K, Pilleron S, Féart C, García García FJ, et al. Associations of fat-soluble micronutrients and redox biomarkers with frailty status in the FRAILOMIC initiative. J Cachexia Sarcopenia Muscle. 2019; 10: 1339-46.
27. Pilleron S, Weber D, Pérès K, Colpo M, Gomez-Cabrero D, Stuetz W, et al. Patterns of circulating fat-soluble vitamins and carotenoids and risk of frailty in four European cohorts of older adults. Eur J Nutr. 2019; 58: 379-89.



N.J. Krnavek1, S. Ajasin2, E.C. Arreola2, M. Zahiri1, M. Noun1, P.J. Lupo2, B. Najafi1, M.M. Gramatges2

1. Baylor College of Medicine, Department of Surgery, Division of Vascular Surgery and Endovascular Therapy, USA; 2. Baylor College of Medicine, Department of Pediatrics, Section of Hematology and Oncology, USA
Corresponding author: Maria Monica Gramatges, Baylor College of Medicine, Department of Pediatrics, Feigin Center, 1102 Bates St, Suite 1200, Houston, Texas, 77030, (p) 832-824-4678; (f) 832-825-4651, gramatge@bcm.edu

J Frailty Aging 2020;in press

Published online December 23, 2020, http://dx.doi.org/10.14283/jfa.2020.71



Background: Survivors of childhood cancer (CCS) are at risk for early aging and frailty. Frailty in CCS has been assessed with established clinical criteria, a time-intensive approach requiring specialized training. There is an unmet need for cost-effective, rapid methods for assessing frailty in at-risk adolescent and young adult (AYA) CCS, which are scalable to large populations. Objectives: To validate a sensor-based frailty assessment tool in AYA CCS, compare frailty status between CCS and controls, and assess the correlation between frailty and number of CCS comorbidities. Design, Setting, and Participants: Mean frailty index (MFI) was assessed by a frailty wrist sensor in 32 AYA CCS who were ≥1 year off therapy and in remission. Results were compared with 32 AYA controls without cancer or chronic disease. Measurements: Frailty assessments with and without a simultaneous cognitive task were performed to obtain MFI. Results were compared between cases and controls using a Student t test, and the number of pre-frail/frail subjects by Chi Square test. The contribution of radiation therapy (RT) exposure to MFI was assessed in a sub-analysis, and the correlation between the number of comorbidities and MFI was measured using the Pearson method. Results: MFI was strongly correlated with gait speed in AYA CCS. CCS were more likely to be pre-frail than controls without cancer history (p=0.032), and CCS treated with RT were more likely to be pre-frail than CCS not treated with RT (p<0.001). The number of comorbidities was strongly correlated with MFI (ρ=0.65), with a 0.028 increase in MFI for each added condition (p<0.001). Conclusions: Results from this study support higher risk for frailty among CCS, especially those with multiple comorbidities or who were treated with RT. A wrist-worn sensor-based method is feasible for application in AYA CCS, and provides an opportunity for cost-effective, rapid screening of at-risk AYA CCS who may benefit from early interventions.

Key words: Survivorship, frailty, slowness, weakness.



Over 80% of children diagnosed with cancer survive their disease (1), but survivors of childhood cancer (CCS) are at risk for late effects including early onset aging and frailty (2-10). An increase in frailty accompanies physiologic aging, affecting ~9% of persons >65 years (11) and 25-40% of persons >80 years (12). Frailty identifies individuals more vulnerable to adverse health outcomes (13) and predicts risk for early mortality (14, 15). Compared with sibling controls, CCS are more likely to report poor general health, functional impairment, and activity limitations with a prevalence that increases with age (16, 4, 3). Among 1,922 adult CCS with a mean age of 33.6 +/- 8.1 years, up to 13.1% were frail and up to 31.5% were pre-frail when assessed by the Fried clinical criteria, similar to the frailty prevalence in the general population that is at least three decades older (5). Frailty in CCS has been associated with exposure to cranial radiation and higher doses of abdominal or pelvic radiation (17).
In addition to the Fried clinical criteria, a number of methods assess frailty by clinical or survey-based assessment of factors such as weight loss, exhaustion, poor grip strength, slow gait speed, low physical activity, chronic health conditions, functional impairments, clinical symptoms, and behavior/psychological factors such as poor sleep quality or mood (18). These methods were largely developed for the frail older patient and validated in geriatric populations. Given the high prevalence of frailty in adolescent and young adult (AYA) CCSs, there is an unmet need for efficient, reliable, and low-cost methods that assess frailty across the lifespan.
We conducted a pilot study in AYA CCS leveraging a sensor-based upper extremity frailty assessment tool that has a strong correlation with well-established clinical and survey-based methods of frailty assessment in adults (sensitivity and specificity of 85% and 93% for predicting pre-frailty and 100% sensitivity and specificity for predicting frailty compared with the Fried criteria, sensitivity and specificity of 78% and 82% compared with a validated modified Rockwood questionnaire, and good correlation (ρ = 0.78) with the 6 minute walk test) (19-22). The Fried criteria and Rockwood questionnaire are considered gold standards for measuring frailty in elderly populations, but include components that are not appropriate for application to AYA populations. Therefore, our first objective was to validate the frailty assessment tool by determining correlation between mean frailty index and gait speed, a component of the Fried criteria, in AYA CCS. We then 1) tested feasibility of applying the frailty assessment tool in AYA CCS in an outpatient setting, 2) compared mean frailty index and frailty status between CCS and controls and CCS with or without high-risk treatment exposures (i.e. radiation), and 3) assessed the correlation between mean frailty index and the number of comorbidities present at the time of enrollment.



Study Population

For the frailty meter validation, eligible cases were recruited from the Texas Children’s Cancer and Hematology Centers (TCCC), and included CCS who were ≥15 years old, English-speaking, ≥1 year off therapy, and in remission.
For the feasibility assessment and comparisons between cases and controls, cases were recruited from CCS who met the same above-described eligibility criteria, but were non-overlapping with subjects recruited to the validation step. Controls were ≥15 years old, without cancer history, and recruited from routine well visit patients at a Texas Children’s Pediatric Clinic or Baylor Clinic. For CCS, comorbidities present at the time of enrollment were abstracted from the electronic medical record (EMR). Comorbidities were defined as chronic health conditions listed as an active problem in the EMR Problem List, present for at least one year (determined by problem start date) and requiring ongoing medical care (determined by scheduled subspecialty clinic visits). All cases and controls provided informed consent, as well as assent when applicable, for participation and were enrolled to an IRB-approved research protocol. This research was conducted in accordance with the ethical standards of the Baylor College of Medicine Institutional Review Board (H-38994) and with the 1964 Helsinki declaration and its later amendments.

Frailty Assessment

Mean frailty index (MFI) was determined for each arm by trained personnel using a wrist-worn frailty meter (Figure 1) (26). A wearable sensor is placed on the patient’s wrist, and, while seated, the participant performs a repetitive elbow flexion and extension task as quick and steadily as possible for 20 seconds (single task assessment). The dual task assessment adds a cognitive load to the single task, and in older adults leads to worse performance and higher MFI scores, an effect that is most pronounced among adults with cognitive impairment (27). During the dual task assessment, the subject counts down from a random number provided by the test administrator while performing the task (27). The sensor contains a tri-axial gyroscope that estimates three-dimensional angular velocity of the upper arm and forearm segments. Outcomes measures representing the kinematics and kinetics of elbow flexion, including speed, rise time, and flexion number per 20-second interval (indicators of slowness); flexibility (indicator of rigidity); power and moment (indicators of weakness); speed variability (indicator of steadiness); and speed reduction (indicator of exhaustion) are derived from the angular velocity, anthropometric data, and sex, and captured wirelessly through a tablet device (BioSensics, LLC, Newton, MA, USA). MFI is derived from these outcome measures using an optimized linear regression model previously described by Lee et al., with a numeric output on a continuous scale between 0 and 1 (26). In addition to the MFI, these outcome measures quantify slowness, weakness, and exhaustion (20). Participants completed the single and dual task assessments for both the dominant and non-dominant upper extremities in an average time of just under 5 minutes.

Figure 1
Upper extremity frailty meter sensor and example output. Mean frailty index (MFI) is derived from the measures obtained, in the context of sex, BMI, height, and weight.


Gait Speed Assessment

For the validation step, gait speed was determined by the Timed Up and Go (TUG) test (23), which is both reliable and reproducible in AYAs and for which normative values in this age range are available (24, 25). Briefly, the time (in seconds) required for a subject to rise from a standard armchair, walk a distance of 3 meters, turn, walk back to the chair, and sit down again is measured for each subject.

Quality of Life Assessment

Given the well-described inverse relationship between frailty and quality of life (28), and evidence for impact of frailty on quality of life among CCS (29), we included the PROMIS measure to assess this outcome in conjunction with frailty assessment in our case-control comparisons. Each participant completed the PROMIS questionnaire, a reliable measure of patient-reported health status for mental and physical well-being that has been validated in both children and adults (30). Metrics included in this study were the PROMIS Global Physical Health, Global Mental Health, and Global Health.

Data Analysis

For the validation step, the relationship between MFI and TUG time was determined by the Pearson correlation test. For the subsequent comparisons, participants were classified based on pre-defined MFI thresholds obtained from the dominant arm as robust (MFI <0.20), pre-frail (0.20≤ MFI <0.35), or frail (MFI ≥0.35), using the benchmarks proposed by Rockwood et al (31). After confirming that the data were normally distributed, the mean MFI and MFI subcomponents were compared between cases and controls using a Student t test, and the number of pre-frail/frail subjects by a Chi Square test. For CCS, the relationship between the number of comorbidities and MFI was determined by linear regression, and correlation was tested using the Pearson method. Factors previously associated with frailty include any exposure to cranial radiation in both sexes and abdominal/pelvic radiation in excess of 34Gy/40Gy, respectively, in males) (17). Therefore, we conducted a sub-analysis to compare outcomes between CCS who met criteria for at-risk RT exposure to CCS who did not meet criteria for at-risk RT. PROMIS data were categorized by T score and established cut points for excellent, very good, good, fair, and poor, and then analyzed by the Chi Square test (32).



Seven CCS were recruited for the validation step (two females and five males), mean age of 22.8 years (15-30 years). The mean for the Timed Up and Go (TUG) was 7.87 seconds (range, 5.23-10.12 seconds). Results were comparable to age-normative values for all but two subjects, who each exceeded the upper bound of the 95% confidence interval for the mean by 0.68 seconds and by 1.15 seconds. MFI was 0.19, range 0.13-0.25, and a strong correlation was observed between MFI and TUG time (ρ =0.83, p=0.02).
There were 48 potentially eligible CCS seen for an office visit in the TCCC Late Effects Clinic between June 29 and July 31, 2019. Subjects were recruited four days per week, so that 34 CCS were approached for participation, of which 32 consented and two declined (94% participation rate). All controls who were approached consented to study, so that there were 32 CCS and 32 age-comparable controls that were enrolled. The distribution of primary diagnoses among CCS were as follows: leukemia/lymphoblastic lymphoma, 20; germ cell tumor, 1; Hodgkin/non-Hodgkin lymphoma, 2; rhabdomyosarcoma, 2; Wilm’s tumor, 2; central nervous system tumor, 3; retinoblastoma, 1; ovarian cancer, 1. Participants were a mean of 9.8 years since completion of cancer treatment (median 8.0 years, range 1-36 years). Table 1 shows the distribution of baseline characteristics among CCS and controls.

Table 1
Distribution of demographic characteristics in CCSs and controls


The mean MFI for the dominant arm, both single and dual tasks, was higher among CCS than controls (p=0.002 and p<0.001, respectively, Table 2). The difference in MFI was primarily driven by the weakness and slowness components of MFI, rather than exhaustion. For both the single and dual task assessments in the dominant arm, CCS were more likely to be pre-frail than controls (p=0.032 and p=0.003, respectively). None of the participants met the pre-determined MFI cutoff criteria for frailty.

Table 2
Mean Frailty Index, frailty subcomponents, quality of life, fatigue assessment in CCSs compared with controls

* MFI thresholds for pre-frail and frail were pre-determined from the benchmarks proposed by Rockwood et al, (31) i.e. robust: MFI <0.20, pre-frail: 0.20≤ MFI <0.35, and frail: MFI ≥0.35. All results are displayed as a mean score ± SD


Out of 32 CCS, 13 had at-risk RT exposures: 12 were treated with cranial radiation (12-55.8 Gy), and one male was treated with pelvic RT (50.4 Gy). No females were treated with abdominal or pelvic RT. In the single task assessment, CCS with at-risk RT exposures had a higher MFI (p<0.001) and were more likely to be pre-frail (p<0.001) than CCS without RT exposure (Table 2). As expected, the 12 CCS whose treatment included CRT had a significantly higher MFI than CCS not exposed to CRT, and though there was some evidence of dose-dependence, this difference was not statistically significant (p=0.15, Table 3). No substantial difference in MFI was noted after the addition of a cognitive load (dual task assessment) in controls and cases, regardless of CRT exposure (Tables 2 and 3).

Table 3
Dominant arm, single and dual task MFI determined in controls compared with CCSs without (n=20) and with (n=12) history of exposure to cranial radiation (CRT). Mean MFI increased with increasing dose of CRT

All results are displayed as mean score ± SD


Eleven CCS had no comorbidities, 10 had one condition, 5 had two conditions, 4 had three conditions, 1 had four conditions, and 1 had six conditions, described in more detail in Figure 2. The number of comorbidities was correlated with MFI (ρ = 0.65), with a 0.028 increase in MFI for each added condition (p<0.001).
No significant differences in the proportion of subjects reporting ‘poor’ or ‘fair’ vs. ‘good,’ very good,’ or ‘excellent’ global health, global mental, or global physical health status were observed between controls and CCS.

Figure 2
Distribution of chronic health conditions by system among CCS (n=32)

Cardiometabolic: congestive heart failure (1), hypertension (2), obesity (6); Endocrine: hypogonadism (1), hypothyroidism (2), primary ovarian failure (2), growth hormone deficiency (3); Vision/hearing: cataracts (2), severe visual impairment (2), hearing loss or tinnitus (3); Peripheral/Central nervous system: peripheral neuropathy (1), hemiparesis (1), chronic migraine (1), epilepsy (2); Musculoskeletal: arthritis (1), osteopenia/osteonecrosis (4); Neuropsychological: generalized anxiety disorder (2) major depressive disorder (2); Pulmonary: chronic lung disease (1), obstructive sleep apnea (1); Renal: chronic kidney disease (1); Hematological: chronic pancytopenia (1)



There is considerable need for methods that rapidly, effectively, and inexpensively screen CCS for evidence of pre-frailty or frailty conferred by cancer treatment. To date, studies conducted in CCS have used the clinical Fried criteria, which is both time-consuming and requires training to administer. In elderly persons, the wrist-worn sensor-based method for frailty status determination has a strong correlation with frailty status determined by both the Fried criteria and the Rockwood Frailty Index (19-22), and is a strong predictor of adverse outcomes such as prolonged hospitalizations and prospective falls (35, 36, 27, 37). Given that this tool has largely been applied in older adult populations, we first validated its use in AYA CCS by demonstrating correlation between frailty status determined by wrist-worn sensor and gait speed.
Our study suggests that this method is feasible for application in the outpatient setting. In our pilot study, CCS were more likely to have a higher MFI and be pre-frail than controls, and CCS with at-risk RT exposures were more likely to have a higher MFI and be pre-frail than CCS without a history of at-risk RT, in line with prior reports (5). Of note, the mean age of CCS in this study was 20.5 years, so it is not surprising that no CCS met thresholds for frailty. CRT-exposed CCS showed no discernible difference in MFI with the addition of a cognitive load (dual task) compared with the single task, and compared with non-CRT-exposed CCS. The absence of an effect on MFI observed with addition of a cognitive load suggests that the dual task may be of less importance when assessing frailty in CCS. MFI in CCS was strongly correlated with the number of comorbidities, but the study was not powered to detect associations with diagnosis, treatment dose, or modality other than RT exposure.
Detection of pre-frailty is important, because it offers an opportunity for early intervention in an anticipated trajectory of continued physical decline. The frailty assessment tool described here is a safe, low cost, and time-efficient method, requiring less than 5 minutes to complete compared with the 10-20 minutes required for the Fried method (33, 34). Moreover, it is a stand-alone tool that requires minimal training to use, and does not require additional equipment such as a dynamometer, stopwatch, or 15 foot floor tape. Our pilot data suggest higher MFI in CCS, especially CCS treated with RT compared with controls, and support prospective application of this method to predict risk for morbidity and mortality in CCS, correlated with objective functional and biological measures.


Acknowledgements: The authors would like to thank the patients who participated in this study as well as their families.
Funding: This work was supported by the Texas Children’s Cancer and Hematology Centers.
Conflicts of Interest: None declared by the Authors.
Ethical Standards: This study was approved by the Baylor College of Medicine institutional review board, and conducted in accordance with the Helsinki Declaration of 1975, as revised in 2000.



1. Howlader N, Noone AM, Krapcho N, Neyman R, Aminou S, Altekruse C et al. SEER cancer statistics review, 1975-2009 (vintage 2009 populations). Bethesda, MD: National Cancer Institute; 2012.
2. Hudson MM, Oeffinger KC, Jones K, Brinkman TM, Krull KR, Mulrooney DA et al. Age-dependent changes in health status in the Childhood Cancer Survivor cohort. J Clin Oncol. 2015;33(5):479-91.
3. Ness KK, Hudson MM, Ginsberg JP, Nagarajan R, Kaste SC, Marina N et al. Physical performance limitations in the Childhood Cancer Survivor Study cohort. J Clin Oncol. 2009;27(14):2382-9.
4. Ness KK, Mertens AC, Hudson MM, Wall MM, Leisenring WM, Oeffinger KC et al. Limitations on physical performance and daily activities among long-term survivors of childhood cancer. Ann Intern Med. 2005;143(9):639-47.
5. Ness KK, Krull KR, Jones KE, Mulrooney DA, Armstrong GT, Green DM et al. Physiologic frailty as a sign of accelerated aging among adult survivors of childhood cancer: a report from the St Jude Lifetime cohort study. J Clin Oncol. 2013;31(36):4496-503.
6. Rueegg CS, Michel G, Wengenroth L, von der Weid NX, Bergstraesser E, Kuehni CE et al. Physical performance limitations in adolescent and adult survivors of childhood cancer and their siblings. PLoS One. 2012;7(10):e47944.
7. Armstrong GT, Kawashima T, Leisenring W, Stratton K, Stovall M, Hudson MM et al. Aging and risk of severe, disabling, life-threatening, and fatal events in the childhood cancer survivor study. J Clin Oncol. 2014;32(12):1218-27.
8. Armstrong GT, Reddick WE, Petersen RC, Santucci A, Zhang N, Srivastava D et al. Evaluation of memory impairment in aging adult survivors of childhood acute lymphoblastic leukemia treated with cranial radiotherapy. J Natl Cancer Inst. 2013;105(12):899-907.
9. Henderson TO, Ness KK, Cohen HJ. Accelerated aging among cancer survivors: from pediatrics to geriatrics. American Society of Clinical Oncology educational book / ASCO American Society of Clinical Oncology Meeting. 2014:e423-30.
10. Ness KK, Kirkland JL, Gramatges MM, Wang Z, Kundu M, McCastlain K et al. Premature Physiologic Aging as a Paradigm for Understanding Increased Risk of Adverse Health Across the Lifespan of Survivors of Childhood Cancer. J Clin Oncol. 2018;36(21):2206-15.
11. Collard RM, Boter H, Schoevers RA, Oude Voshaar RC. Prevalence of frailty in community-dwelling older persons: a systematic review. J Am Geriatr Soc. 2012;60(8):1487-92.
12. Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci. 2001;56(3):M146-56.
13. Macklai NS, Spagnoli J, Junod J, Santos-Eggimann B. Prospective association of the SHARE-operationalized frailty phenotype with adverse health outcomes: evidence from 60+ community-dwelling Europeans living in 11 countries. BMC Geriatr. 2013;13:3.
14. Chang SF, Lin PL. Frail phenotype and mortality prediction: a systematic review and meta-analysis of prospective cohort studies. Int J Nurs Stud. 2015;52(8):1362-74.
15. Shamliyan T, Talley KM, Ramakrishnan R, Kane RL. Association of frailty with survival: a systematic literature review. Ageing Res Rev. 2013;12(2):719-36.
16. Ness KK, Morris EB, Nolan VG, Howell CR, Gilchrist LS, Stovall M et al. Physical performance limitations among adult survivors of childhood brain tumors. Cancer. 2010;116(12):3034-44.
17. Hayek S, Gibson TM, Leisenring WM, Guida JL, Gramatges MM, Lupo PJ et al. Prevalence and Predictors of Frailty in Childhood Cancer Survivors and Siblings: A Report From the Childhood Cancer Survivor Study. J Clin Oncol. 2019:JCO1901226.
18. Aguayo GA, Donneau AF, Vaillant MT, Schritz A, Franco OH, Stranges S et al. Agreement Between 35 Published Frailty Scores in the General Population. Am J Epidemiol. 2017;186(4):420-34.
19. Toosizadeh N, Wendel C, Hsu CH, Zamrini E, Mohler J. Frailty assessment in older adults using upper-extremity function: index development. BMC Geriatr. 2017;17(1):117.
20. Toosizadeh N, Mohler J, Najafi B. Assessing Upper Extremity Motion: An Innovative Method to Identify Frailty. J Am Geriatr Soc. 2015;63(6):1181-6.
21. Toosizadeh N, Berry C, Bime C, Najafi B, Kraft M, Mohler J. Assessing upper-extremity motion: An innovative method to quantify functional capacity in patients with chronic obstructive pulmonary disease. PLoS One. 2017;12(2):e0172766.
22. Toosizadeh N, Joseph B, Heusser MR, Orouji Jokar T, Mohler J, Phelan HA et al. Assessing Upper-Extremity Motion: An Innovative, Objective Method to Identify Frailty in Older Bed-Bound Trauma Patients. J Am Coll Surg. 2016;223(2):240-8.
23. Podsiadlo D, Richardson S. The timed “Up & Go”: a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991;39(2):142-8.
24. Nicolini-Panisson RD, Donadio MV. Normative values for the Timed ‘Up and Go’ test in children and adolescents and validation for individuals with Down syndrome. Dev Med Child Neurol. 2014;56(5):490-7.
25. Kear BM, Guck TP, McGaha AL. Timed Up and Go (TUG) Test: Normative Reference Values for Ages 20 to 59 Years and Relationships With Physical and Mental Health Risk Factors. J Prim Care Community Health. 2017;8(1):9-13.
26. Lee H, Joseph B, Enriquez A, Najafi B. Toward Using a Smartwatch to Monitor Frailty in a Hospital Setting: Using a Single Wrist-Wearable Sensor to Assess Frailty in Bedbound Inpatients. Gerontology. 2018;64(4):389-400.
27. Toosizadeh N, Najafi B, Reiman EM, Mager RM, Veldhuizen JK, O’Connor K et al. Upper-Extremity Dual-Task Function: An Innovative Method to Assess Cognitive Impairment in Older Adults. Front Aging Neurosci. 2016;8:167.
28. Kojima G, Iliffe S, Jivraj S, Walters K. Association between frailty and quality of life among community-dwelling older people: a systematic review and meta-analysis. J Epidemiol Community Health. 2016;70(7):716-21.
29. Ness KK, Howell CR, Bjornard KL. Frailty and quality of life in adult survivors of childhood cancer. Expert Rev Qual Life Cancer Care. 2017;2(2):79-85.
30. Cella D, Yount S, Rothrock N, Gershon R, Cook K, Reeve B et al. The Patient-Reported Outcomes Measurement Information System (PROMIS): progress of an NIH Roadmap cooperative group during its first two years. Med Care. 2007;45(5 Suppl 1):S3-S11.
31. Rockwood K, Andrew M, Mitnitski A. A comparison of two approaches to measuring frailty in elderly people. J Gerontol A Biol Sci Med Sci. 2007;62(7):738-43.
32. Hays RD, Spritzer KL, Thompson WW, Cella D. U.S. General Population Estimate for “Excellent” to “Poor” Self-Rated Health Item. J Gen Intern Med. 2015;30(10):1511-6.
33. Kim H, Higgins PA, Canaday DH, Burant CJ, Hornick TR. Frailty assessment in the geriatric outpatient clinic. Geriatr Gerontol Int. 2014;14(1):78-83.
34. Dent E, Kowal P, Hoogendijk EO. Frailty measurement in research and clinical practice: A review. Eur J Intern Med. 2016;31:3-10.
35. Joseph B, Pandit V, Rhee P, Aziz H, Sadoun M, Wynne J et al. Predicting hospital discharge disposition in geriatric trauma patients: is frailty the answer? The journal of trauma and acute care surgery. 2014;76(1):196-200.
36. Joseph B, Pandit V, Zangbar B, Kulvatunyou N, Tang A, O’Keeffe T et al. Validating trauma-specific frailty index for geriatric trauma patients: a prospective analysis. J Am Coll Surg. 2014;219(1):10-7 e1.
37. Joseph B, Toosizadeh N, Orouji Jokar T, Heusser MR, Mohler J, Najafi B. Upper-Extremity Function Predicts Adverse Health Outcomes among Older Adults Hospitalized for Ground-Level Falls. Gerontology. 2017;63(4):299-307.