Abstract
Purpose:
To determine the time-varying associations between 1) decreased handgrip strength and disabilities in each activity of daily living (ADL) function, and 2) disaggregated ADL limitations and time to mortality in older adults.
Methods:
A United States nationally representative sample of 17,747 older adults from the Health and Retirement Study were followed up for 8 yr. Maximal handgrip strength was measured with a hand-held dynamometer. Ability to perform ADL was self-reported. Date of death was identified by the National Death Index and exit interviews. Separate covariate-adjusted hierarchical logit models were used to examine the time-varying associations between decreased handgrip strength and each ADL outcome. Distinct covariate-adjusted Cox models were used to analyze the time-varying associations between disaggregated ADL limitations and time to mortality.
Results:
Every 5-kg decrease in handgrip strength was associated with increased odds for the following ADL limitations: 20% for eating, 14% for walking, 14% for bathing, 9% for dressing, 8% for transferring, and 6% for toileting. The presence of a bathing, walking, toileting, eating, and dressing ADL disability was associated with a 47%, 43%, 32%, 30%, and 19% higher hazard for mortality, respectively. A transferring ADL disability was not significantly associated with mortality.
Conclusions:
Decreased handgrip strength was associated with increased odds for each ADL limitation, and in turn, most individual ADL impairments were associated with a higher hazard for mortality in older adults. These findings provide insights into the disabling process by identifying which ADL limitations are most impacted by decreased handgrip strength and the subsequent time to mortality for each ADL disability.
Keywords: ACTIVITIES OF DAILY LIVING, GERIATRICS, DEATH, DISABLED PERSONS, MUSCLE STRENGTH, MUSCLE WEAKNESS
Age-related declines in muscle mass and strength occurs in older adults, even in those that are relatively healthy and weight stable (1). Handgrip strength is a simple and inexpensive, clinically viable screening tool for estimating overall muscle strength, and robust biomarker of aging (2). Previous longitudinal investigations have revealed that low handgrip strength is associated with the incidence of an activities of daily living (ADL) disability and all-cause mortality in older adults (3,4). Handgrip strength has been shown to be associated with several other clinically relevant health outcomes and is therefore an important health factor when evaluating the disabling process (5).
Questionnaires regarding ADL are often used to determine functional status in older adults. Respondents who have difficulty or are unable to perform one or more ADL have an ADL disability. The prevalence of a limitation is higher for some individual ADL than others; however, a disability in any of the ADL is associated with an elevated risk of mortality (6,7). Using traditional definitions of ADL disability may not provide the necessary details for understanding how different health factors are associated with each ADL function, and subsequent health outcomes. For example, older adults with cognitive impairments often experience limitations with bathing, and the presence of a bathing ADL impairment is associated with incident nursing home placement (8,9). Therefore, it is possible that certain health factors may differentially impact disabilities in individual ADL tasks, which, in turn, affects future health outcomes.
Cross-sectional and some longitudinal study designs may not account for the changes in handgrip strength that occur throughout the life course, nor do they capture the mutable nature or fluidity of ADL functions (10,11). For example, longitudinal investigations with repeated measures will often include participants’ handgrip strength and exclude those with any ADL limitation at baseline when determining the association between handgrip strength and time to incident ADL disability. However, incorporating longitudinal study designs when examining the association between handgrip strength and ADL functions, with the use of time-varying covariates, provides a deeper understanding for how changes in handgrip strength and functional status change over time. Acknowledging such fluctuations in handgrip strength and functional status may influence how handgrip strength, ADL, and clinically relevant health outcomes are connected.
More research is needed for investigating the links between functional disabilities and mortality in older adults (12). The preservation of muscle strength is considered to be an important factor for slowing the disabling process. Understanding how handgrip strength affects the capacity of individual ADL, and the subsequent time to mortality for specific ADL impairments using a longitudinal, time-varying approach, may provide insights into the disabling process. Such information could be used to inform interventions designed to preserve function and delay mortality (13). Accordingly, the purposes of this study were to determine the time-varying associations between 1) decreased handgrip strength and disabilities in individual ADL tasks and 2) disaggregated ADL limitations and time to mortality in older adults.
METHODS
Participants
Data from the Health and Retirement Study (HRS) were used. Cleaned and standardized RAND HRS data were joined with other necessary data components from the HRS files. The HRS is designed to monitor the health and financial status of aging Americans and provides data for a nationally representative sample of community-dwelling adults older than 50 yr, including surveys from approximately 23,000 households (14). Since 1992, HRS participants have been reinterviewed biannually. The HRS follows respondents longitudinally until death, and new cohorts have been added to the original sample to maintain national representation of the survey over time. A multistage probability design is used by the HRS, including geographical stratification and oversampling of certain demographic groups (blacks and Hispanics). Sample weights are provided to account for the multistage, area probability design, and they were used in the analyses. Details for the HRS are described elsewhere (15).
Individuals who participated in the 2006 wave of the HRS were included and followed for 8 yr (2008, 2010, 2012, and 2014 waves). Since the 2006 wave, the HRS has used a mixed-mode design for follow-up, wherein half of the sample completes enhanced face-to-face interviews with physical and biological measures, and a psychosocial questionnaire. The other half only completes the core interview, typically by telephone (14). Therefore, the half-samples alternate completion of enhanced interviews at each wave to minimize participant burden and costs. Interview response rates for each wave of the HRS are >80%. Proxy respondents are conducted for those who are unwilling or unable to complete interviews themselves. Approximately 9% of proxy interviews occur at each wave.
Each participant from the HRS provided written informed consent and the HRS protocols were approved by the University of Michigan Behavioral Sciences Committee Institutional Review Board. Data used in these analyses contained no direct identifiers, thereby ensuring participant anonymity.
Measures
Descriptive variables.
Participants self-reported their height and weight at each wave. Body mass index (BMI) was calculated as weight in kilograms divided by height in meters-squared. Race and ethnicity (White, Black, Hispanic), age, and sex were also reported.
Handgrip strength.
A Smedley spring-type hand-held dynamometer (Scandidact, Denmark) was used to assess handgrip strength. After adjusting the dynamometer to fit the hand size of each participant, a practice trial was performed in the standing position with the arm at the side and elbow flexed at a 90° angle. If a participant was unable to stand or hold the dynamometer, the measurement was performed in the seated position with the upper arm resting on an object for support. Starting on the nondominant hand, participants squeezed the dynamometer with maximal effort, and then let go. Participants performed two measurements with each hand, alternating between hands. If only one hand could be used for the handgrip strength assessment, participants waited 30 s between measurements (16). The highest handgrip strength measurement was included in the analyses. Those who had surgery, swelling, inflammation, severe pain, or an injury in both hands did not engage in handgrip strength testing. Handgrip strength assessments were a part of the enhanced face-to-face interviews. Participants had handgrip strength measures for the 2006, 2010, and 2014 waves, or 2008 and 2012 waves. Therefore, measures from either 2006 or 2008 were used for both waves and likewise for the 2010 or 2012 waves.
Cognition.
At each wave, cognitive function was evaluated with an array of tests that were modified from the Telephone Interview of Cognitive Status, a validated cognitive screening instrument from the Mini-Mental State Examination that was designed for population based studies (17). For self-respondents younger than 65 yr, a 27-point composite measure that included immediate and delayed word recall from a list of 10 words (0–20 points), serial sevens subtraction test starting from 100 (0–5 points), and counting backward as quickly as possible for 10 continuous numbers beginning with the number 20 (0–2 points). Those with scores of six or less were considered severely cognitive impaired (18).
Cognitive function for self-respondents age at least 65 yr was assessed with a 35-point scale. The additional assessments that were included on the 35-point scale were mental status questions that included object naming (0–2 points), date naming (0–4 points), and correct identification of the current president and vice president of the United States (0–2 points). Participants with scores of seven or less were considered severely cognitive impaired (19).
Those represented by a proxy had their cognitive function assessed with an 11-point scale. The proxy’s assessment of the respondent’s memory had responses ranging from excellent to poor (0–4 points), limitations in five instrumental ADL (managing money, taking medication, preparing hot meals, using a telephone, and shopping for groceries; 0–5 points), and the survey interviewer’s evaluation of if the respondent had difficulty completing the interview due to a cognitive impairment (none, some, and prevents completion; 0–2 points). Individuals with scores of at least six were considered as having a severe cognitive impairment (18).
Morbidity.
Participants reported if a doctor had ever diagnosed them with high blood pressure or hypertension, diabetes or high blood sugar, cancer or a malignant tumor (excluding minor skin cancer), lung disease (e.g., bronchitis or emphysema), a heart condition (e.g., coronary heart disease, angina, congestive heart failure), stroke, emotional or psychiatric problems, and arthritis or rheumatism. At each wave, the number of affirmative diagnoses for each health condition was included in the analyses.
Cigarette smoking.
Separate single-item indicators of cigarette smoking were measured at each wave. Participants reported if they were currently smoking cigarettes (yes, no). Likewise, participants told interviewers if they have ever smoked over 100 cigarettes in their lifetime (yes, no).
Depression.
An eight-item Center for Epidemiologic Studies Depression (CES-D) scale was used to assess mental health at each wave. The eight-item version of the CES-D scale used in the HRS has an equivalent reliability and validity to the 20-item CES-D scale (20). Participants indicated whether any of the negative (was depressed, everything was an effort, sleep was restless, felt lonely, felt sad, and could not get going) or positive indicators (was happy, enjoyed life; reverse scored) for depressive symptoms occurred in the week before the interview date. Scores ranged from zero-to-eight with higher scores indicating more depressive symptoms. The continuous score was used in the analyses.
Self-rated health status.
A single-item measure of self-rated health was taken at each wave. Participants were asked to assess their health as excellent, very good, good, fair, or poor. The categorical score was used in the analyses.
Activities of daily living.
Participants self-reported their ability to perform six ADL at each wave: dressing, eating, transferring in or out of bed, toileting, bathing, and walking across a room. Those indicating difficulty or an inability to perform an ADL were considered disabled for that particular ADL function.
Mortality.
Date of death was monitored through linkage to the National Death Index. Approximately 93% of exit interviews were conducted with a surviving spouse, child, or other informant to collect information about medical expenditures, family interactions, outlook of assets after death, and other circumstances that may have occurred toward the end of life (14).
Statistical Analysis
Separate hierarchical logistic regression models were used to examine the time-varying associations between decreased handgrip strength and disabilities in each ADL function after adjusting for age, sex, race and ethnicity, BMI, cognitive impairment, morbidity, CES-D score, current smoking status, smoking history, and self-rated health. Participants were nested within waves with a random residual and unstructured covariance. A between- and within-participant degrees of freedom was used.
Distinct Cox proportional hazard regression models were used to assess the time-varying associations between disaggregated ADL limitations and time to mortality after adjusting for handgrip strength, age, sex, race and ethnicity, BMI, cognitive impairment, morbidity, CES-D score, current smoking status, smoking history, and self-rated health. The number of days since entering the study until death or date of last interview was the time variable, and date of birth was the entry variable. Participants were censored if they had not died by the 2014 wave, or if they were lost to follow-up and their death status could not be ascertained. A flowchart for those included in this investigation is presented in Figure 1. Procedures were used to account for the complex sampling design of the HRS (proc surveyphreg). Analyses were performed with SAS 9.4 software (Cary, NC) and an alpha level of 0.05 was used.
FIGURE 1— Data flow diagram for those included in the present investigation. ‡Handgrip strength measures in this wave could not be concatenated with the other half sample (2016 wave) because those data have not yet been released.
RESULTS
Of the 18,469 participants in the 2006 wave, exclusions occurred for unknown death dates (n = 14) and one or more missing covariates at all five waves (n = 708). After exclusions, there were 17,747 participants included (96.1%) and their descriptive characteristics are shown in Table 1. A Sankey bar chart (21) is presented in Figure 2 to illustrate how the number of ADL limitations fluctuated across waves for those with at least one ADL impairment. Although participants may have had improvements, persistence, or declines in their function over time, overall participants experienced increases in ADL limitations as they age. For example, of the participants that had at least one ADL limitation, 41% had a single impairment and 3% were disabled in all six ADL functions for the 2006 wave; whereas, 16% had a single impairment and 12% were disabled in all six ADL functions for the 2014 wave.
TABLE 1.
Descriptive characteristics of the participants.
| 2006 Wave (n = 17,747) | 2008 Wave (n = 15,771) | 2010 Wave (n = 14,033) | 2012 Wave (n = 12,805) | 2014 Wave (n = 11,346) | |
|---|---|---|---|---|---|
|
| |||||
| Handgrip strength (kg) | 31.4 ± 11.2 | 31.5 ± 11.2 | 31.5 ± 14.1 | 31.6 ± 14.2 | 28.9 ± 10.9 |
| Age (yr) | 68.0 ± 11.1 | 69.3 ± 10.7 | 70.7 ± 10.3 | 71.8 ± 9.9 | 72.9 ± 9.6 |
| Female, n (%) | 10,458 (58.9%) | 9370 (59.4%) | 8379 (59.7%) | 7701 (60.1%) | 6924 (61.0%) |
| White, n (%) | 14,367 (81.0%) | 12,801 (81.2%) | 11,357 (80.9%) | 10,362 (80.9%) | 9142 (80.6%) |
| Black, n (%) | 2496 (14.1%) | 2184 (13.9%) | 1954 (13.9%) | 1760 (13.7%) | 1577 (13.9%) |
| Hispanic, n (%) | 1603 (9.0%) | 1438 (9.1%) | 1280 (9.1%) | 1208 (9.4%) | 1078 (9.5%) |
| BMI (kg-m −2) | 27.8 ± 5.9 | 28.0 ± 5.9 | 28.1 ± 6.0 | 28.1 ± 6.0 | 28.1 ± 6.0 |
| Cognitive impairment, n (%) | 1474 (8.3%) | 1129 (7.2%) | 1048 (7.5%) | 857 (6.7%) | 882 (7.8%) |
| Number of morbid conditions | 2.0 ± 1.5 | 2.2 ± 1.5 | 2.3 ± 1.5 | 2.5 ± 1.5 | 2.6 ± 1.5 |
| Depression score | 1.5 ± 2.0 | 1.4 ± 2.0 | 1.4 ± 1.9 | 1.4 ± 2.0 | 1.4 ± 2.0 |
| Current smoker, n (%) | 2438 (13.8%) | 2000 (12.7%) | 1611 (11.5%) | 1383 (10.8%) | 1074 (9.5%) |
| Previous smoker, n (%) | 10,090 (57.1%) | 8902 (56.5%) | 7876 (56.1%) | 7090 (55.4%) | 6206 (54.7%) |
| Self-rated health, n (%) | |||||
| Excellent | 1954 (11.0%) | 1403 (8.9%) | 1250 (8.9%) | 1067 (8.4%) | 814 (7.2%) |
| Very good | 5075 (28.6%) | 4536 (28.8%) | 4291 (30.6%) | 3871 (30.3%) | 3307 (29.2%) |
| Good | 5413 (30.6%) | 5063 (32.1 %) | 4521 (32.2%) | 4127 (32.2%) | 3860 (34.0%) |
| Fair | 3707 (20.9%) | 3294 (20.9%) | 2817 (20.1%) | 2621 (20.5%) | 2463 (21.7%) |
| Poor | 1575 (8.9%) | 1464 (9.3%) | 1147 (8.2%) | 1104 (8.6%) | 894 (7.9%) |
| Bathing ADL limitation, n (%) | 1455 (8.2%) | 1339 (8.5%) | 1325 (9.4%) | 1211 (9.5%) | 1121 (9.9%) |
| Transferring ADL limitation, n (%) | 1307 (7.4%) | 1105 (7.0%) | 1099 (7.8%) | 986 (7.7%) | 885 (7.8%) |
| Dressing ADL limitation, n (%) | 1958 (11.0%) | 1721 (10.9%) | 1735 (12.4%) | 1509 (11.8%) | 1451 (12.8%) |
| Eating ADL limitation, n (%) | 759 (4.3%) | 620 (3.9%) | 701 (5.0%) | 633 (4.9%) | 561 (4.9%) |
| Toileting ADL limitation, n (%) | 1280 (7.2%) | 1064 (6.8%) | 1088 (7.8%) | 968 (7.6%) | 888 (7.8%) |
| Walking ADL limitation, n (%) | 1469 (8.3%) | 1333 (8.5%) | 1255 (8.9%) | 1150 (9.0%) | 1040 (9.2%) |
| Died before next wave, n (%) | 1191 (6.7%) | 1458 (9.2%) | 1033 (7.3%) | 1030 (8.0%) | |
FIGURE 2— Sankey bar chart for depicting changes in the number of ADL limitations at each wave.
The results for the time-varying associations between handgrip strength and each ADL outcome are displayed in Table 2. Every 5-kg decrease in handgrip strength was associated with the following odds ratios (OR) for each ADL limitation: 1.20 (95% confidence intervals [CI], 1.14–1.26) for eating, 1.14 (95% CI, 1.12–1.19) for walking, 1.14 (95% CI, 1.12–1.19) for bathing, 1.09 (95% CI, 1.07–1.11) for dressing, 1.08 (95% CI, 1.06–1.12) for transferring, and 1.06 (95% CI, 1.04–1.08) for toileting.
TABLE 2.
The time-varying associations between handgrip strength and individual ADL limitations.
| Dressing |
Eating |
Transferring |
Toileting |
Bathing |
Walking |
|||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | OR | 95% CI | OR | 95% CI | OR | 95% CI | OR | 95% CI | OR | 95% CI | |
|
| ||||||||||||
| Handgrip strength (5-kg decrease) | 1.09 | 1.07–1.11 | 1.20 | 1.14–1.26 | 1.08 | 1.06–1.12 | 1.06 | 1.04–1.08 | 1.14 | 1.12–1.19 | 1.14 | 1.12–1.19 |
| Age | 1.03 | 1.03–1.04 | 1.02 | 1.01–1.02 | 0.99 | 0.99–1.00 | 1.03 | 1.02–1.03 | 1.04 | 1.04–1.05 | 1.04 | 1.04–1.05 |
| Male (ref: female) | 1.88 | 1.71–2.07 | 1.25 | 1.04–1.50 | 1.00 | 0.90–1.12 | 0.74 | 0.66–0.83 | 1.16 | 1.04–1.30 | 1.09 | 0.98–1.22 |
| White (ref: not white) | 1.01 | 0.85–1.19 | 1.24 | 0.91–1.68 | 0.84 | 0.71–1.00 | 0.88 | 0.72–1.07 | 0.94 | 0.78–1.13 | 0.87 | 0.72–1.06 |
| Black (ref: not black) | 1.37 | 1.13–1.66 | 1.41 | 0.99–2.02 | 1.10 | 0.89–1.34 | 1.21 | 0.96–1.52 | 1.55 | 1.26–1.92 | 1.34 | 1.07–1.66 |
| Hispanic (ref: not Hispanic) | 1.46 | 1.29–1.66 | 1.41 | 1.13–1.75 | 1.68 | 1.47–1.92 | 0.82 | 0.69–0.97 | 0.99 | 0.85–1.15 | 0.74 | 0.63–0.88 |
| BMI | 1.07 | 1.06–1.08 | 0.97 | 0.96–0.98 | 1.03 | 1.02–1.03 | 1.04 | 1.03–1.04 | 1.04 | 1.03–1.04 | 1.05 | 1.04–1.06 |
| Cognitive impairment (ref: not impaired) | 1.60 | 1.32–1.94 | 2.67 | 2.07–3.43 | 1.77 | 1.45–2.16 | 1.11 | 0.88–1.41 | 1.96 | 1.62–2.37 | 1.59 | 1.29–1.95 |
| Morbiditya | 1.12 | 1.09–1.15 | 1.27 | 1.21–1.33 | 1.13 | 1.09–1.16 | 1.15 | 1.12–1.19 | 1.23 | 1.20–1.27 | 1.18 | 1.14–1.21 |
| Depression score | 1.21 | 1.19–1.23 | 1.18 | 1.15–1.22 | 1.24 | 1.21–1.26 | 1.20 | 1.18–1.22 | 1.20 | 1.18–1.22 | 1.22 | 1.19–1.24 |
| Current smoker (ref: nonsmoker) | 0.90 | 0.80–1.01 | 0.88 | 0.71–1.09 | 0.92 | 0.81–1.05 | 0.93 | 0.80–1.07 | 1.22 | 1.08–1.39 | 1.18 | 1.04–1.34 |
| Previous smoker (ref: nonsmoker) | 1.05 | 0.097–1.14 | 0.90 | 0.78–1.05 | 1.06 | 0.97–1.16 | 0.95 | 0.86–1.04 | 0.96 | 0.88–1.06 | 1.13 | 1.03–1.24 |
| Self-rated health status (ref: excellent) | ||||||||||||
| Very good | 1.65 | 1.25–2.19 | 1.56 | 0.84–2.90 | 1.47 | 0.98–2.20 | 1.61 | 1.13–2.30 | 1.23 | 0.87–1.74 | 1.31 | 0.88–1.95 |
| Good | 3.28 | 2.51–4.29 | 3.21 | 1.77–5.81 | 4.61 | 3.15–6.76 | 3.35 | 2.37–4.72 | 2.53 | 1.82–3.51 | 3.50 | 2.40–5.09 |
| Fair | 6.14 | 4.68–8.05 | 7.32 | 4.04–13.26 | 9.81 | 6.69–14.38 | 6.55 | 4.64–9.26 | 5.90 | 4.24–8.20 | 7.54 | 5.18–10.99 |
| Poor | 13.66 | 10.32–18.08 | 13.67 | 7.46–25.02 | 20.13 | 13.64–29.73 | 11.83 | 8.29–16.88 | 13.40 | 9.57–18.76 | 19.45 | 13.28–28.50 |
For every one condition.
The model includes data for the 17,747 participants included.
ref, reference.
Table 3 presents the results for the time-varying associations between disaggregated ADL limitations and time to mortality. The presence of a bathing, walking, toileting, eating, or dressing ADL limitation was associated with a 1.47 (95% CI, 1.27–1.72), 1.43 (95% CI, 1.23–1.68), 1.32 (95% CI, 1.12–1.56), 1.30 (95% CI, 1.05–1.63), and 1.19 (95% CI, 1.03–1.37) higher hazard ratio for mortality, respectively. However, a transferring ADL limitation was not significantly associated with time to mortality (hazard ratio, 1.16; 95% CI, 0.97–1.39).
TABLE 3.
The time-varying associations between disaggregated ADL limitations and mortality.
| Dressing |
Eating |
Transferring |
Toileting |
Bathing |
Walking |
|||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HR | CI | HR | CI | HR | CI | HR | CI | HR | CI | HR | CI | |
|
| ||||||||||||
| ADL limitation (ref: no ADL limitation) | 1.19 | 1.03–1.37 | 1.30 | 1.05–1.63 | 1.16 | 0.97–1.39 | 1.32 | 1.12–1.56 | 1.47 | 1.27–1.72 | 1.43 | 1.23–1.68 |
| Handgrip strength (5-kg decrease) | 1.06 | 1.03–1.09 | 1.06 | 1.03–1.09 | 1.06 | 1.03–1.09 | 1.06 | 1.03–1.09 | 1.06 | 1.02–1.09 | 1.06 | 1.03–1.09 |
| Age | 1.05 | 1.04–1.06 | 1.05 | 1.04–1.06 | 1.05 | 1.04–1.06 | 1.05 | 1.04–1.06 | 1.05 | 1.04–1.06 | 1.05 | 1.04–1.06 |
| Male (ref: female) | 1.63 | 1.42–1.86 | 1.64 | 1.43–2.88 | 1.64 | 1.43–1.88 | 1.65 | 1.44–1.89 | 1.62 | 1.42–1.86 | 1.65 | 1.44–1.89 |
| White (ref: not white) | 1.55 | 1.11–2.16 | 1.53 | 1.10–2.13 | 1.54 | 1.11–2.14 | 1.55 | 1.10–2.14 | 1.55 | 1.11–2.17 | 1.57 | 1.12–2.18 |
| Black (ref: not black) | 1.65 | 1.14–2.37 | 1.64 | 1.14–2.36 | 1.65 | 1.15–2.37 | 1.63 | 1.13–2.35 | 1.62 | 1.12–2.33 | 1.66 | 1.15–2.39 |
| Hispanic (ref: not Hispanic) | 0.80 | 0.65–1.00 | 0.81 | 0.65–1.00 | 0.80 | 0.65–1.00 | 0.81 | 0.65–1.01 | 0.81 | 0.65–1.01 | 0.82 | 0.65–1.02 |
| BMI | 0.96 | 0.94–0.97 | 0.96 | 0.95–0.97 | 0.96 | 0.95–0.97 | 0.96 | 0.95–0.97 | 0.96 | 0.95–0.97 | 0.96 | 0.94–0.97 |
| Cognitive impairment (ref: not impaired) | 1.21 | 0.90–1.63 | 1.20 | 0.90–1.62 | 1.20 | 0.89–1.62 | 1.21 | 0.90–1.62 | 1.16 | 0.86–1.56 | 1.19 | 0.88–1.60 |
| Morbiditya | 1.10 | 1.06–1.15 | 1.10 | 1.05–1.14 | 1.10 | 1.06–1.15 | 1.10 | 1.06–1.15 | 1.09 | 1.05–1.14 | 1.10 | 1.05–1.14 |
| Depression score | 1.02 | 0.99–1.04 | 1.02 | 0.99–1.04 | 1.02 | 0.99–1.05 | 1.02 | 0.99–1.04 | 1.01 | 0.98–1.04 | 1.01 | 0.98–1.04 |
| Current smoker (ref: nonsmoker) | 1.41 | 1.18–1.67 | 1.41 | 1.19–1.67 | 1.40 | 1.18–1.66 | 1.40 | 1.18–1.66 | 1.41 | 1.19–1.68 | 1.41 | 1.18–1.67 |
| Previous smoker (ref: nonsmoker | 1.44 | 1.28–1.63 | 1.44 | 1.28–1.63 | 1.44 | 1.27–1.63 | 1.44 | 1.28–1.63 | 1.43 | 1.27–1.62 | 1.44 | 1.27–1.63 |
| Self-rated health status (ref: excellent) | ||||||||||||
| Very good | 1.28 | 0.95–1.75 | 1.26 | 0.93–1.71 | 1.28 | 0.95–1.71 | 1.26 | 0.93–1.71 | 1.29 | 0.96–1.76 | 1.26 | 0.93–1.70 |
| Good | 1.93 | 1.43–2.61 | 1.92 | 1.43–2.59 | 1.93 | 1.43–2.58 | 1.91 | 1.42–2.57 | 1.94 | 1.44–2.62 | 1.91 | 1.42–2.58 |
| Fair | 3.78 | 2.78–5.14 | 3.83 | 2.82–5.20 | 3.82 | 2.81–5.19 | 3.80 | 2.80–5.16 | 3.78 | 2.79–5.14 | 3.80 | 2.80–5.15 |
| Poor | 6.86 | 4.93–9.55 | 7.01 | 5.04–9.74 | 6.98 | 5.10–9.69 | 6.87 | 4.94–9.55 | 6.67 | 4.80–9.27 | 6.60 | 4.74–9.20 |
For every one condition.
The model includes data for the 17,747 participants included.
DISCUSSION
The principal results of this investigation demonstrate that handgrip strength was robustly associated with individual ADL outcomes, and most ADL limitations were also strongly associated with time to mortality. Specifically, decreased handgrip strength was associated with increased odds of developing an ADL disability in each function, with the greatest odds for eating, walking, and bathing tasks. Disabilities in nearly all ADL functions were associated with a higher hazard for mortality, although the highest hazard ratios were seen in those with a bathing, walking, and toileting limitation. These results provide insights into the disabling process by revealing how handgrip strength differentially influences each ADL outcome, and in turn, how the presence of individual ADL limitations impacts time to mortality.
The variation across OR for the association between handgrip strength and each ADL outcome suggests that muscle weakness may contribute to the development of an ADL limitation for some functions more than others. For example, individuals who had an eating limitation may have been frail from being underweight and malnourished (22). This likely explains why increased BMI was also associated with decreased odds for developing an eating ADL limitation. Similarly, our results suggesting decreased handgrip strength increased the odds of a walking ADL limitation may be explained by how muscle weakness contributes to slow walking speed, which subsequently increases the risk of future movement disabilities including walking (23). Chronic pain is also a known contributor to muscle weakness and the presence of pain may create instability during walking, thereby increasing the risk for walking limitations (24).
Although individual ADL are typically aggregated when assessing functional status in older adults, it is important to acknowledge the subtasks and spectrum of physical and psychological problems that may contribute to each functional disability. For example, while our results suggests decreased handgrip strength was associated with increased odds of developing an ADL limitation across tasks, other health domains such as cognition, body function and structure, and environmental factors may also contribute to the development of functional disabilities (25). Bathing limitations could be driven by sensory impairments or physical declines related to dementia; whereas, poor mobility and the presence of multiple ADL limitations may factor into the development of a toileting limitation (26,27).
Our results are compatible with the results of another investigation suggesting depressive symptoms are connected to transferring and dressing limitations (28). Therefore, although decreased handgrip strength may increase a person’s odds for developing a functional limitation, a variety of health factors each differentially contribute to the development of specific ADL limitations and several factors may be linked.
Given that different health factors influence the development of an ADL limitation, the presence of a limitation in specific ADL functions may also differentially influence health outcomes such as time to mortality. Previous research has demonstrated that the presence of an ADL disability increases the risk for premature mortality in older adults (6); however, we found that individual ADL limitations were differentially associated with an increased risk for early mortality, except for transferring impairments. Our results were similar to an investigation of older adults from the HRS, wherein a bathing ADL limitation was associated with nursing home placement (8). The intense psychological distress from needing help with bathing, along with the high levels of assistance required from others to bathe, may explain why having a bathing ADL limitation represents the later stages of the disabling process (29). Our results were also similar to another investigation that revealed older adults who had difficulty or were unable to walk a quarter-mile had greater odds for mortality compared to those that had no difficulty with walking (30). Using holistic framework that considers mobility determinants and life-space locations may improve assessments and treatments of mobility for older adults (31).
Having difficulty or being unable to eat may accelerate the disabling process via malnutrition in those with an eating ADL limitation. The increased risk for urinary tract infections in older adults that experience incontinence may explain why the hazard for mortality was higher for those with a toileting ADL disability, even after controlling for cognitive impairments (32). Dressing oneself involves a series of motor (e.g., reach, grab, lift clothing) and process skills (e.g., locate clothes and dress in a proper order), and losing these skills may exacerbate time to mortality if a person has a dressing ADL disability (33). The nonsignificant hazard ratio for mortality seen in persons with transferring ADL limitations may be explained by a person’s ability to regain function after rehabilitation when a health event occurred that created difficulties in the transferring ADL function (34).
Interventions that aim to preserve function and prolong survival in older adults should acknowledge how different health factors influence individual ADL outcomes, and how the presence of certain ADL limitations accelerates time to mortality. Targeted, multidimensional intervention strategies that incorporate muscle strengthening physical activities, healthy dietary counseling, and social engagement may help retain strength and function in older adults, especially if healthy behaviors are practiced earlier in life (35,36). However, particular emphasis should be placed on the role of physical activity in reducing the risk of functional disabilities and all-cause mortality in older adults, especially because increased physical activity participation improves muscle strength, cognition, mental and physical health, and life expectancy (37–39).
Further, we suggest that the traditional binary definition of a functional limitation should be modified. More research is needed for advancing evaluations of function including refining what tasks are assessed, incorporating changes in functional capacity with repeated measures over time, identifying how functions interact, and developing an enhanced outcome for better capturing functional status that is clinically meaningful for health practitioners and their patients. Such information will help inform practitioners about what factors put their patients at greater risk for impairments in individual functions, and how certain functional outcomes may be linked with other clinically relevant health outcomes that potentiate the disabling process.
Some limitations should be noted. Self-reported physical activity participation was not included in our models because the questionnaire used to assess physical activity in the HRS has not been well validated. Although treating morbidity as the summation of diseases reported by participants at each wave allowed us to control for several diseases and avoid issues related to multicollinearity across diseases, doing so did not allow us to identify how specific diseases were associated with the development of an ADL limitation and time to mortality. Handgrip strength data for the 2016 wave has not yet been released and therefore could not be concatenated with the 2014 wave. Additionally, we did not have complete data for all participants at each wave. Participants were followed starting at the 2006 wave of the HRS, but if they were missing covariate information for a given wave, they were excluded from the models for that wave. This was taken into account when calculating the time variable for our Cox models. Detailed missing covariate information is presented in Appendix 1 (Supplemental Digital Content 1; The Amount of Missing Observations at Each Wave, http://links.lww.com/MSS/B312).
Despite these limitations, our study provides insights into the disabling process by providing time-varying information for how handgrip strength is associated with each ADL task, and how the presence of a specific ADL limitation is associated with time to mortality. This is important considering that the number of older adults in the United States is projected to increase by approximately 112% by the year 2060 (40). Future investigations should continue examining the time-varying associations between handgrip strength, function, and clinically relevant health outcomes in older adults. The use of more causal study designs (e.g., matching) will also help with unraveling the disabling process.
In conclusion, decreased handgrip strength was associated with increased odds of developing an ADL disability for each task, and in turn, most ADL limitations were associated with a higher hazard for mortality in older adults. These findings provide insights into the disabling process by identifying how decreased handgrip strength influences each ADL, and how the presence of certain ADL limitations increases the risk for mortality. Our results suggest that refining and developing interventions that aim to preserve function and delay mortality for older adults take a multidisciplinary approach, in that several aspects of health (e.g., physical, mental, cognitive) are addressed. Such intervention approaches may help older adults retain function and live longer lives.
Supplementary Material
Acknowledgments
R. P. M. was supported in part by an Advanced Rehabilitation Research Training award from the National Institute on Disability and Rehabilitation Research (90AR5020–0200). M. D. P. is supported in part by the Claude D. Pepper Center (AG024824) and Michigan Institute for Clinical and Health Research (UL1TR002240).
Footnotes
The authors declare no conflicts of interest and the results of the present study do not constitute endorsement by ACSM. The results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation.
REFERENCES
- 1.Newman AB, Lee JS, Visser M, et al. Weight change and the conservation of lean mass in old age: the Health, Aging and Body Composition Study. Am J Clin Nutr. 2005;82(4):872–8. [DOI] [PubMed] [Google Scholar]
- 2.Sayer AA, Kirkwood T. Grip strength and mortality: a biomarker of ageing? Lancet. 2015;386(9990):226–7. [DOI] [PubMed] [Google Scholar]
- 3.Leong DP, Teo KK, Rangarajan S, et al. Prognostic value of grip strength: findings from the Prospective Urban Rural Epidemiology (PURE) study. Lancet. 2015;386(9990):266–73. [DOI] [PubMed] [Google Scholar]
- 4.McGrath RP, Vincent BM, Snih SA, et al. The association between handgrip strength and diabetes on activities of daily living disability in older mexican americans. J Aging Health 2017:0898264317715544. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Syddall H, Cooper C, Martin F, Briggs R, Aihie Sayer A. Is grip strength a useful single marker of frailty? Age Ageing. 2003;32(6):650–6. [DOI] [PubMed] [Google Scholar]
- 6.Majer IM, Nusselder WJ, Mackenbach JP, Klijs B, van Baal PH. Mortality risk associated with disability: a population-based record linkage study. Am J Public Health. 2011;101(12):e9–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Martin LG, Freedman VA, Schoeni RF, Andreski PM. Trends in disability and related chronic conditions among people ages fifty to sixty-four. Health Aff (Millwood). 2010;29(4):725–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Fong JH, Mitchell OS, Koh BS. Disaggregating activities of daily living limitations for predicting nursing home admission. Health Serv Res. 2015;50(2):560–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gure TR, Langa KM, Fisher GG, Piette JD, Plassman BL. Functional limitations in older adults who have cognitive impairment without dementia. J Geriatr Psychiatry Neurol. 2013;26(2):78–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hardy SE, Gill TM. Recovery from disability among community-dwelling older persons. JAMA. 2004;291(13):1596–602. [DOI] [PubMed] [Google Scholar]
- 11.Dodds RM, Syddall HE, Cooper R, et al. Grip strength across the life course: normative data from twelve British studies. PLoS One. 2014;9(12):e113637. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Millan-Calenti JC, Tubío J, Pita-Fernández S, et al. Prevalence of functional disability in activities of daily living (ADL), instrumental activities of daily living (IADL) and associated factors, as predictors of morbidity and mortality. Arch Gerontol Geriatr. 2010;50(3):306–10. [DOI] [PubMed] [Google Scholar]
- 13.McGrath R, Robinson-Lane SG, Peterson MD, Bailey RR, Vincent BM. Muscle strength and functional limitations: preserving function in older Mexican Americans. J Am Med Dir Assoc. 2018;S1525–8610(17):30696–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sonnega A, Faul JD, Ofstedal MB, Langa KM, Phillips JW, Weir DR. Cohort profile: the Health and Retirement Study (HRS). Int J Epidemiol. 2014;43(2):576–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Health and Retirement Study Data Book [Internet]. Ann Arbor, (MI): Health and Retirement Study; [cited 2018 Feb 15]. Available from: https://hrs.isr.umich.edu/about/data-book. [Google Scholar]
- 16.Crimmins E, Guyer H, Langa K, Ofstedal M, Wallace R, Weir D. Documentation of physical measures, anthropometrics and blood pressure in the Health and Retirement Study. HRS Documentation Report DR-011. 2008;14:47–59. [Google Scholar]
- 17.Plassman BL, Newman TT, Welsh KA, Helms M, Breitner JC. Application in epidemiological and longitudinal studies. Cogn Behav Neurol. 1994;7(3):235–41. [Google Scholar]
- 18.Crimmins EM, Kim JK, Langa KM, Weir DR. Assessment of cognition using surveys and neuropsychological assessment: the Health and Retirement Study and the Aging, Demographics, and Memory Study. J Gerontol B Psychol Sci Soc Sci. 2011;66(1 Suppl):i162–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Langa KM, Larson EB, Karlawish JH, et al. Trends in the prevalence and mortality of cognitive impairment in the United States: is there evidence of a compression of cognitive morbidity? Alzheimers Dement. 2008;4(2):134–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Turvey CL, Wallace RB, Herzog R. A revised CES-D measure of depressive symptoms and a DSM-based measure of major depressive episodes in the elderly. Int Psychogeriatr. 1999;11(2):139–48. [DOI] [PubMed] [Google Scholar]
- 21.Rosanbalm S Getting Sankey with Bar Charts. PharmaSUG [Internet]. 2015. [cited 2018 Feb 15]. Available from: https://www.pharmasug.org/proceedings/2015/DV/PharmaSUG-2015-DV07.pdf.
- 22.Matos L, Tavares M, Amaral T. Handgrip strength as a hospital admission nutritional risk screening method. Eur J Clin Nutr. 2007;61(9):1128–35. [DOI] [PubMed] [Google Scholar]
- 23.Shimada H, Makizako H, Doi T, Tsutsumimoto K, Suzuki T. Incidence of disability in frail older persons with or without slow walking speed. J Am Med Dir Assoc. 2015;16(8):690–6. [DOI] [PubMed] [Google Scholar]
- 24.Eggermont LH, Leveille SG, Shi L, et al. Pain characteristics associated with the onset of disability in older adults: the maintenance of balance, independent living, intellect, and zest in the Elderly Boston Study. J Am Geriatr Soc. 2014;62(6):1007–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.den Ouden ME, Schuurmans MJ, Mueller-Schotte S, Brand JS, van der Schouw YT. Domains contributing to disability in activities of daily living. J Am Med Dir Assoc. 2013;14(1):18–24. [DOI] [PubMed] [Google Scholar]
- 26.Talley KM, Wyman JF, Bronas UG, Olson-Kellogg BJ, McCarthy TC, Zhao H. Factors associated with toileting disability in older adults without dementia living in residential care facilities. Nurs Res. 2014;63(2):94–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Cohen-Mansfield J, Parpura-Gill A. Bathing: a framework for intervention focusing on psychosocial, architectural and human factors considerations. Arch Gerontol Geriatr. 2007;45(2):121–35. [DOI] [PubMed] [Google Scholar]
- 28.Boström G, Conradsson M, Rosendahl E, Nordström P, Gustafson Y, Littbrand H. Functional capacity and dependency in transfer and dressing are associated with depressive symptoms in older people. Clin Interv Aging. 2014;9:249–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Rader J, Barrick AL, Hoeffer B, et al. The bathing of older adults with dementia: easing the unnecessarily unpleasant aspects of assisted bathing. Am J Nurs. 2006;106(4):40–8. [DOI] [PubMed] [Google Scholar]
- 30.Hardy SE, Kang Y, Studenski SA, Degenholtz HB. Ability to walk 1/4 mile predicts subsequent disability, mortality, and health care costs. J Gen Intern Med. 2011;26(2):130–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Webber SC, Porter MM, Menec VH. Mobility in older adults: a comprehensive framework. Gerontologist. 2010;50(4):443–50. [DOI] [PubMed] [Google Scholar]
- 32.Tal S, Guller V, Levi S, et al. Profile and prognosis of febrile elderly patients with bacteremic urinary tract infection. J Infect. 2005;50(4):296–305. [DOI] [PubMed] [Google Scholar]
- 33.Liu CJ, Jones LY, Formyduval AR, Clark DO. Task-oriented exercise to reduce activities of daily living disability in vulnerable older adults: a feasibility study of the 3-step workout for life. J Aging Phys Act. 2016;24(3):384–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Alexander NB, Grunawalt JC, Scott C, Augustine J. Bed mobility task performance in older adults. J Rehabil Res Dev. 2000;37(5):633–8. [PubMed] [Google Scholar]
- 35.Mendes de Leon CF, Rajan KB. Psychosocial influences in onset and progression of late life disability. J Gerontol B Psychol Sci Soc Sci. 2014;69(2):287–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Frimel TN, Sinacore DR, Villareal DT. Exercise attenuates the weight-loss-induced reduction in muscle mass in frail obese older adults. Med Sci Sports Exerc. 2008;40(7):1213–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Paulo TR, Tribess S, Sasaki JE, et al. A cross-sectional study of the relationship of physical activity with depression and cognitive deficit in older adults. J Aging Phys Act. 2016;24(2):311–21. [DOI] [PubMed] [Google Scholar]
- 38.Reimers CD, Knapp G, Reimers AK. Does physical activity increase life expectancy? A review of the literature. J Aging Res. 2012;2012:243958. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Peterson MD, Rhea MR, Sen A, Gordon PM. Resistance exercise for muscular strength in older adults: a meta-analysis. Ageing Res Rev. 2010;9(3):226–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Colby SL, Ortman JM. Projections of the size and composition of the US population: 2014 to 2060: population estimates and projections. US Census Bureau;2015:1–13. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.