Abstract
Using new weakness cut-points individually may help estimate time to mortality, but their collective use could improve value. We sought to determine the associations of 1) each absolute and body size normalized cut-point, and 2) collective weakness on time to mortality in Americans. The analytic sample included 14,178 subjects aged ≥50-years from the 2006-2018 waves of the Health and Retirement Study. Date of death was confirmed from the National Death Index. Handgrip dynamometry measured handgrip strength (HGS). Men were categorized as weak if their HGS was <35.5-kg (absolute), <0.45 kg/kg (body mass normalized), or <1.05 kg/kg/m2 (body mass index normalized). Women were classified as weak if their HGS was <20.0-kg, <0.337 kg/kg, or <0.79 kg/kg/m2. Collective weakness categorized persons as below 1, 2, or all 3 cut-points. Cox proportional hazard regression models were used for analyses. Subject values below each absolute and normalized cut-point for the three weakness parameters had a higher hazard ratio for early all-cause mortality: 1.45 (95% Confidence Interval (CI): 1.36-1.55) for absolute weakness, 1.39 (CI: 1.30-1.49) for body mass index normalized weakness, and 1.33 (CI: 1.24-1.43) for body mass normalized weakness. Those below 1, 2, or all 3-weakness cut-points had a 1.37 (CI: 1.26-1.50), 1.47 (CI: 1.35-1.61), and 1.69 (CI: 1.55-1.84) higher hazard for mortality, respectively. Weakness determined by a composite measure of absolute and body size adjusted strength capacity provides robust prediction of time to mortality, thus potentially informing sports medicine and health practitioner discussions about the importance of muscle strength during aging.
Keywords: Aging, Geriatric Assessment, Longevity, Muscle Strength, Muscle Strength Dynamometer
INTRODUCTION
Weakness, as measured by handgrip strength (HGS), represents the onset of physical disablement (3). The presence of weakness is widespread and associated with several age-related health conditions such as cognitive impairment, chronic disease, and functional disability (5,25,27). As such, it is not surprising that low HGS is associated with early cause-specific (e.g., cardiovascular) and all-cause mortality (7,16,21). Further, low HGS could be a stronger predictor of all-cause and cardiovascular mortality than systolic blood pressure such that factors including autonomic imbalance, endothelial dysfunction, and arterial stiffness may mediate the association between strength capacity and cardiovascular events (20). Given that weakness is robustly associated with disability, disease, and death, strength capacity is a considered a critical health-related biomarker that warrants routine assessment by healthcare providers and sports medicine practitioners (27,33).
Despite the wide-spread use of HGS in exercise, clinical, and research settings, there is considerable heterogeneity in HGS procedures including the existence of several cut-points for determining weakness (24). Such inconsistencies in weakness cut-points pose threats to internal validity, generates difficulty in making comparisons across studies examining weakness, and may lead to misclassification of weakness (i.e., false positive), which in turn, could misinform referral to intervention (e.g., strength programs). Moreover, while there are differences in the specific thresholds for absolute peak force between weakness cut-points, there are also cut-points that normalize absolute strength to body size characteristics such as body mass and body mass index (BMI) because body size may influence strength capacity. Therefore, having more consistent weakness cut-points may help to standardize HGS methodologies.
The Sarcopenia Definition and Outcomes Consortium (SDOC) presented new absolute and body size normalized (body mass and BMI) weakness cut-points (6,22). The cut-points from the SDOC are unique such that they provide more uniform thresholds for operationalizing with absolute strength capacity, and HGS normalized to a body size metric, which in turn, may help to reduce HGS procedural heterogeneity. Although these cut-points provide flexibility in accounting for absolute or body size normalized strength capacity, the prevalence of Americans below one or more of these cut-points is not equal (29), and the reasons for this disproportion are relatively unknown. Opportunities may also exist for collectively utilizing these absolute and body size normalized cut-points, which may subsequently improve weakness operationalization and mitigate misclassification of weakness. However, the role of these weakness cut-points on time to mortality is unclear. Understanding how these cut-points are associated with time to mortality may help to present health and sports medicine practitioners with insights on how weakness is defined. We sought to control for relevant covariates in quantifying the associations of 1) individual absolute and body size normalized weakness cut-points, and 2) collective weakness on time to mortality in Americans.
METHODS
Experimental Approach to the Problem
We performed a secondary analysis of data from the 2006-2018 waves of the RAND Health and Retirement Study (HRS). The HRS utilizes a longitudinal-panel design for observing health-related factors in Americans as they age (34). Those that participate in the HRS are interviewed biennially and followed until death. New cohorts of HRS subjects are added at a pre-specified time for helping the HRS maintain a national sample of Americans (13). HRS interview response rates have consistently been >80%. Beginning in the 2006 wave, the HRS started conducting enhanced face-to-face interviews that included physical measures such as HGS (13). These face-to-face interviews occurred on a random half sample of HRS subjects and alternated in their completion at each wave, while the other half sample completed the core interviews (32). Additional details about the HRS are available elsewhere (13).
Subjects
The sample included 14,329 subjects aged ≥50-years that completed physical and survey assessments during this period. Those with missing covariates were excluded (n=151). Table 1 shows the baseline descriptive characteristics of the 14,178 subjects. Overall, subjects were aged 68.8±10.8 years and were mostly female (55.2%). Of those included, 6,514 (45.9%) were not-weak, 2,607 (18.4%) were below 1 weakness cut-point, 2,626 (15.5%) were below 2 weakness cut-points, and 2,431 (17.1%) were below all 3-weakness cut-points. All HRS subjects provided written informed consent before study entry and the University of Michigan’s Behavioral Sciences Committee Institutional Review Board approved study protocols. This secondary analysis was considered exempt by the North Dakota State University Institutional Review Board.
Table 1.
Baseline Descriptive Characteristics of the Subjects.
| Overall (n=14,178) | 0 Weakness Categories (n=6,514) | 1 Weakness Category (n=2,607) | 2 Weakness Categories (n=2,626) | 3 Weakness Categories (n=2,431) | |
|---|---|---|---|---|---|
|
|
|
||||
| Age (years) | 68.8±10.8 | 65.6±10.0 | 68.6±10.4 | 70.7±10.5 | 75.6±10.3 |
| Age Group (n (%)) | |||||
| 50-64 years | 3,913 (27.6) | 2,358 (36.2) | 707 (27.1) | 572 (21.8) | 276 (11.4) |
| 65-74 years | 5,824 (41.1) | 2,925 (44.9) | 1,117 (42.8) | 1,041 (39.6) | 741 (30.5) |
| 75-84 years | 3,390 (23.9) | 1,057 (16.2) | 638 (24.5) | 777 (29.6) | 918 (37.8) |
| ≥85 years | 1,051 (7.4) | 174 (2.7) | 145 (5.6) | 236 (9.0) | 496 (20.4) |
| Body Mass Index (kg/m2) | 28.2±5.9 | 26.2±4.0 | 29.4±5.8 | 31.4±7.6 | 28.9±6.2 |
| Obesity (n (%)) | 4,658 (32.8) | 1,158 (17.7) | 1,256 (48.1) | 1,337 (50.9) | 907 (37.3) |
| Female (n (%)) | 7,837 (55.2) | 3,500 (53.7) | 1,453 (55.7) | 1,603 (61.0) | 1,580 (64.9) |
| Race and Ethnicity (n (%)) | |||||
| Hispanic Black | 54 (0.4) | 26 (0.4) | 11 (0.4) | 11 (0.4) | 6 (0.2) |
| Hispanic White | 1,027 (7.2) | 429 (6.6) | 165 (6.3) | 210 (8.0) | 223 (9.2) |
| Non-Hispanic Black | 2,594 (18.3) | 1,178 (18.1) | 505 (19.4) | 576 (21.9) | 335 (18.8) |
| Non-Hispanic White | 9,456 (66.7) | 4,344 (66.7) | 1,759 (67.5) | 1,653 (62.9) | 1,700 (69.9) |
| Other | 1,047 (7.4) | 537 (8.2) | 167 (6.4) | 176 (6.7) | 167 (6.9) |
| Multimorbidity (n (%)) | 8,841 (62.3) | 3,192 (49.0) | 1,716 (65.8) | 1,978 (75.3) | 1,955 (80.4) |
| Cigarette Smoking Status (n (%)) | |||||
| Current Smoker | 1,994 (14.1) | 1,165 (17.9) | 346 (13.3) | 305 (11.6) | 178 (7.3) |
| Previous Smoker | 6,122 (43.2) | 2,548 (39.1) | 1,232 (47.2) | 1,177 (44.8) | 1,105 (45.5) |
| Never Smoked | 6,062 (42.8) | 2,801 (43.0) | 1,029 (39.5) | 1,144 (43.6) | 1,148 (47.2) |
| Self-Rated Health (n (%)) | |||||
| Excellent | 1,340 (9.5) | 888 (13.6) | 212 (8.1) | 140 (5.3) | 100 (4.1) |
| Very Good | 4,000 (28.2) | 2,218 (34.1) | 716 (27.5) | 576 (21.9) | 490 (20.2) |
| Good | 4,592 (32.4) | 2,056 (31.6) | 880 (33.8) | 935 (35.6) | 721 (29.7) |
| Fair | 3,154 (22.3) | 1,076 (16.5) | 613 (23.5) | 703 (26.8) | 762 (31.3) |
| Poor | 1,092 (7.7) | 276 (4.2) | 186 (7.1) | 272 (10.4) | 358 (14.7) |
| Depressed (n (%)) | 1,980 (13.9) | 768 (11.8) | 323 (12.3) | 418 (15.9) | 471 (19.3) |
| MVPA Participation (n (%)) | 8,029 (56.6) | 4,338 (66.6) | 1,446 (55.4) | 1,218 (46.3) | 1,027 (42.2) |
| ADL Limitation (n (%)) | 2,459 (17.3) | 539 (8.3) | 404 (15.5) | 640 (24.3) | 876 (36.0) |
| Cognitive Impairment (n (%)) | 327 (2.3) | 68 (1.0) | 50 (1.9) | 73 (2.7) | 136 (5.5) |
| High School Graduate or Above (n (%)) | 11,113 (78.3) | 5,370 (82.4) | 2,077 (79.6) | 1,947 (74.1) | 1,719 (70.7) |
Note: ADL=activities of daily living, MVPA=moderate-to-vigorous physical activity.
Procedures
Mortality
Date of death was verified by the HRS through linkage to the National Death Index. Death verification was also confirmed with postmortem interviews with a surviving family member or other informant. In effort to maintain subject anonymity and deidentification, the RAND HRS provides the month and year wherein a subject death occurred. Mortality validation for the HRS has demonstrated that the National Death Index and postmortem interviews appropriately ascertains 99% of subject deaths (39).
Weakness
A Smedley spring-type handgrip dynamometer (Scandidact; Odder, Denmark) was used to measure HGS. Interclass correlations for the Smedley dynamometer suggest excellent reliability (4). Interviewers provided subjects with instructions for how to perform the HGS assessment. The dynamometer was fit to the hand size of each subject, and a practice trial was allowed for familiarization. Beginning on the self-reported non-dominant hand, subjects positioned their arm at the side and flexed their elbow at 90° while grasping the dynamometer. Subjects then squeezed the dynamometer with maximal effort, exhaling while squeezing, twice on each hand, alternating between hands. Persons unable to stand or properly position their arm during HGS testing could sit and place their arm on a supporting object. Those reporting a surgical procedure, swelling, inflammation, severe pain, or an injury to both hands 6-months prior to the face-to-face interview did not complete HGS assessments. More details about the HRS protocols in the HRS are available elsewhere (8).
The trial with the highest recorded HGS regardless of hand was included in the analyses. Men with HGS <35.5-kg (absolute), <0.45 kg/kg (normalized to body mass), or <1.05 kg/kg/m2 (normalized to BMI) were considered weak. Likewise, women with HGS <20.0-kg, <0.337 kg/kg, or <0.79 kg/kg/m2 were also classified as weak (6,22,29). These cut-points were used collectively, whereby subjects were categorized as being below 1, 2, or all 3 cut-points.
Covariates
Respondents told interviewers their age, sex (male, female), race (Black, Other, White), ethnicity (Hispanic, non-Hispanic), educational achievement (not a high school graduate, high school graduate and equivalent or above), height, and body mass. Persons with a BMI ≥30 kg/m2 were considered obese. Subjects reported if they were currently smoking cigarettes, or if they have ever smoked more than 100 cigarettes in their lifetime (previous smoker). Subjects were also asked if a healthcare provider had diagnosed them with hypertension, a heart condition, chronic lung disease, arthritis or rheumatism, emotional or psychiatric problems, stroke, diabetes, and cancer. Those affirming the presence of at least 2 conditions were considered as having multimorbidity. A single-item indicator of perceived health was collected, and respondents rated their health as either “excellent”, “very good”, “good”, “fair”, or “poor”.
Cognitive function was assessed with the modified version of the Telephone Interview of Cognitive Status (TICS). The modified TICS was created from the Mini-Mental State Examination for population-based studies such as the HRS (31). Assessments included immediate and delayed word recall from a list of 10 words, serial sevens subtraction test beginning with the number 100, counting backward for 10 consecutive numbers starting with the number 10 at maximal speed, object naming, date naming, and identifying the current president and vice-president of the United States. A 35-point composite scale was used and persons with scores <11 were considered as having a cognitive impairment (19). The 8-item Center for the Epidemiologic Studies Depression scale examined depressive symptoms (36). Subjects indicated if they experienced any positive or negative emotions during the week before the interview. Scores ranged from 0-8 with higher scores suggesting more depressive symptoms. Those with scores ≥3 were classified as depressed (36).
Subjects reported their ability to dress, eat, transfer in-or-out of bed, toilet, bathe, and walk across a small room. Persons suggesting difficulty or an inability in completing any basic self-care task were considered as having an activities of daily living (ADL) limitation. Persons reporting engagement in moderate-to-vigorous physical activity (MVPA) at least “once a week” were considered as participating in MVPA (12).
Statistical Analysis
All analyses were performed with SAS 9.4 software (SAS Institute; Cary, NC). Distinct Kaplan-Meier estimators created survival curves and quantified median time to mortality after study entry using 1) individual absolute and body size normalized weakness cut-points, and 2) the collective weakness groups as the strata. Separate Cox proportional hazard regression models analyzed the associations of 1) absolute weakness (reference: not-below absolute cut-point), 2) body mass normalized weakness (reference: not-below the body mass normalized weakness cut-point), 3) BMI normalized weakness (reference: not-below the BMI normalized cut-point), and 4) collective weakness (reference: below 0 cut-points) on time to mortality.
Each Cox model was adjusted for age, sex, race, ethnicity, multimorbidity, cigarette smoking status, self-rated health, obesity, depression, MVPA participation, cognitive impairment, and educational achievement. Data were left-truncated because subjects entered the HRS at different ages and had to be aged ≥50-years to be included. With observational designs such as those from the HRS, persons enter the investigation at different ages, but those who have entered the investigation after the origin point (≥50-years) have a delayed entry into the observation (left-truncation) and their peers who died before the study started are not observed (18), and therefore, baseline age was the entry variable. Because body size may skew both absolute weakness and the denominator of the normalization equations (23), supplementary analyses were conducted to evaluate the associations of the 1) individual absolute and body size normalized weakness cut-points, and 2) collective weakness on time to mortality by underweight (BMI <18.5 kg/m2) and obesity status using the same Cox modeling procedures. An alpha level of 0.05 was used for all analyses.
RESULTS
Table 2 shows the proportions of subjects below the weakness cut-points. The greatest proportion of subjects were considered weak under all 3 cut-points, when any weakness was present, for the overall sample (17.1%) and those who died (29.1%). Figure 1 displays the survival curves for the individual weakness groups on time to mortality. The median survival time (i.e., the shortest time at which the survival probability drops to 0.5 or below) was 8.0 (95% confidence interval (CI): 7.9, 8.3) years for persons below the absolute weakness cut-point, 9.5 (CI: 9.2, 9.8) for those below the BMI normalized weakness cut-point, and 10.0 (CI: 9.8, 10.2) years for those below the body mass normalized weakness cut-point. Figure 2 presents survival curves for the collective weakness groups on time to mortality. Likewise, persons below 1, 2, or all 3-weakness cut-points had a median time to mortality of 11.6 (CI: 11.2, 12.0), 10.3 (10.0, 10.7), and 8.0 (CI: 7.8, 8.5) years, respectively.
Table 2.
Distribution of Subjects Considered Weak for Each Threshold.
| n (%) |
|
|---|---|
| Overall | |
| Not-Weak | 6,514 (45.9) |
| Absolute Weakness | 491 (3.5) |
| Body Mass Normalized Weakness | 2,088 (14.7) |
| Body Mass Index Normalized Weakness | 28 (0.2) |
| Absolute Weakness & Body Mass Normalized Weakness | 1,057 (7.5) |
| Absolute Weakness & Body Mass Index Normalized Weakness | 16 (0.1) |
| Body Mass Normalized Weakness & Body Mass Index Normalized Weakness | 1,553 (11.0) |
| Absolute Weakness & Body Mass Normalized Weakness & Body Mass Index Normalized Weakness | 2,431 (17.1) |
| Future Death | |
| Not-Weak | 1,274 (28.6) |
| Absolute Weakness | 279 (6.3) |
| Body Mass Normalized Weakness | 579 (13.0) |
| Body Mass Index Normalized Weakness | 6 (0.1) |
| Absolute Weakness & Body Mass Normalized Weakness | 559 (12.5) |
| Absolute Weakness & Body Mass Index Normalized Weakness | 10 (0.2) |
| Body Mass Normalized Weakness & Body Mass Index Normalized Weakness | 454 (10.2) |
| Absolute Weakness & Body Mass Normalized Weakness & Body Mass Index Normalized Weakness | 1,297 (29.1) |
Figure 1.
Kaplan-Meier Curves for Individual Weakness Cut-Points and Time to Mortality.
Note: A) Absolute Cut-Points, B) Body Mass Normalized Cut-Points, C) Body Mass Index Normalized Cut-Points.
Figure 2.
Kaplan-Meier Curves for Collective Weakness and Time to Mortality.
The results for the associations of the individual weakness cut-points on time to mortality are presented in Table 3. Persons below each cut-point had a higher hazard for mortality: 1.45 (CI: 1.36, 1.55) for absolute weakness, 1.39 (CI: 1.30, 1.49) for BMI normalized weakness, and 1.33 (CI: 1.24, 1.43) for body mass normalized weakness. Table 4 displays the results for the associations of the collective weakness categories on time to mortality. Those below 1, 2, or all 3-weakness cut-points had 1.37 (CI: 1.26, 1.50), 1.47 (CI: 1.35, 1.61), and 1.69 (CI: 1.55, 1.84) higher hazard for mortality, respectively.
Table 3.
Results for the Associations of the Individual Weakness Categories on Time to Mortality.
| Number of People | Number of Deaths | Mean & 95% CI Follow-Up Years | Mortality Rate per 1,000 Person-Years | Hazard Ratio (95% CI) | |
|---|---|---|---|---|---|
|
|
|||||
| Absolute Categories | |||||
| Not-Weak | 10,183 (71.8%) | 2,313 (51.9%) | 7.0 (6.9, 7.1) | 32.5 | Reference |
| Weak | 3,995 (28.1%) | 2,145 (48.1%) | 6.0 (5.9, 6.1) | 88.4 | 1.45 (1.36, 1.55) |
| BMI Normalized Categories | |||||
| Not-Weak | 10,150 (71.6%) | 2,691 (60.4%) | 6.9 (6.8, 7.0) | 38.5 | Reference |
| Weak | 4,028 (28.4%) | 1,767 (39.6%) | 6.3 (6.2, 6.4) | 69.2 | 1.39 (1.30, 1.49) |
| Body Mass Normalized Categories | |||||
| Not-Weak | 7,049 (49.7%) | 1,569 (35.2%) | 6.9 (6.8, 7.0) | 31.8 | Reference |
| Weak | 7,129 (50.3%) | 2,889 (64.8%) | 6.4 (6.3, 6.5) | 62.6 | 1.33 (1.24, 1.43) |
Note: BMI=body mass index, CI=confidence interval.
Table 4.
Results for the Associations of the Collective Weakness Categories on Time to Mortality.
| Number of People | Number of Deaths | Mean & 95% CI Follow-Up Years | Mortality Rate per 1,000 Person-Years | Hazard Ratio (95% CI) | |
|---|---|---|---|---|---|
|
|
|||||
| Overall | |||||
| 0 Weakness Categories | 6,514 (46.0%) | 1,274 (28.6%) | 7.0 (6.9, 7.1) | 27.6 | Reference |
| 1 Weakness Category | 2,607 (18.4%) | 864 (19.4%) | 6.6 (6.5, 6.8) | 49.6 | 1.37 (1.26, 1.50) |
| 2 Weakness Categories | 2,626 (18.5%) | 1,023 (22.9%) | 6.5 (6.3, 6.6) | 59.7 | 1.47 (1.35, 1.61) |
| 3 Weakness Categories | 2,431 (17.1%) | 1,297 (29.1%) | 6.0 (5.9, 6.2) | 87.9 | 1.69 (1.55, 1.84) |
Note: CI=confidence interval.
Appendix 1 shows the results for the associations of the individual weakness cut-points on time to mortality by underweight and obesity status. Those categorized as underweight and below the BMI normalized cut-point had a 2.04 (CI: 1.17, 3.48) higher hazard for mortality, but no significant associations were observed for the absolute and body mass normalized cut-points. Persons with obesity below each cut-point had a higher hazard for mortality: 1.30 (CI: 1.14, 1.48) for absolute weakness, 1.51 (CI: 1.32, 1.73) for BMI normalized weakness, and 1.67 (CI: 1.38, 2.02) for body mass normalized weakness.
The results for the associations of the collective weakness cut-points on time to mortality by underweight and obesity status are in Appendix 2. Those categorized as underweight but below all 3-weakness cut-points had a 2.22 (CI: 1.16, 4.24) higher hazard for mortality, but no significant associations were observed for those below 1 or 2 weakness cut-points. Persons with obesity that were below the collective weakness categories had a higher hazard for mortality: 1.43 (CI: 1.16, 1.78) for 1 cut-point, 1.75 (CI: 1.41, 2.17) for 2 cut-points, and 2.00 (CI: 1.62, 2.50) for all 3 cut-points.
DISCUSSION
The principal findings of this investigation revealed that persons below the SDOC weakness cut-points, used individually or collectively, had a higher risk for early all-cause mortality. Specifically, persons below each cut-point had a higher mortality risk: 45% for absolute weakness, 39% for BMI normalized weakness, 33% for body mass normalized weakness. When the weakness cut-points were used collectively, persons below 1, 2, or all 3 cut-points had a 37%, 47%, and 69% higher mortality risk. The role of underweight or obesity status may influence these cut-points for time to mortality. Our findings suggest that the absolute and body size normalized cut-points are useful for predicting early all-cause mortality risk. When these cut-points are used collectively instead of individually, mortality risk elevates as the being below more weakness categories increases, especially when below all 3 cut-points. While using these absolute and body size normalized weakness cut-points individually may help to determine mortality risk, the prognostic value of these cut-points could be improved when they are used collectively, and when underweight or obesity status are acknowledged.
Our results align with findings from other investigations examining the association between weakness and early all-cause mortality. For example, another study that categorized men with HGS <26-kg and women with HGS <16-kg as weak revealed that they had a 67% and 39% increased risk for early all-cause mortality, respectively (7). Our findings also showed that persons considered weak with the BMI and body mass normalized cut-points had a higher all-cause mortality risk, suggesting that absolute or body size normalized weakness cut-points each have utility for predicting mortality. There are some indications that a causal association may exist for weakness and early all-cause mortality, but this association could be attributed to weaker persons inabilities to overcome acute (e.g., myocardial infarction, stroke) and daily stressors (e.g., chronic disease) that may influence their longevity (26). The selection of absolute or body size normalized weakness cut-points might be informed by several factors such that the absolute cut-points are feasible and relatively accurate while the normalized cut-points could be more precise as a stand-alone estimate of strength capacity (24). Nonetheless, the SDOC weakness cut-points can be used individually for estimating early mortality risk.
Although the SDOC cut-points each have utility for predicting early all-cause mortality, the collective use of these cut-points may elevate their prognostic value. Specifically, we found that being below more weakness cut-points elevated mortality risk. The proportions of persons below a weakness cut-point may differ, and given there are several weakness cut-points that exist (24), using relevant cut-points collectively may help to improve how weakness is operationalized. Further, assessing other characteristics of physical functioning alongside collective weakness may better determine disablement (3,10). The role of body size should likewise be considered in weakness assessments among persons categorized as underweight or obese because such body size may influence the denominator of the normalization equations and serve as an indicator for other cardiometabolic conditions such as sarcopenic and osteosarcopenic obesity (11). Accordingly, continued monitoring of collective weakness status and body size may better inform how HGS is used.
The results of this investigation are consistent with other research suggesting low HGS is a risk factor for early all-cause mortality (7,16,21). Our findings provide support for using the SDOC weakness cut-points, and interventions targeting strength capacity may help longevity. For example, older inactive women may benefit from sprint interval concurrent training (38), while short-term power training may help to improve strength and function (14). Moreover, higher daily protein intake may also help to improve muscle mass and function (9), while Vitamin D supplementation may influence strength (2). Such physical activity and nutrition interventions may serve as framework for elevating strength, mass, and function to reduce mortality risk during aging, but feasibly measuring strength remains crucial. Thus, we suggest that HGS be measured as part of exercise assessments in relevant settings and populations, and during routine healthcare and sports medicine visits over time.
Some limitations should be acknowledged. Date of death was specified to the month and year, but not the exact day to help protect HRS subject anonymity and deidentification, thereby potentially biasing our estimates. However, population-based studies such as the HRS may not have exact dates for other health events (e.g., disease diagnosis). Persons that may have been institutionalized or unable to engage in HGS testing may have been weak and died during the study period, leading to underestimations of our findings. Although the HRS is a large population-based study, multiple interviewers may have collected HGS measurements, which could have threatened validity. Persons that were unable to execute HGS testing because of recent hand injuries or surgeries may have experienced nerve dysfunction (37), which may introduce selection bias for our study. Some factors influencing the association between HGS and time to mortality may not have been available in the HRS (e.g., specificity and duration of medication usage, environmental factors).Other body size metrics (e.g., stature) (28) that could be used to normalize HGS were not part of the SDOC. Examinations of muscle quality for the weakness assessments used in our investigation were unavailable. While the SDOC presents absolute and body size normalized weakness cut-points, an overall lack of consensus still exists for which cut-points should be used.
This investigation revealed that each absolute and body size normalized weakness cut-point from the SDOC helped to predict time to mortality in Americans. Although the SDOC weakness cut-points help to provide standardization to HGS methods, other opportunities may exist. For example, different dynamometers, protocols (e.g., body positions), and reporting (e.g., number of trials) may systematically bias HGS values and impact the proportions of persons above and below weakness cut-points. Such differences will limit drawing comparative conclusions across studies, but not all cut-points may recommend a HGS test protocol. Sports medicine practitioners and healthcare providers should collect HGS and track changes over time as appropriate.
PRACTICAL APPLICATIONS
Although low muscle strength is a robust indicator of poor human performance (35), weakness is likewise associated with deteriorating health during aging (27). Sports medicine and healthcare practitioners should continue emphasizing the importance of muscle strength for health and longevity, and such efforts should be advocated earlier in life because of how the link between strength and health changes over time (17). Measures of HGS remain a feasible and viable assessment of overall strength capacity, and could especially valuable when equipment and person-level functional abilities are limited. Acknowledgement for how weakness is differentially operationalized with cut-points should be considered. The Physical Activity Guidelines for Americans (30), recommendations for resistance training (15), and relevant dietary counseling (1) provide framework for helping our rapidly growing older American demographic increase or preserve strength.
ACKNOWLEDGEMENTS
Research reported in this publication was supported by the National Institute on Aging of the National Institutes of Health under Award Number R15AG072348 (to RM). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The results of the present study do not constitute endorsement of the product by the authors or the NSCA. The authors report no conflicts of interest.
REFERENCES
- 1.Bauer J, Biolo G, Cederholm T, Cesari M, Cruz-Jentoft AJ, Morley JE, et al. Evidence-based recommendations for optimal dietary protein intake in older people: a position paper from the PROT-AGE Study Group. J Am Med Dir Assoc 14: 542–559, 2013. [DOI] [PubMed] [Google Scholar]
- 2.Beaudart C, Buckinx F, Rabenda V, Gillain S, Cavalier E, Slomian J, et al. The effects of vitamin D on skeletal muscle strength, muscle mass, and muscle power: a systematic review and meta-analysis of randomized controlled trials. J Clin Endocrinol Metab 99: 4336–4345, 2014. [DOI] [PubMed] [Google Scholar]
- 3.Beaudart C, Rolland Y, Cruz-Jentoft AJ, Bauer JM, Sieber C, Cooper C, et al. Assessment of muscle function and physical performance in daily clinical practice. Calcif Tissue Int 105: 1–14, 2019. [DOI] [PubMed] [Google Scholar]
- 4.Benton MJ, Spicher JM, and Silva-Smith AL. Validity and reliability of handgrip dynamometry in older adults: A comparison of two widely used dynamometers. Plos One 17: e0270132, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bohannon RW. Grip strength: an indispensable biomarker for older adults. Clin Interv Aging 14: 1681, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Cawthon PM, Manini T, Patel SM, Newman A, Travison T, Kiel DP, et al. Putative cut-points in sarcopenia components and incident adverse health outcomes: an SDOC analysis. J Am Geriatr Soc 68: 1429–1437, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Celis-Morales CA, Welsh P, Lyall DM, Steell L, Petermann F, Anderson J, et al. Associations of grip strength with cardiovascular, respiratory, and cancer outcomes and all cause mortality: prospective cohort study of half a million UK Biobank participants. bmj 361, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Crimmins E, Guyer H, Langa K, Ofstedal MB, Wallance R, and Weir D. Documentation of physical measures, anthropometrics and blood pressure in the Health and Retirement Study. Available from: https://hrs.isr.umich.edu/sites/default/files/biblio/dr-011.pdf. Accessed August 1, 2023.
- 9.Deer RR and Volpi E Protein intake and muscle function in older adults. Curr Opin Clin Nutr Metab Care 18: 248, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Dommershuijsen LJ, Isik BM, Darweesh SK, van der Geest JN, Ikram MK, and Ikram MA. Unraveling the association between gait and mortality—one step at a time. J Gerontol Ser A 75: 1184–1190, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Donini LM, Busetto L, Bischoff SC, Cederholm T, Ballesteros-Pomar MD, Batsis JA, et al. Definition and diagnostic criteria for sarcopenic obesity: ESPEN and EASO consensus statement. Obes Facts 15: 321–335, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Feng X, Croteau K, Kolt GS, and Astell-Burt T Does retirement mean more physical activity? A longitudinal study. BMC Public Health 16: 1–7, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Fisher GG and Ryan LH. Overview of the health and retirement study and introduction to the special issue. Work Aging Retire 4: 1–9, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Fraga-Germade E, Carballeira E, and Iglesias-Soler E Effect of Resistance Training Programs With Equated Power on Older Adults’ Functionality and Strength: A Randomized Controlled Trial. J Strength Cond Res 38: 153–163, 2024. [DOI] [PubMed] [Google Scholar]
- 15.Fragala MS, Cadore EL, Dorgo S, Izquierdo M, Kraemer WJ, Peterson MD, et al. Resistance training for older adults: position statement from the national strength and conditioning association. J Strength Cond Res 33, 2019. [DOI] [PubMed] [Google Scholar]
- 16.García-Hermoso A, Cavero-Redondo I, Ramírez-Vélez R, Ruiz JR, Ortega FB, Lee D-C, et al. Muscular strength as a predictor of all-cause mortality in an apparently healthy population: a systematic review and meta-analysis of data from approximately 2 million men and women. Arch Phys Med Rehabil 99: 2100–2113, 2018. [DOI] [PubMed] [Google Scholar]
- 17.Jung HN, Jung CH, and Hwang Y-C. Sarcopenia in youth. Metabolism 155557, 2023. [DOI] [PubMed] [Google Scholar]
- 18.Lamarca R, Alonso J, Gomez G, and Muñoz Á Left-truncated data with age as time scale: an alternative for survival analysis in the elderly population. J Gerontol A Biol Sci Med Sci 53: M337–M343, 1998. [DOI] [PubMed] [Google Scholar]
- 19.Langa KM, Larson EB, Karlawish JH, Cutler DM, Kabeto MU, Kim SY, 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 4: 134–144, 2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Leong DP, Teo KK, Rangarajan S, Lopez-Jaramillo P, Avezum A, Orlandini A, et al. Prognostic value of grip strength: findings from the Prospective Urban Rural Epidemiology (PURE) study. The Lancet 386: 266–273, 2015. [DOI] [PubMed] [Google Scholar]
- 21.López-Bueno R, Andersen LL, Koyanagi A, Núñez-Cortés R, Calatayud J, Casaña J, et al. Thresholds of handgrip strength for all-cause, cancer, and cardiovascular mortality: a systematic review with dose-response meta-analysis. Ageing Res Rev 101778, 2022. [DOI] [PubMed] [Google Scholar]
- 22.Manini TM, Patel SM, Newman AB, Travison TG, Kiel DP, Shardell MD, et al. Identification of sarcopenia components that discriminate slow walking speed: a pooled data analysis. J Am Geriatr Soc 68: 1419–1428, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.McGrath R Comparing absolute handgrip strength and handgrip strength normalized to body weight in aging adults. Aging Clin Exp Res 31: 1851–1853, 2019. [DOI] [PubMed] [Google Scholar]
- 24.McGrath R, Cawthon P, Clark B, Fielding R, Lang J, and Tomkinson G Recommendations for Reducing Heterogeneity in Handgrip Strength Protocols. J Frailty Aging 11: 143–150, 2022. [DOI] [PubMed] [Google Scholar]
- 25.McGrath R, FitzSimmons S, Andrew S, Black K, Bradley A, Christensen BK, et al. Prevalence and Trends of Weakness Among Middle-Aged and Older Adults in the United States. J Strength Cond Res 10–1519, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.McGrath R, Tomkinson GR, LaRoche DP, Vincent BM, Bond CW, and Hackney KJ. Handgrip strength asymmetry and weakness may accelerate time to mortality in aging Americans. J Am Med Dir Assoc 21: 2003–2007, 2020. [DOI] [PubMed] [Google Scholar]
- 27.McGrath RP, Kraemer WJ, Snih SA, and Peterson MD. Handgrip strength and health in aging adults. Sports Med 48: 1993–2000, 2018. [DOI] [PubMed] [Google Scholar]
- 28.Nevill AM, Tomkinson GR, Lang JJ, Wutz W, and Myers TD. How should adult handgrip strength be normalized? Allometry reveals new insights and associated reference curves. Med Sci Sports Exerc, 2021. [DOI] [PubMed] [Google Scholar]
- 29.Patel SM, Duchowny KA, Kiel DP, Correa-de-Araujo R, Fielding RA, Travison T, et al. Sarcopenia Definition & Outcomes Consortium Defined Low Grip Strength in Two Cross-Sectional, Population-Based Cohorts. J Am Geriatr Soc 68: 1438–1444, 2020. [DOI] [PubMed] [Google Scholar]
- 30.Piercy KL, Troiano RP, Ballard RM, Carlson SA, Fulton JE, Galuska DA, et al. The physical activity guidelines for Americans. JAMA 320: 2020–2028, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Plassman BL, Newman TT, Welsh KA, and Helms M Properties of the Telephone Interview for Cognitive Status: application in epidemiological and longitudinal studies. Neuropsychiatry Neuropsychol Behav Neurol 7: 235–241, 1994. [Google Scholar]
- 32.Sample sizes and response rates. Health and Retirement Study. Available from: https://hrs.isr.umich.edu/sites/default/files/biblio/ResponseRates_2017.pdf. Accessed August 1, 2023.
- 33.Sayer AA and Kirkwood TB. Grip strength and mortality: a biomarker of ageing? The lancet 386: 226–227, 2015. [DOI] [PubMed] [Google Scholar]
- 34.Sonnega A, Faul JD, Ofstedal MB, Langa KM, Phillips JW, and Weir DR. Cohort profile: the health and retirement study (HRS). Int J Epidemiol 43: 576–585, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Suchomel TJ, Nimphius S, and Stone MH. The importance of muscular strength in athletic performance. Sports Med 46: 1419–1449, 2016. [DOI] [PubMed] [Google Scholar]
- 36.Turvey CL, Wallace RB, and Herzog R A revised CES-D measure of depressive symptoms and a DSM-based measure of major depressive episodes in the elderly. Int Psychogeriatr 11: 139–148, 1999. [DOI] [PubMed] [Google Scholar]
- 37.Verdecchia N, Johnson J, Baratz M, and Orebaugh S Neurologic complications in common wrist and hand surgical procedures. Orthop Rev 10, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Wadsworth DD, Rodriguez-Hernandez M, Huffman LS, McDonald JR, Spring KE, and Pascoe DD. Adaptations to a concurrent exercise training program in inactive aging women. J Strength Cond Res 36: 3217–3223, 2022. [DOI] [PubMed] [Google Scholar]
- 39.Weir D Validating mortality ascertainment in the Health and Retirement Study. Available from: https://hrs.isr.umich.edu/sites/default/files/biblio/Weir_mortality_ascertainment.pdf. Accessed January 9, 2024.