Grip Strength as a Marker of Ageing and Health: Evidence from Six Low - - PDF document

1 | P a g e

Grip Strength as a Marker of Ageing and Health: Evidence from Six Low and Middle- Income Countries By

• P. Arokiasamy and Y. Selvamani

Abstract Grip strength is a widely accepted marker of ageing and health. While chronological age is well recognized as a marker of frailty, many studies have investigated the close association between grip strength and frailty. Other studies have shown grip strength as a predictor of disability, morbidity and mortality. The objective of this paper is to assess the association between grip strength, overall and specific dimensions of multidimensional health such as functional loss, general health, cognition, chronic multimorbidity conditions and quality of life among older adults in age 50+. Cross-country analysis is conducted using the WHO‟s Study

• n global AGEing and adult health (SAGE) data for Low and Middle-Income Countries

(LMICs) of China, China, Ghana India, Mexico, Russia and South Africa. Regression models are estimated to predict the association between grip strength, functional health, multimorbidity, cognition and quality of life incorporating age, health risk factors and SES

• covariates. Results showed prevalence of morbidity and multi-morbidity conditions are

associated with reduced grip strength. Regression analysis revealed significant association of multimorbidity conditions with grip strength in these six countries. Moreover, this association is stronger and consistent among men. We observed a strong association between grip strength and subjective health measures and cognition. Background Low and middle income countries (LIMCs) currently experience high speed of population

• aging. Consequently, the burden of NCDs is escalating; previous research indicates that NCDs

have a disproportionately high impact on low- and middle-income countries. The implications

• f aging on health and health care are major public health challenge across globe. Aging has

greatly contributed to steep increase in the prevalence of chronic morbidity and frailty. Grip strength is a widely accepted marker of ageing and health. While chronological age is well recognized as a marker of frailty, many studies have investigated the close association between grip strength and frailty. Other studies have shown grip strength as a predictor of disability,

2 | P a g e

morbidity and mortality (Giampaoli et al., 1999; Al Snih et al., 2004; Gale et al., 2005; Newman et al., 2006; Leong et al., 2015). Studies have shown combination of one or more diseases, also termed as multimorbidity is a major risk factor for individual survival and quality of life. Multimordity is referred as the presence of one or more morbidity conditions which has strong implication

• n the health care expenditure (Lee et al., 2015). Multimorbidity has been associated with

adverse health outcomes, such as reduced physical function, poor quality of life, poor self-rated health (SRH), increased use of inpatient and ambulatory care and mortality. While, many studies have examined the association of multimorbidity with subjective health, mortality and quality of life (Marengoni et al., 2008; Garin et al., 2014; Arokiasamy et al., 2015), relatively, much less is known about the association of grip strength with ageing and health in LMICs. This study has two objectives; first, to examine the association between multimorbidity and grip strength. Secondly, to study the association of grip strength with the health and quality

• f life outcomes to understand the role of physical strength in determining later life general

health and cognition. Muscle weakness in old age is associated with morbidity and mortality (Rantanen et al., 2000; Rosero-Bixby and Dow, 2012; Leong et al., 2015; Xu and Hao, 2017), disability (Giampaoli et al., 1999; Taekema et al., 2010). Yet, very few studies have examined the role of grip strength on subjective health and cognition measures in developing countries. The aim of this study is to assess the association between grip strength, overall and specific dimensions of multidimensional frailty such as functional loss, chronic multimorbidity conditions and quality of life among older adults using the WHO‟s Study on global AGEing and adult health (SAGE) data for the six Low and Middle Income Countries (LMICs) of China, China, Ghana India, Mexico, Russia and South Africa. Materials and Methods We use data from the WHO’s Study on Global Ageing and Adult Health (SAGE). SAGE is a nationally representative household health survey conducted in six countries: China, Ghana, India, Mexico, the Russian Federation and South Africa. This analysis focussed on sample of

• lder adults aged 50 years and above. The analysis was carried out using the nationally

representative sample with the total sample of 30155 older adults in age 50+. Details on sampling and the methodology are available in Kowal et al, 2012. SAGE is nationally representative and cohort study conducted in six countries; India, China, Ghana, Mexico, Russia and South Africa during 2007-10. The aim of the SAGE was to fulfil the data gaps to

3 | P a g e

understand the health and wellbeing of the growing aging population in six low and middle- income countries. SAGE measures are comparable with other studies from high-income countries such as the Health and Retirement Study (HRS), the Survey of Health, Ageing and Retirement in Europe (SHARE). SAGE collects data on self-reported as well as biomarkers data on different domains of health, wellbeing and anthropometric indicators. A detailed information about data description given in Kowal et al. (2012). Definition and Variable Construction Grip strength Grip strength is a measure of muscle strength which is a strong predictor of individual health, functionality and mortality. In WHO-SAGE grip strength was assessed in both the hands using a hand dynamometer. In the analysis, we have taken the best of four measurements. Since, grip strength is one of the main outcome variable in this study, we excluded outliers. The main analysis excluded less than 1 and above 99 percentile of grip strength measurements. Multi-morbidity Mult-morbidity is defined as the presence of one or more chronic condition at the time of data

• collection. We have included eight chronic health conditions, namely: arthritis, stroke, angina

pectoris, diabetes mellitus, asthma, hypertension, chronic lung disease and visual acuity. Among these, for arthritis, angina pectoris, asthma, lung disease, SAGE survey provides two types of measures: First, self-reports of the diagnosis of individual diseases and second is the symptom based assessment or direct health examination of abovementioned diseases. The specific question asked in SAGE for self-reports is: “Have you ever been diagnosed with/told that you have disease name? Thus, we have considered an individual as suffering from these diseases if he/she is found positive in the symptom based assessment. For, stroke and diabetes mellitus we have relied on the self-reports of diagnosis and for hypertension and visual acuity, we have used measured outcomes of blood pressure monitor and vision test. Self-Rated Health (SRH) This study used self-rated health as one of the outcome variables. In SAGE, self-rated health was assessed on a five-point scale with the following question: In general, how would you rate your health today? The response categories were: ‘very good’, ‘good’, ‘moderate’, ‘bad’ and

4 | P a g e

‘very bad’. In the analysis, ‘bad’ and ‘very bad’ health categories were combined to represent poor self-rated health. WHODAS score The 12-item version of WHODAS 2.0 encompasses all six domains of the full version: cognition, mobility, self-care, getting along, life activities and participation in society. Its psychometric properties in older people from low and middle income countries have been validated previously. WHODAS score ranges from 0 (no disability) to 100 (full disability). Cognitive Score Index To understand the composite effect of cognition we made a cognitive index combining four variables: verbal fluency, verbal recall, digit span forward and digit span backward. The index was derived using Principal Components Analysis (PCA), a mathematical tool which helps in creating a composite index using uncorrelated components, where each component captures the largest possible variation in the original variables. Selected raw scores for cognitive tasks were bundled into three domains (digit span, memory and executive functioning) to yield compound cognitive scores. This was done to condense the number of cognitive variables while refining the robustness of the underlying cognitive construct. We followed two steps to make a cognitive index: Step 1: All four variables were in different scales. So first, we standardized these variables. A standardized variable (sometimes called a z-score or a standard score) is a variable that has been rescaled to have a mean of zero and a standard deviation of one. Each case's value on the standardized variable designates its difference from the mean of the primary variable in some standard deviations (of the original variable). Step 2: PCA is a multivariate statistical technique used for extracting from a set of variables those few orthogonal linear combinations that capture the common information most

• successfully. Further, this index comprises both values, positive and negative. So we converted

this index into a 0–100 scale which facilitates easier interpretation of the data. Increasing the score meant better cognitive abilities. Quality of Life Index (WHO-QoL) We used the quality of life questionnaire (S-QoL 30) from the WHO-SAGE data set, which was a particular, self-administered and multidimensional QoL questionnaire designed for

• people. It included 30 items describing five dimensions: physical health (PHY), psychological

5 | P a g e

health (PSY), level of independence (IND), social relation (SOC) and environment (ENV). It also included a total score (Index). The five dimensions and the Index score ranged from 0 to 100; higher scores indicated better quality of life. Covariates Age, place of residence[urban/rural], years of education [0-4 years of education, 6-9 years of education and 10+ years], wealth quintile[poorest, poorer, middle, richer, richest], BMI [underweight, normal,

• verweight

and

• bese],

tobacco use[no/yes] physical activity[active/inactive]. Statistical Analysis First, we assessed the association of multi-morbidity and grip strength using stratified multivariate linear regression methods. Further, we generated sex and country-specific grip quintile to examine the association of grip strength with four outcomes- poor self-rated health, WHO-DAS, quality of life and cognition. Logistic regression was used to examine the association of grip quintile with poor self-rated health and functional health. The multivariate linear regression model was adopted to assess the linkages of grip strength with cognition and quality of life. The predicted probabilities of grip strength by multimorbidity was generated adjusting for socio-demographic and health risk factors. Similarly, adjusted WHO-DAS, quality of life and cognition scores were predicted by grip quintile from multivariate linear regression models. The adjusted prevalence of poor self-rated health was generated from logit regression model. Results Table 1 shows descriptive statistics of the study population. Of the total sample of 30155 older adults for all six countries, China has proportionately a higher sample size with 40 percent. The distribution of sample by age group indicate higher the sample size in the age group 50-59 in all the countries. In Ghana and Russia, the sample size in 70+ age group is higher compared to

• ther countries. In Mexico, Russia and South Africa, the share of older women participants are

much higher than older men. The distribution of sample by place of residence describe a higher rural sample was from India, China and Ghana. Education attainment of the older adults was lower in India, Ghana and Mexico, where more than 50 percent of older adults attained only no schooling or up to four years of schooling. In India, a larger proportion of older adults are underweight and use tobacco. Fortyone percent of study population in Mexico and 57% in South Africa are physically inactive.

6 | P a g e

Figure 1 shows the prevalence of multiple morbidity conditions in six countries. The prevalence of any chronic morbidity was higher in Ghana with 41 percent. In Russia, 35 percent

• f older adults had 3 or more chronic diseases.

We predicted grip strength by multi-morbidity conditions in six countries as shown in Figure 2. Multi-morbidity is seen associated with lower grip strength in all countries. Moreover, the decline is much sharper among men. Regression analysis shows that multi- morbidity significantly lowers grip strength controlling for sociodemographic and health

• variables. This association is very strong for men. The association is significant for women in

China, Ghana, Russia and South Africa. The adjusted prevalence of poor self-rated health shown in figure 3. Across the countries, a declining trend is observed in poor self-rated health for higher grip quintile. This association is stronger for India, Ghana and Russia. Figure 4 shown the adjusted WHO-DAS score by grip quintile. There is notable decline in WHO-DAS score in higher grip quintile and this observed in all the countries except China. Figure 5 displays the association of grip quintile with quality of life and cognition. Higher grip quintile is seen positively associated with better quality of life and cognition which is consistent across all the six countries. Results from regression analysis on the association of grip quintile with self-rated health, WHO-DAS, quality of life and cognition are shown in table 3. The results are adjusted for age, residence, gender, schooling, wealth quintile, height, body mass index, tobacco use and physical activity. Older adults in lowest grip quintile were 2.68 times more likely to report poor health in India compared to highest grip quintile. Similar findings are evident from other countries too; older adults in China, Ghana, Russia and South Africa had poor general health with lower grip strength. The pattern is similar with functional health that lowest grip quintile is positively associated higher WHO-DAS score in six countries. The fully adjusted results of grip quintile with quality of life and cognition indicate a negative and significant association in all the countries. Older adults in lowest grip quintile had [β =-3.62, 95%CI -4.51, -2.72] lower quality of life score as compared to highest grip quintile in India. Similarly, older adults in lowest grip quintile had lower quality life score in China, Ghana, Mexico, Russia and South

• Africa. Higher grip quintile was strongly associated with better cognitive abilities in all the

countries except China.

7 | P a g e

Conclusions In this study, we analysed SAGE data from six low and middle-income countries. Results revealed negative implications of chronic morbidity conditions on objective measure of health. Previous studies have shown a close association between combinations of multiple chronic morbidities with mental health and subjective health measures such as self-rated health, functional health and quality of life measures (Arokiasamy et al., 2015). This study has deepened the understanding of multiple chronic diseases on grip strength which is a strong predictor of morbidity and mortality (Rantanen et al., 2000; Rosero-Bixby and Dow, 2012; Leong et al., 2015; Xu and Hao, 2017), disability (Giampaoli et al., 1999; Taekema et al., 2010). Secondly, we examined the association of grip strength with four outcomes- self-rated health, functional health, quality of life and cognition. Results revealed a significant and strong association between multiple chronic conditions and grip strength. So far, only few studies have tested the association of chronic diseases and grip strength and found a strong association between chronic morbidities and grip strength in men (Cheung et al., 2013). In this context, the results of this study contribute to a better understanding of the implications of chronic morbidities in low and middle-income countries. Further, the results of this study further adds to the literature: a significant association

• f grip strength with subjective health, quality of life and cognitive functions is evident.

Previous studies have observed an association between grip strength and quality of life mainly from developed countries (Sayer et al., 2006), cognition (Sternäng et al., 2016) and disability measures (Giampaoli et al., 1999; Taekema et al., 2010). In this context, this study strongly highlights the importance of grip strength as marker of health in old age for low and middle- income countries. The association between grip strength and health, cognition and quality of

• utcomes can be bidirectional.

While, the prevalence of chronic diseases increases across the developing countries with the ongoing demographic transition, the measures to prevent chronic morbidities is necessary which will improve the general health and quality of life of the older population. Further, measures to improve the grip strength through physical activity and nutritional interventions are important. Also, older adults with chronic morbidities need to be given special interventions to improve the physical strength which can help them their general and functional health.

8 | P a g e

The strengths of this study are; we have used a national representative data sets from six countries. Therefore, the results can be generalized at the national level. Second, while, many studies have found a heterogeneity in reported measure of chronic diseases, especially in developing countries (Sen, 2002; Vellakkal et al., 2013; Arokiasamy et al., 2017), in this study, we have used objective measures of chronic diseases. Limitations: The results of the study are based on cross-sectional data. In conclusion, this study highlights significant association of multiple morbidity and grip strength. A robust impact of grip strength with subjective health and cognition indicates the multi-dimensional predictive power of grip strength. The results underline the importance

• f the disease free old age and better physical strength for healthy aging in low and middle-

income countries. References

• 1. Al Snih, S., Markides, K. S., Ottenbacher, K. J., & Raji, M. A. (2004). Hand grip

strength and incident ADL disability in elderly Mexican Americans over a seven-year

• period. Aging clinical and experimental research, 16(6), 481-486.
• 2. Arokiasamy, P., Kowal, P., Capistrant, B. D., Gildner, T. E., Thiele, E., Biritwum, R.

B., ... & Guo, Y. (2017). Chronic Noncommunicable Diseases in 6 Low-and Middle- Income Countries: Findings From Wave 1 of the World Health Organization's Study

• n

Global Ageing and Adult Health (SAGE). American Journal

• f

Epidemiology, 185(6), 414-428.

• 3. Arokiasamy, P., Uttamacharya, U., Jain, K., Biritwum, R. B., Yawson, A. E., Wu, F.,

... & Afshar, S. (2015). The impact of multimorbidity on adult physical and mental health in low-and middle-income countries: what does the study on global ageing and adult health (SAGE) reveal?. BMC medicine, 13(1), 1.

• 4. Cheung, C. L., Nguyen, U. S. D., Au, E., Tan, K. C., & Kung, A. W. (2013). Association
• f handgrip strength with chronic diseases and multimorbidity. Age, 35(3), 929-941.
• 5. Gale, C. R., Martyn, C. N., Cooper, C., & Sayer, A. A. (2007). Grip strength, body

composition, and mortality. International journal of epidemiology, 36(1), 228-235.

• 6. Garin, N., Olaya, B., Moneta, M. V., Miret, M., Lobo, A., Ayuso-Mateos, J. L., & Haro,
• J. M. (2014). Impact of multimorbidity on disability and quality of life in the Spanish
• lder population. PloS one, 9(11), e111498.
• 7. Giampaoli, S., Ferrucci, L. U. I. G. I., Cecchi, F., Noce, C. L., Poce, A., Dima, F., ... &

Menotti, A. L. E. S. S. A. N. D. R. O. (1999). Hand-grip strength predicts incident disability in non-disabled older men. Age and ageing, 28(3), 283-288.

• 8. Kowal, P., Chatterji, S., Naidoo, N., Biritwum, R., Fan, W., Ridaura, R. L., ... &

Snodgrass, J. J. (2012). Data resource profile: the World Health Organization Study on global AGEing and adult health (SAGE). International journal of epidemiology, 41(6), 1639-1649.

• 9. Lee, J. T., Hamid, F., Pati, S., Atun, R., & Millett, C. (2015). Impact of

noncommunicable disease multimorbidity on healthcare utilisation and out-of-pocket

9 | P a g e

expenditures in middle-income countries: cross sectional analysis. PloS one, 10(7), e0127199.

• 10. Leong, D. P., Teo, K. K., Rangarajan, S., Lopez-Jaramillo, P., Avezum, A., Orlandini,

A., ... & Rahman, O. (2015). Prognostic value of grip strength: findings from the Prospective Urban Rural Epidemiology (PURE) study. The Lancet, 386(9990), 266- 273.

• 11. Marengoni, A., Von Strauss, E., Rizzuto, D., Winblad, B., & Fratiglioni, L. (2009). The

impact of chronic multimorbidity and disability on functional decline and survival in elderly persons. A community‐based, longitudinal study. Journal of internal medicine, 265(2), 288-295.

• 12. Newman, A. B., Kupelian, V., Visser, M., Simonsick, E. M., Goodpaster, B. H.,

Kritchevsky, S. B., ... & Harris, T. B. (2006). Strength, but not muscle mass, is associated with mortality in the health, aging and body composition study cohort. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 61(1), 72-77.

• 13. Rantanen, T., Harris, T., Leveille, S. G., Visser, M., Foley, D., Masaki, K., & Guralnik,
• J. M. (2000). Muscle strength and body mass index as long-term predictors of mortality

in initially healthy men. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 55(3), M168-M173.

• 14. Rosero-Bixby, L., & Dow, W. H. (2012). Predicting mortality with biomarkers: a

population-based prospective cohort study for elderly Costa Ricans. Population health metrics, 10(1), 11.

• 15. Taekema, D. G., Gussekloo, J., Maier, A. B., Westendorp, R. G., & de Craen, A. J.

(2010). Handgrip strength as a predictor of functional, psychological and social health. A prospective population-based study among the oldest old. Age and ageing, 39(3), 331-337.

• 16. Vellakkal, S., Subramanian, S. V., Millett, C., Basu, S., Stuckler, D., & Ebrahim, S.

(2013). Socioeconomic inequalities in non-communicable diseases prevalence in India: disparities between self-reported diagnoses and standardized measures. PloS one, 8(7), e68219.

• 17. Xu, L., & Hao, Y. T. (2017). Effect of handgrip on coronary artery disease and

myocardial infarction: a Mendelian randomization study. Scientific reports, 7.

10 | P a g e

Table 1 Characteristics of the Study Population in Six Countries, WHO-SAGE Wave 1

Characteristics India China Ghana Mexico Russia South Africa (n=6,283) (n=12,084) (n=3,977) (n=1,933) (n=2,954) (n=2,924) Age 50-59 49.5 45.6 41.1 50.4 47.3 49.2 60-69 30.8 32.2 27.9 25.3 25.5 31.2 70+ 19.7 22.2 31.0 24.3 27.2 19.6 Sex Male 51.1 49.8 53.1 47.0 39.9 44.8 Female 48.9 50.2 46.9 53.0 60.1 55.2 Residence Urban 28.3 45.5 41.2 79.6 71.6 65.0 Rural 71.7 54.5 58.8 20.4 28.4 35.1 Education 0-5 years 70.4 49.3 61.9 54.5 6.3 46.7 6-9 years 13.1 35.7 8.3 34.9 18.1 31.7 10+ 16.5 15.0 29.8 10.5 75.7 21.6 Wealth Quintile Lowest 18.1 16.6 18.0 14.8 14.3 20.6 2 18.9 18.7 19.1 25.8 16.9 20.6 3 19.3 20.8 20.3 16.3 19.4 18.5 4 19.7 23.5 21.2 16.4 22.5 19.0 Highest 24.0 20.4 21.4 26.7 26.9 21.3 Body mass index Underweight 37.9 4.0 14.1 0.5 1.3 3.4 Normal 48.7 59.8 55.1 21.0 21.9 23.9 Overweight 10.7 30.0 20.0 50.3 43.0 26.3 Obese 2.7 6.1 10.9 28.2 33.7 46.5 Tobacco use No 52.6 72.9 92.2 86.6 78.3 81.8 Yes 47.4 27.1 7.8 13.4 21.7 18.2 Physical Inactivity No 69.9 63.9 72.8 59.2 78.2 43.0 Yes 30.1 36.1 27.2 40.8 21.8 57.0 Total 20.8 40.1 13.2 6.4 9.8 9.7

11 | P a g e

Figure 1. Weighted Prevalence of Chronic Diseases Count in 6 Countries, WHO-SAGE Wave 1

21 19 19 13 17 10 36 37 41 37 26 38 24 28 27 32 22 30 19 16 12 18 35 22 5 10 15 20 25 30 35 40 45 India China Ghana Mexico Russia South Africa 0 disease 1 disease 2 disease 3+ disease

12 | P a g e

Figure 2. Predicted grip strength adjusted for sociodemographic and health indicators Grip Strength (kg)

19.18 23.87 25.37 20.29 28.67 33.99 18.8 21.53 23.98 19.21 25.17 28.56 10 15 20 25 30 35 40 India China Ghana Mexico Russia South Africa

Multimorbidity and Grip Stregnth (kg) in Women

3+ 2 disease 1 disease No disease 28.81 35.94 34.63 31.46 43.6 42.21 27.7 32.92 30.72 28.67 40.98 35.34 10 15 20 25 30 35 40 45 50 India China Ghana Mexico Russia South Africa

Multimorbidity and Grip Stregnth(kg) in Men

3+ 2 disease 1 disease No disease

13 | P a g e

Table 2. Regression results of multimorbidity and grip strength (kg) in 6 Countries, WHO-SAGE Wave 1

India China Ghana Mexico Russia South Africa

β [95% CI] β [95% CI] β [95% CI] β [95% CI] β [95% CI] β [95% CI]

Men No disease 1.11***[0.31,1.92] 3.01***[2.12,3.90] 3.91***[2.23,5.60] 2.78***[0.86,4.70] 2.61**[0.55,4.67] 6.87***[3.46,10.2] 1 disease 0.95**[0.21,1.68] 2.39***[1.59,3.19] 3.14***[1.55,4.72] 1.18[-0.38, 2.75] 0.92[-0.90, 2.76] 5.19***[2.54, 7.83] 2 disease 0.65[-0.13, 1.44] 1.34***[0.52,2.17] 1.14[-0.52, 2.81]

• 0.07[-1.69,1.55]

0.109[-1.71,1.92] 2.10[-0.65, 4.87] 3+ disease Ref Ref Ref Ref Ref Ref

R2

0.2486 0.215 0.129 0.2447 0.2493 0.0604 Sample 3115 5374 2022 693 1010 958 Women No disease 0.37[-0.28, 1.04] 2.34***[1.64,3.03] 1.39*[-0.08, 2.86] 1.08[-0.25, 2.42] 3.49***[2.19,4.79] 5.43***[2.95,7.91] 1 disease 0.45[-0.10, 1.00] 1.90***[1.31,2.50] 1.76***[0.55,2.98] 0.13[-0.84, 1.11] 1.54***[0.52,2.56] 2.90***[1.18,4.61] 2 disease 0.51*[-0.08, 1.10] 1.17***[0.57,1.78] 1.18*[-0.09, 2.46] 0.28[-0.68, 1.25] 1.67***[0.72,2.63] 2.49***[0.74,4.24] 3+ disease Ref Ref Ref Ref Ref Ref

R2

0.1631 0.1491 0.0736 0.1008 0.2276 0.0481 Sample 3014 6272 1804 1044 1765 1388

Results are adjusted for age, residence, years of education and wealth quintile, height, BMI, tobacco use and physical activity.

* p,.05; **p,.01; ***p,.001.

14 | P a g e

Figure 3. Adjusted prevalence of poor self-rated health by Grip Quintile in 6 Countries, WHO-SAGE Wave 1 Figure 4. Adjusted WHO-DAS score by grip quintile in six countries, WHO-SAGE Wave 1

23.67 21.52 18.49 8.99 25.41 16.26 10.35 15.56 9.39 10.35 16.39 9.34 0.00 5.00 10.00 15.00 20.00 25.00 30.00 India China Ghana Mexico Russia South Africa

Grip Quintile and Self-Rated Health

Q1 (lowest) Q2 Q3 Q4 Q5 (highest) 29.45 10.62 21.31 19.16 25.35 22.35 25.14 8.70 18.54 15.83 17.71 16.39 0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 India China Ghana Mexico Russia South Africa

Grip Quintile and WHO-DAS Score

Q1 (lowest) Q2 Q3 Q4 Q5 (highest)

15 | P a g e

Figure 5. Adjusted quality of life and cognition scores by grip quintile in six countries, WHO-SAGE Wave 1

36.86 51.25 42.51 41.73 47.10 41.09 39.99 51.75 47.62 43.72 54.16 48.62 20.00 25.00 30.00 35.00 40.00 45.00 50.00 55.00 60.00 India China Ghana Mexico Russia South Africa

Grip Quintile and Cognition Score

Q1 (lowest) Q2 Q3 Q4 Q5 (highest) 47.47 51.71 45.51 52.01 48.04 46.47 51.09 53.35 48.42 54.38 50.14 49.39 20.00 25.00 30.00 35.00 40.00 45.00 50.00 55.00 60.00 India China Ghana Mexico Russia South Africa

Grip Quintile and Quality of Life Score

Q1 (lowest) Q2 Q3 Q4 Q5 (highest)

16 | P a g e

Table 3 Multivariate regression results of poor self-rated health, functional health, quality of life and cognition.

India China Ghana Mexico Russia South Africa OR [95% CI] OR [95% CI] OR [95% CI] OR [95% CI] OR [95% CI] OR [95% CI] Self-rated health Grip Quintile Q1 (lowest) 2.68***[2.07, 3.47] 1.48***[1.25, 1.75] 2.18***[1.55, 3.06] 0.85[0.49, 1.46] 1.73***[1.23, 2.45] 1.88***[1.27, 2.79] 2 1.82***[1.39, 2.37 ] 1.28***[1.08, 1.51] 2.03***[1.44, 2.85] 0.80[0.47, 1.36] 1.48**[1.06, 2.08] 1.55**[1.03, 2.33] 3 1.67***[1.28, 2.19] 1.12[0.95, 1.32] 1.18[0.83, 1.69] 1.16[0.69, 1.95] 1.20[0.85, 1.69] 0.98[0.64, 1.51] 4 1.48***[1.13, 1.94] 1.03[0.86, 1.22] 1.13[0.79, 1.62] .95[0.56, 1.61] 1.00[0.69, 1.44] 1.17[0.75, 1.82] 5 (highest) Ref Ref Ref Ref Ref Ref WHO-DAS β [95% CI] β [95% CI] β [95% CI] β [95% CI] β [95% CI] β [95% CI] Grip Quintile Q1 (lowest) 4.30***[2.89, 5.71] 1.92***[1.26, 2.58] 2.77***[1.07, 4.46] 3.33***[0.87, 5.80] 7.64***[5.90, 9.38] 5.96***[3.76, 8.16] 2 2.67***[1.25, 4.10] 0.09[-0.55, 0.74] 6.67***[4.95, 8.38] 0.95[-1.41, 3.33] 2.60***[0.97, 4.23] 2.92**[0.71, 5.13] 3 2.11***[0.73, 3.50]

• 0.68**[-1.31, -0.05]

3.85***[2.18, 5.52] 0.96[-1.46, 3.39] 0.75[-0.83, 2.34]

• 1.27[-3.43, 0.87]

4 0.88[-0.47, 2.23]

• 0.57*[-1.20, 0.05]

2.42***[0.79, 4.05] 0.11[-2.30, 2.53] 0.006[-1.58, 1.60] 1.61[-0.61, 3.83] 5 (highest) Ref Ref Ref Ref Ref Ref Quality of Life (QoL) Grip Quintile Q1 (lowest)

• 3.62***[-4.51, -2.72]
• 1.63***[-2.38, -0.89]
• 2.90***[-4.12, -1.69]
• 2.36***[-3.90, -0.83]
• 2.09***[-3.59, -0.59]
• 2.91***[-4.41, -1.41]

2

• 2.44***[-3.34, -1.54]
• 1.18***[-1.91, -0.44]
• 5.19***[-6.42, -3.97]
• 1.52**[-3.00, -0.04]
• 2.34***[-3.75, -0.94]
• 1.46*[-2.97, 0.03]

3

• 1.89***[-2.76, -1.01]
• 0.15[-0.87, 0.56]
• 3.25***[-4.44, -2.06]
• 1.31*[-2.82, 0.20[
• 1.47**[-2.85, -0.10]
• .64[-2.11, 0.82]

4

• 1.52***[-2.38, -0.66]
• 0.07[ -0.78, 0.64]
• 1.60***[-2.77, -0.44]
• 0.58[-2.09, 0.91]

0.43[-0.94, 1.81]

• 0.30[-1.82, 1.21]

5 (highest) Ref Ref Ref Ref Ref Ref Cognition Grip Quintile Q1 (lowest)

• 3.13***[-3.84, -2.41]
• 0.49[-1.12, 0.13]
• 5.10***[-6.10, -4.11]
• 1.99***[-3.36, -0.62]
• 7.06***[-8.46, -5.66]
• 7.53***[ -8.89, -6.17]

2

• 2.24***[-2.96, -1.53]
• 0.55*[-1.17, 0.06]
• 5.36***[-6.37, -4.36]
• 1.62**[-2.94, -0.30]
• 3.40***[-4.71, -2.10]
• 5.47***[-6.84, -4.11]

3

• 0.89**[-1.59, -0.19]
• 0.001[-0.60, 0.60]
• 3.85***[-4.82, -2.88]
• 1.42**[-2.77, -0.06]
• 1.94***[-3.22, -.67]
• 4.32***[-5.64, -2.99]

4

• 0.90***[-1.58, -0.23]

0.06[-0.53, 0.66]

• 2.71***[-3.66, -1.76]
• 0.54[-1.88, 0.80]
• 1.50**[-2.79, -0.22]
• 4.45***[-5.82, -3.08]

5 (highest) Ref Ref Ref Ref Ref Ref

Regression models were adjusted for age, sex, place of residence, schooling, wealth quintile, physical activity, tobacco use and height.

* p,.05; **p,.01; ***p,.001