jfa journal

AND option

OR option

A SIMPLIFIED APPROACH FOR CLASSIFYING PHYSICAL RESILIENCE AMONG COMMUNITY-DWELLING OLDER ADULTS: THE HEALTH, AGING, AND BODY COMPOSITION STUDY

 

C. Wu1, T.-Z. Lin1, J.L. Sanders2

 

1. Global Health Research Center, Duke Kunshan University, Kunshan, Jiangsu, China; 2. Vertex Pharmaceuticals, Boston, USA.

Corresponding Author: Chenkai Wu, PhD, MPH, MS, Global Health Research Center,Duke Kunshan University, Academic Building 3038, No. 8 Duke Avenue, Kunshan, Jiangsu, China, 215316, Phone: (+86) 512 36657235, E-mail: chenkai.wu@dukekunshan.edu.cn

J Frailty Aging 2022;in press
Published online May 5, 2022, http://dx.doi.org/10.14283/jfa.2022.38

 


Abstract

Background: Physical resilience is an emerging concept within the context of aging and geriatric medicine, and we previously developed and validated one such indicator based on the mismatch between persons’ frailty level and multimorbidity burden. We sought to develop a simplified version for classifying physical resilience. We also examined the agreement between the simplified version and the original approach and evaluated its predictive validity.
Methods: Participants were 2,457 older adults from the Health, Aging, and Body Composition Study. We constructed a simplified version for quantifying physical resilience based on the multimorbidity burden and level of frailty (score: 0-10). Participants were grouped by the number of diseases and classified into three groups—adapters, expected agers, and premature frailers—based on the mean and SD of frailty score (less than, within, or above one standard deviation of the mean).
Results: The Cohen’s kappa between the novel resilience classification and the original approach was 0.70, and the percentage of absolute agreement was 85.4%. We observed a steep increase in years of healthy and able life from premature frailers to adapters in the simplified resilience classifications.
Conclusions: We developed a simplified version for quantifying physical resilience in a cohort of initially well-functioning older Black and White adults. The agreement between the simplified version and the original approach is high. Adapters had a longer healthy lifespan than expected agers and premature frailers. This user-friendly indicator could help assess patients’ physical resilience in clinical settings.

Key words: Resilience, frailty, multimorbidity burden.


 

Introduction

Resilience— “the capacity to bounce back to normal functioning after a perturbation” —is a key concept that has been widely used in various scientific disciplines, including ecology, engineering, social sciences, economics, and public health (1). Multiple conceptual frameworks have been proposed to understand and measure resilience. Bruneau et al. proposed that the concept of resilience was a trait and a stable profile for individuals and could be measured directly (2). However, some studies claimed that resilience was a latent construct which could only be predicted but not measured before the presence of stressors (3, 4). In the absence of a gold standard, previous investigations have proposed a number of ways to quantify resilience guided by different theoretical frameworks, such as the effect modification approach, the psychometric approach, the priori approach, the clustering approach, the complex system approach, and the residual approach (5).
Physical resilience, commonly defined as the ability to recover from aging-related diseases or losses, is an emerging concept within the context of aging and geriatric medicine (6). Researchers have proposed a number of indicators recently to operationalize physical resilience following different conceptual frameworks (7, 8). We previously developed and validated one such indicator—the frailty-disease mismatch approach—in two large longitudinal cohorts of community-dwelling older adults using the residual approach (9). This approach is based on the conjecture that frailty, a multisystem, emergent phenomenon, should manifest synchronously with the development of organ-specific diseases, and that individuals whose degree of frailty is highly mismatched to their level of organ-specific disease may exhibit particular patterns of resilience. Our approach classifies persons into three groups: individuals whose frailty level matches the extent of clinical disease are termed “expected agers”; those with frailty level lower than expected from multimorbidity burden are termed “adapters”; and those whose frailty level is greater than expected from multimorbidity burden are considered “premature frailers”. One intriguing feature of our approach is that it does not rely on measuring health status following a stressor, a commonly employed method. Our approach allows one’s resilience to be classified at any time. This may facilitate the development of interventions that promote resilience preceding a stressor (e.g., surgery). However, widespread clinical adoption of our previously published approach is limited, primarily because identifying an individual’s resilience requires a relatively large number of demographic and health characteristics and a non-negligible amount of time.
In this study, we sought to develop a simplified version of the frailty-disease mismatch approach for classifying physical resilience among older Black and White adults from the Health, Aging and Body Composition (ABC) cohort in the US. Participants were grouped by the number of clinical diseases and classified into three resilience groups—adapters, expected agers, and premature frailers—based on the mean and standard deviation of frailty score (less than, within, or above one SD of the mean). We also examined the agreement between the simplified version and the original approach we have developed and evaluated its predictive validity on various outcomes, including years of able life, years of healthy life, and years of healthy and able life.

 

Methods

Data and Study Population

The Health ABC Study is a longitudinal cohort designed to examine age-related changes in health and body composition and functional limitations in initially well-functioning older adults. Between March 1997 and July 1998, 3,075 Black and White individuals aged 70–79 years were recruited from a list of Medicare beneficiaries provided by the Health Care Financing Administration at two study sites across the United States, Pittsburgh, Pennsylvania, and Memphis, Tennessee. For individuals outside the United States, they are eligible to sign up for Medicare if they meet the following criteria: (i) must be aged 65 years or older, (ii) must be either a US citizen or an alien who has been lawfully admitted for permanent residence and has been residing in the United States for five continuous years prior to the month of filing an application for Medicare (10). The inclusion criteria for our study were (i) free of life-threatening illness, (ii) self-reported ability to walk a quarter of a mile, to climb 10 steps without resting, and to perform basic activities of daily living without assistance, and (iii) no intention to move out of the current geographic area for at least three years. These inclusion criteria resulted in a study population that was healthier than an age-matched general population at enrollment. Details about the Health ABC study design have been described elsewhere (11). All participants provided written informed consent. The study protocol was approved by the institutional review boards of the two clinical sites (University of Pittsburgh and University of Tennessee) and the Data Coordinating Center at the University of California, San Francisco.

Analytic Sample

We used data from the second annual clinic visit, when direct calculation of weight loss—one component for measuring physical frailty—between two consecutive visits was possible. Of the 3,075 participants enrolled at baseline, 2,457 (79.9%) with complete data to construct the resilience variable were included in the present analysis.

Frailty

The level of frailty was measured by the Scale of Aging Vigor in Epidemiology (SAVE) (12), a 10-point frailty scale that was developed to remove the ceiling effect of the original Fried’s physical frailty phenotype and to achieve better differentiation of frailty. Tertiles from the Health ABC Study were considered for each of the five components: walk time, grip strength, exhaustion, physical activity, and weight change. The best tertile received a score of 0, the middle tertile received a score of 1, and the worst tertile received a score of 2. The total score was the sum of the five components, ranging from 0 (least frail) to 10 (most frail). The predictive validity of the SAVE has been demonstrated in multiple cohorts (12, 13). Operational definitions and cut-points of frailty components have been published (9).

Clinical Disease

We used algorithms based on self-reported physician diagnoses, recorded medication use, and laboratory test to define the presence of cancer (excluding non-melanoma skin cancer), coronary heart disease (angina, myocardial infarction, bypass surgery, and angioplasty), heart failure, hypertension, cerebrovascular disease (stroke, transient ischemic attack, and carotid endarterectomy), diabetes mellitus, osteoporosis, osteoarthritis, kidney disease, lung disease (chronic bronchitis, chronic obstructive pulmonary disease, and emphysema), and Parkinson’s disease. Depression was defined as a score greater than 16 on a 20-item Center for Epidemiology Study-Depression scale or use of antidepressant medications (14).

Physical Resilience

Original Approach: We developed and validated a novel physical resilience indicator using a residual approach to capture the degree of mismatch between persons’ frailty level and multimorbidity burden in the Health ABC cohort (9). We used linear regression to regress the level of frailty (score: 0-10) on each clinical disease, multimorbidity burden (indicated by self-rated health and number of medications), age, age (2), age (3), race/ethnicity (Black and White), and sex. We included the polynomial forms of age to allow for non-linear relationship between age and frailty. Inclusion of the quadratic and cubic terms of age also allows for better comparison of our findings to previous studies. Subsequently, residuals from the regression were used to define three aging groups based on values less than, within, or above one SD (1.89) of the mean residual value. Participants whose observed frailty scores were at least 1.89 points higher than their regression-estimated scores were considered premature frailers, those whose observed frailty scores were at least 1.89 points lower than their regression-estimated scores were considered adapters, and those whose frailty scores were within one SD of their regression-estimated scores were considered expected agers.

Simplified Approach

In this study, we proposed a simplified version for quantifying physical resilience merely based on the total number of clinical diseases and the level of frailty (score: 0-10). First, we computed the overall SD of the frailty score for all the participants. Next, we grouped the participants by the total count of clinical diseases (0-8) and computed the average frailty score for each group. Within each group, participants whose observed frailty scores were at least one SD above the group average were considered “premature frailers”, those whose observed frailty scores were at least one SD below the group average were considered “adapters”, and those whose frailty scores were within one SD of the group average were considered “expected agers”.

Outcomes

Outcomes include years of able life (YAL), years of healthy life (YHL), and years of healthy and able life (YHAL). YAL was calculated as the number of years the participant did not have disability. YHL was calculated as the number of years the participant reported good or better health on a scale of excellent, very good, good, fair, or poor. YHAL was calculated as the number of years the participant was in good health and had no disability. All outcomes were assessed starting after the second clinic visit.

Covariates

Socio-demographics included study site (Pittsburgh and Memphis), sex (male and female), age (years), and education (less than high school, high school or equivalent, and more than high school). Behavioral characteristics included the smoking status (current, former, and never).

Statistical Analyses

We first examined the agreement between the two simplified methods and the physical resilience indicator developed using the residual approach with regard to the categorical definition of physical resilience (premature frailer, expected ager, and adapter) using the percentage of absolute agreement and Cohen’s kappa statistic. For each simplified method, we then compared demographic and lifestyle characteristics between three groups—premature frailer, expected ager, and adapter—using analysis of variance F test or non-parametric equivalent (e.g., Kruskal-Wallis test) for continuous variables and chi-square test for categorical variables. We calculated the mean and SD of each outcome (YAL, YHL, and YHAL) by each resilience group. To estimate the adjusted associations between the resilience classifications and the health outcomes, we also conducted a series of linear regressions to compare YAL, YHL, and YHAL between premature frailers, expected agers, and adapters, adjusting for study site (Memphis or Pittsburgh), education in years, sex (male or female), race (Black or White), age and smoking status (never, previous, or current). All tests were two-sided with a significance level of P <.05. All analyses were conducted using Stata 15.0 (Stata Corp, College Station, TX) and R version 4.1.0.

 

Results

Table 1 showed the cut-points of resilience classification based on the total number of clinical diseases and frailty score for two simplified classification approaches. Of the 2,457 study participants, 1,671 (68.0%), 373 (15.2%), and 413 (16.8%) were classified as expected agers, premature frailers, and adapters, respectively. The overall SD for classifying these groups was 2.08. The percentage of absolute agreement between this simplified approach and the original resilience indicator was 85.4%, where 1,479 (60.2%) expected agers, 312 (12.7%) prematurely frailers, and 308 (12.6%) adapters shared the same classification. The Cohen’s kappa between the novel resilience classification and the existing resilience indicator was 0.70. There were no differences in sex, race, education level, smoking and BMI between the three resilience groups. The premature frailers were older than the expected agers and adapters (p for comparison < 0.001).

Table 1. Cut-points of resilience classification based on the total number of clinical diseases and frailty, N=2,457

* The frailty score of expected agers is the interval between the cut-points for adapters and premature frailers; † We computed the overall SD (2.08) of the frailty score. Participants whose frailty scores were at least one SD above, within one SD of, and at least one SD below the group average were defined as “premature frailers”, “expected agers”, and “adapters”, respectively.

 

We observed a steep increase in YAL, YHL, and YHAL from premature frailers to adapters in the simplified resilience classification (Figure 1). The average YAL was 5.5, 9.1, and 11.0 years among premature frailers, expected agers, and adapters, respectively (p for comparison < 0.001), and the corresponding YHL was 6.5, 9.7 and 11.6 years (p for comparison < 0.001). The average YHAL was 4.0, 7.7, and 9.9 years among premature frailers, expected agers, and adapters, respectively (p for comparison < 0.001)

Figure 1. Unadjusted association between the resilience group and years of able life (YAL), years of healthy life (YHL), and years of healthy and able life (YHAL)

P-value for each of three comparisons across resilience group was <.001. Note: The 95% confidence interval for YAL was (10.5, 11.4), (8.8, 9.4), and (5.1, 6.0) for adapters, expected agers, and premature frailers, respectively. The 95% confidence interval for YHL was (11.1, 12.1), (9.5, 10.0), and (6.0, 7.0) for adapters, expected agers, and premature frailers, respectively. The 95% confidence interval for YHAL was (9.4, 10.4), (7.5, 8.0), and (3.6, 4.4) for adapters, expected agers, and premature frailers, respectively.

 

We conducted a series of linear regressions to compare YAL, YHL, and YHAL between premature frailers, expected agers, and adapters. In each regression model, we adjusted for study site (Memphis or Pittsburgh), education in years, sex (male or female), race (Black or White), age and smoking status (never, previous, or current). Persons classified as adapters had a longer YHAL than expected agers (β = 1.81), while being premature frailers was associated with a shorter healthy and able lifespan (β = -3.35). Such association between other health outcomes (YAL and YHL) and resilience groups persisted in the multivariable regression model. We also conducted the additional analyses of resilience indicator with multimorbidity burden. After accounting for the number of clinical diseases in the regression model, both being adapter and having fewer diseases was associated with a longer YAL, YHL, and YHAL, and none of the resilience indicators was associated with the count of clinical diseases (data not shown).

Table 2. Adjusted association between resilience groups and health outcomes in the multivariable regression model, N=2,457

Abbreviations: CI, confidence interval; Note: We adjusted for study site (Memphis or Pittsburgh), education in years, sex (male or female), race (Black or White), age and smoking status (never, previous, or current) in each regression model. * We computed the overall SD (2.08) of the frailty score. Participants whose frailty scores were at least one SD above, within one SD of, and at least one SD below the group average were defined as “premature frailers”, “expected agers”, and “adapters”, respectively.

 

Discussion

The present study aimed to develop a simplified version of the frailty-disease mismatch approach for classifying physical resilience among black and white US older adults. We found that the agreement between the simplified version and the original approach we have developed is satisfactory. We also validated the predictive validity of the new simplified approach and found that adapters—whose frailty level is lower than expected from their clinical multimorbidity burden—spent more of their remaining life in health and disability-free status than expected agers and premature frailers.
Physical resilience has been widely accepted as a construct that captures the ability to recover from stressors in gerontology and geriatrics. In the absence of a gold standard, multiple operational definitions have been proposed (15–19). We have proposed the frailty-disease mismatch theory to conceptualize physical resilience as the capacity to adapt to and mitigate the consequences of cumulative damage in organ systems. In this sense, the degree of mismatch between individuals’ frailty level and multimorbidity burden is suitable to quantify their physical resilience. Our approach for classifying resilience is potentially useful to build clinical prediction tools for quantifying the likelihood to recover when facing an acute stressor (e.g., stroke and surgery), which may aid in the identification of older patients needing more intensive care. The simplified physical resilience indicator we developed in the present study is able to overcome major limitations of the original approach (i.e., data unavailability and long administration time) while preserving predictive value. The original approach takes into account for demographic features. It, therefore, provides a more accurate and granular indicator of physical resilience and would be an ideal choice in research. The simplified approach adopts a cut-point based definition and takes much less time to implement, which might facilitate a rapid assessment of physical resilience in clinical settings.
Our study has several limitations. First, although we constructed the concept of physical resilience based on the mismatch between the multimorbidity burden and the frailty levels, future research could explore other stressors (e.g., physical, social, and environmental) to further validate the construct and predictive validity of our approach for quantifying physical resilience. Second, comparing to the original method, our simplified approach weighted each disease equally and did not account for the severity of diseases for simplicity purposes. These two approaches may serve different purposes. The original approach accounted for the severity of clinical diseases by incorporating medication use and adjusted for demographic differences, providing a more accurate quantification to be employed in the research setting. The simplified approach relies on a simple, unweighted sum of chronic conditions, which could promote the uptake of resilience indicator in busy routine clinical care settings. The physical frailty phenotype and clinical diseases are widely available in population-based studies and can be constructed using electronic medical records (20). Future research using other data sources is needed to examine the generalizability and external validity of our approaches. Lastly, the Health ABC participants were healthier than an age-matched general population at enrollment. We, therefore, may underestimate the frailty levels and the proportion of premature frailers among older adults.
In conclusion, we developed a simplified version of the frailty-disease mismatch approach in a cohort of initially well-functioning older Black and White adults in the US. The agreement between the simplified version and the original approach we have developed is high. The predictive value of the new approach has been demonstrated; persons classified as adapters had a longer healthy lifespan than expected agers and premature frailers. This user-friendly indicator of physical resilience could help assess patients’ physical resilience in clinical settings.

 

Funding: This work is supported by Chinese Ministry of Science and Technology (2020YFC2005600), National Institute on Aging (N01-AG-6-2101; N01-AG-6-2103; N01-AG-6-2106; NIA grant R01-AG028050, and NINR grant R01-NR012459), and Intramural Research Program of the NIH, National Institute on Aging. The sponsors had no role in the design and conduct of the study; in the collection, analysis, and interpretation of data; in the preparation of the manuscript; or in the review or approval of the manuscript.

Acknowledgement: C.W. and T.L. drafted the manuscript, and T.L. performed the analysis. J.L.S. provided critical comments during manuscript revisions. All authors approved the final version of the manuscript. The funding sponsor had no role in the analysis, drafting of the manuscript, or the decision to publish.

Conflict of interest: None declared.

 

SUPPLEMENTARY MATERIAL

References

1. Scheffer M, Bolhuis JE, Borsboom D, et al. Quantifying resilience of humans and other animals. Proc Natl Acad Sci U S A. 2018;115(47):11883-11890. doi:10.1073/pnas.1810630115
2. Bruneau M, Chang SE, Eguchi RT, et al. A Framework to Quantitatively Assess and Enhance the Seismic Resilience of Communities. Earthquake Spectra. 2003;19(4):733-752. doi:10.1193/1.1623497
3. Kalisch R, Baker DG, Basten U, et al. The resilience framework as a strategy to combat stress-related disorders. Nat Hum Behav. 2017;1(11):784-790. doi:10.1038/s41562-017-0200-8
4. Links JM, Schwartz BS, Lin S, et al. COPEWELL: A Conceptual Framework and System Dynamics Model for Predicting Community Functioning and Resilience After Disasters. Disaster Med Public Health Prep. 2018;12(1):127-137. doi:10.1017/dmp.2017.39
5. Kok AAL, Huisman M, Cosco TD, Melis RJF. Quantitative Approaches to Examine Resilience and Aging. In: Wister AV, Cosco TD, eds. Resilience and Aging: Emerging Science and Future Possibilities. Risk, Systems and Decisions. Springer International Publishing; 2020:107-136. doi:10.1007/978-3-030-57089-7_6
6. Whitson HE, Duan-Porter W, Schmader KE, Morey MC, Cohen HJ, Colón-Emeric CS. Physical Resilience in Older Adults: Systematic Review and Development of an Emerging Construct. The Journals of Gerontology: Series A. 2016;71(4):489-495. doi:10.1093/gerona/glv202
7. Colón-Emeric C, Pieper CF, Schmader KE, et al. Two Approaches to Classifying and Quantifying Physical Resilience in Longitudinal Data. J Gerontol A Biol Sci Med Sci. 2020;75(4):731-738. doi:10.1093/gerona/glz097
8. Varadhan R, Walston JD, Bandeen-Roche K. Can a Link Be Found Between Physical Resilience and Frailty in Older Adults by Studying Dynamical Systems? J Am Geriatr Soc. 2018;66(8):1455-1458. doi:10.1111/jgs.15409
9. Wu C, Li YX, Marron MM, Odden MC, Newman AB, Sanders JL. Quantifying and Classifying Physical Resilience Among Older Adults: The Health, Aging, and Body Composition Study. J Gerontol A Biol Sci Med Sci. 2020;75(10):1960-1966. doi:10.1093/gerona/glz247
10. Original Medicare (Part A and B) Eligibility and Enrollment | CMS. Accessed January 28, 2022. https://www.cms.gov/Medicare/Eligibility-and-Enrollment/OrigMedicarePartABEligEnrol
11. Newman AB, Haggerty CL, Goodpaster B, et al. Strength and muscle quality in a well-functioning cohort of older adults: the Health, Aging and Body Composition Study. J Am Geriatr Soc. 2003;51(3):323-330. doi:10.1046/j.1532-5415.2003.51105.x
12. Sanders JL, Boudreau RM, Fried LP, Walston JD, Harris TB, Newman AB. Measurement of organ structure and function enhances understanding of the physiological basis of frailty: the Cardiovascular Health Study. J Am Geriatr Soc. 2011;59(9):1581-1588. doi:10.1111/j.1532-5415.2011.03557.x
13. Sanders JL, Singh J, Minster RL, et al. Association Between Mortality and Heritability of the Scale of Aging Vigor in Epidemiology. J Am Geriatr Soc. 2016;64(8):1679-1683. doi:10.1111/jgs.14190
14. Orme JG, Reis J, Herz EJ. Factorial and discriminant validity of the Center for Epidemiological Studies Depression (CES-D) scale. J Clin Psychol. 1986;42(1):28-33. doi:10.1002/1097-4679(198601)42:1<28::aid-jclp2270420104>3.0.co;2-t
15. Resnick B, Galik E, Dorsey S, Scheve A, Gutkin S. Reliability and Validity Testing of the Physical Resilience Measure. The Gerontologist. 2011;51(5):643-652. doi:10.1093/geront/gnr016
16. Duan-Porter W, Cohen HJ, Demark-Wahnefried W, et al. Physical resilience of older cancer survivors: An emerging concept. Journal of Geriatric Oncology. 2016;7(6):471-478. doi:10.1016/j.jgo.2016.07.009
17. Arbeev KG, Ukraintseva SV, Bagley O, et al. “Physiological Dysregulation” as a Promising Measure of Robustness and Resilience in Studies of Aging and a New Indicator of Preclinical Disease. The Journals of Gerontology: Series A. 2019;74(4):462-468. doi:10.1093/gerona/gly136
18. Gijzel SMW, van de Leemput IA, Scheffer M, Roppolo M, Olde Rikkert MGM, Melis RJF. Dynamical Resilience Indicators in Time Series of Self-Rated Health Correspond to Frailty Levels in Older Adults. The Journals of Gerontology: Series A. 2017;72(7):991-996. doi:10.1093/gerona/glx065
19. Resnick BA, Inguito PL. The Resilience Scale: Psychometric Properties and Clinical Applicability in Older Adults. Archives of Psychiatric Nursing. 2011;25(1):11-20. doi:10.1016/j.apnu.2010.05.001
20. Segal JB, Chang HY, Du Y, Walston J, Carlson M, Varadhan R. Development of a Claims-Based Frailty Indicator Anchored to a Well-Established Frailty Phenotype. Med Care. 2017;55(7):716-722. doi:10.1097/MLR.0000000000000729