jfa journal

AND option

OR option

ASSOCIATION OF PATIENT REPORTED GLOBAL HEALTH MEASURES WITH FRAILTY SCORES IN ACUTELY INJURED OLDER ADULTS

 

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

 


Abstract

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.


 

Introduction

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.

 

Methods

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.

 

Results

Demographics

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.

 

Discussion

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.

 

References

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.