Epidemiology of sarcopenia among community-dwelling older adults in Taiwan: A pooled analysis for a broader adoption of sarcopenia assessments

bs_bs_banner Geriatr Gerontol Int 2014; 14 (Suppl. 1): 52 60 ORIGINAL ARTICLE Epidemiology of sarcopenia among community-dwelling older adults in Taiwan: A pooled analysis for a broader adoption of sarcopenia assessments I-Chien Wu, 1,2 Cheng-Chieh Lin, 2 Chao A. Hsiung, 1 Ching-Yi Wang, 3 Chih-Hsing Wu, 4,5 Ding-Cheng Derrick Chan, 6,7 Tsai-Chung Li, 8 Wen-Yuan Lin, 9 Kuo-Chin Huang, 10 Ching-Yu Chen 1,10 and Chih-Cheng Hsu 1,11 for the Sarcopenia and Translational Aging Research in Taiwan (START) Team* 1 Institute of Population Health Sciences, National Health Research Institutes, Miaoli County, 2 Program for Aging, College of Medicine, 8 Graduate Institute of Biostatistics, College of Public Health, China Medical University, 3 School of Physical Therapy and Center for Education and Research on Geriatrics and Gerontology, Chung Shan Medical University, 9 Department of Family Medicine, China Medical University Hospital, 11 Department of Health Services Administration, China Medical University and Hospital, Taichung, 4 Department of Family Medicine, National Cheng Kung University Hospital, 5 Institute of Gerontology, National Cheng Kung University Medical College, Tainan, Departments of 6 Geriatrics and Gerontology and 7 Internal Medicine, National Taiwan University Hospital, and 10 Department of Family Medicine, College of Medicine, National Taiwan University, Taipei, Taiwan Aim: To develop cut-off points of muscle mass, gait speed and handgrip strength; and to examine the prevalence of sarcopenia, and the relationship between sarcopenia stages and functional limitations and disability by using these cut-off points. Methods: We pooled individual participant data of 2867 community-dwelling older adults from five cohort studies. We defined the cut-off point of a muscle mass index (ASM/ht 2 ) as the values of two standard deviations below the sex-specific means of a young population or as the 20th percentile of the sex-specific distribution in our study population. The gait speed and handgrip strength cut-off points were defined as the 20th percentile of their population distributions. We also measured functional limitations, using the Short Physical Performance Battery, and the number of activities of daily living and instrumental activities of daily living difficulties. Results: We identified the cut-off points of ASM/ht 2, gait speed and handgrip strength. By applying these cut-off points to our study population, the prevalence of sarcopenia varied from 3.9% (2.5% in women and 5.4% in men) to 7.3% (6.5% in women and 8.2% in men). A higher sarcopenia stage was independently associated with a lower summary performance score, as well as more activities of daily living and instrumental activities of daily living difficulties (P < 0.05 for all). Conclusions: The prevalence of sarcopenia in community-dwelling older adults is comparable with those in other populations. A dose response relationship exists between sarcopenia stages and functional limitations/disability. The European Working Group on Sarcopenia in Older People consensus definition using these cut-off points is suitable for determining sarcopenia cases in the elderly population of Taiwan. Geriatr Gerontol Int 2014; 14 (Suppl. 1): 52 60. Keywords: disability evaluation, epidemiology, muscle, physical function, sarcopenia. Introduction Accepted for publication 5 November 2013. Correspondence: Dr Chih-Cheng Hsu MD DRPH, Institute of Population Health Sciences, National Health Research Institutes, 35 Keyan Road, Zhunan, Miaoli County 35053, Taiwan. Email: cch@nhri.org.tw *Other members of the START Team are listed at the end of this article. The global population is aging, and disabilities later in life are becoming an urgent health issue. 1 Skeletal muscles play a critical role in maintaining function later in life. 2 However, skeletal muscle mass declines with age. 2 Sarcopenia, the age-related loss of skeletal muscle mass and function, is prevalent among older people, and represents a major risk factor of disability, falls, 52 doi: 10.1111/ggi.12193 2014 Japan Geriatrics Society

Sarcopenia in an elderly population in Taiwan mortality, and numerous other adverse health outcomes in older adults. 2 4 Therefore, the early recognition and diagnosis of sarcopenia in a primary care setting can enable identifying and treating vulnerable older adults at an early stage of the disablement process. Operational definitions and diagnostic criteria have recently been proposed to facilitate the prompt recognition of sarcopenia in clinical practice. 3 7 According to the consensus developed by the European Working Group on Sarcopenia in Older People (EWGSOP), the diagnosis of sarcopenia requires the measurement of three variables: muscle mass, physical performance and muscle strength. 6 A wide range of assessment techniques with varying costs, availability and ease of use are available for measuring each of these sarcopenia variables. 6 Assessment techniques that are economical, readily available, reliable and simple facilitate the widespread adoption of current sarcopenia case-finding strategies in different healthcare settings and large-scale community-based epidemiological studies. Bioelectrical impedance analysis, usual gait speed and handgrip strength are measurement methods used to assess muscle mass, physical performance and muscle strength, respectively, and are extremely promising in this regard. 4,6,8,9 Sarcopenia is diagnosed in older adults with low muscle mass plus poor physical performance or muscle strength. 6 The EWGSOP also proposed a staging scheme for sarcopenia. 6 However, whether the cut-off points of the muscle mass estimate, usual gait speed and handgrip strength determined in Western studies could be applied to other populations with different ethnicities or cultures remains unclear. That racial and ethnic differences in body composition and physical function exist is well known. 10 14 The results of previous studies have shown that the proper cut-off points of sarcopenia variables for Asians can differ from those for Westerners. 15 23 In the consensus report, the EWGSOP called for the reference values of sarcopenia variables for populations worldwide. 6 Because sarcopenia is believed to represent a major cause of disability, 6,7 sarcopenia defined using cut-off points should be independently associated with the early indicators of the disablement process, as well as a greater degree of disability among older adults. Specifically, a dose response relationship should exist between sarcopenia stages, and the indicators of disability risk and severity. To meet the urgent need of defining the subnormal values of muscle mass and physical function for a sarcopenia diagnosis in a rapidly aging Asian population, we pooled primary data from five major cohort studies in Taiwan, as part of the Sarcopenia and Translational Aging Research in Taiwan (START) project. This pooling project, with its large sample, provides a unique opportunity to address issues that cannot be addressed sufficiently in any single study. The purpose of the present study was to develop muscle mass, gait speed and handgrip strength cut-off points based on population distributions, which were then used in the EWGSOP diagnostic criteria of sarcopenia to determine its prevalence and relationship with functional health in a large sample of community-dwelling older adults in Taiwan. Methods Participants We analyzed the individual participant data of 2867 community-dwelling older adults, which were collected during baseline examinations of five cohort studies carried out between 2003 and 2012 (Table 1). Details of the designs and participant recruitment of these studies have been published previously. 24 26 All of the cohort studies required written informed consent, and were approved by the respective institutional review boards. Body composition Body composition was measured using an 8-contact electrode bioelectrical impedance analysis (BIA) device (Tanita BC-418, Tanita, Tokyo, Japan), and followed the standard procedure and the manufacturer s instructions. 27 This BIA device was used to measure the wholebody and segmental impedance (± 1Ω) at a frequency of 50 khz, and it provided valid muscle mass estimates (kg) of each of the four extremities. 8 Appendicular muscle mass (ASM) was calculated as the sum of the estimated muscle mass for the arms and legs. A relative skeletal muscle mass index (ASM/ht 2 ) normalized for height was defined as the ratio of ASM (kg) and the height squared (m 2 ). 4,6 For this analysis, we defined muscle mass cut-off points according to the distribution of ASM/ht 2 of a young population comprising 998 healthy adults (aged 20 40 years) 16 or the study population. A participant was considered to have low muscle mass if his or her ASM/ht 2 was below 2 standard deviations of the reference young adult values defined in previous studies (6.76 kg/m 2 for men and 5.28 kg/m 2 for women). 16 Alternatively, based on the definition by Delmonico et al. and Newman et al. 3,5 we examined the sex-specific distribution of ASM/ht 2 in our study population, and participants with ASM/ht 2 in the lowest 20% of the sex-specific distribution were considered to have low muscle mass. Gait speed Participants were observed taking a walk over a short distance at their usual pace, starting from a standing position, and an examiner timed the task by using a 2014 Japan Geriatrics Society 53

I-C Wu et al. Table 1 Characteristics of the cohort studies included in the pooled analysis Mean (SD) body mass index Mean (SD) height Men Mean (SD) weight Mean age at entry Year of starting enrolment Study Study location Setting No. participants n (years) (%) (kg) (cm) (kg/m 2 ) HALST Taipei, Kaohsiung Community 713 2012 73.4 43.6 61.1 (10.5) 158.2 (8.5) 24.4 (3.4) TCHS-E Taichung Community 1042 2009 74.2 51.9 60.9 (10.4) 157.8 (8.0) 24.4 (3.6) IPFCEH Hualien Community 340 2003 74.3 55.9 61.7 (9.6) 159.0 (8.2) 24.4 (3.2) COMDLST Taichung Community 223 2008 74.5 46.6 62.5 (10.6) 158.4 (8.3) 24.9 (3.5) TOP Kaohsiung Community 549 2012 76.0 51.9 58.8 (10.8) 154.9 (8.0) 24.4 (3.8) COMDLST, Community-dwelling Older Adult s Mobility Disability Longitudinal Follow Up Study in Taichung City Development of a Mobility Disability Screening Tool and Plan; HALST, Healthy Aging Longitudinal Study in Taiwan; IPFCEH, Investigation of Physical Function Status of Community Elderly Adults in Hualien city; TCHS-E, Taichung Community Health Study-Elders; TOP, Tianliao Old People Study. handheld stopwatch that measured to the nearest hundredth of a second. The walking distance ranged from 3 to 5 m according to different study sites. The gait speed was calculated as the walking distance (m) divided by time (s). We examined the distribution of the gait speed stratified according to the sex-specific median height. Participants with a gait speed in the lowest 20% of the sex- and height-specific distributions were considered to have low gait speed. 6,28 Handgrip strength Handgrip strength (kg) was measured using standard calibrated hand dynamometers. 24,25 We calculated the average grip strength of each hand, and the average grip strength of the hand with superior performance or that of the dominant hand was used for the analysis. We examined the distribution of the handgrip strength stratified according to sex-specific body mass index (BMI) quartiles. Participants with a handgrip strength in the lowest 20% of the sex- and BMI-specific distributions were considered to have low handgrip strength. 6,28 Sarcopenia We determined the presence of sarcopenia according to the EWGSOP consensus definition. 6 A diagnosis of sarcopenia requires the presence of low muscle mass in addition to a low handgrip strength or a slow gait speed. Participants were considered to have severe sarcopenia if all three of these criteria were present, whereas those with low muscle mass, but without a low handgrip strength or a slow gait speed, were considered to be in the presarcopenia stage. Functional health status We assessed functional limitations by using the Short Physical Performance Battery (SPPB). 29 A participant received a summary performance score on this scale, with a higher score representing better performance. A lower summary performance score predicts disability in a linear fashion across the entire range of values. 29 We also assessed the perceived function in the activities of daily living (ADL) and the instrumental activities of daily living (IADL) by carrying out structured interviews with the Katz Index, the Barthel Index and the Lawton Brody Instrumental Activities of Daily Living Scale, which evaluate the level of dependence in carrying out each task. 30 32 The severity of a disability was defined using the number of ADL (i.e. eating, transferring, personal hygiene, bathing, walking, dressing and continence) and IADL (i.e. shopping, housekeeping, handling finances, food preparation, using transportation, using a telephone, doing the laundry and taking medication) that the person was unable to carry out independently. 33 54 2014 Japan Geriatrics Society

Sarcopenia in an elderly population in Taiwan Covariates Table 2 Cut-off points of the sarcopenia variables The covariates were the participant s age, sex, education level ( 6 years or >6 years), smoking status (smoker or non-smoker), obesity status and comorbidities. Participants who reported smoking cigarettes at the time of the interview were defined as smokers. We calculated the BMI as the bodyweight in kg divided by the square of the height in m 2, and categorized it according to the definition of the Department of Health in Taiwan (BMI 24 26.9 kg/m 2 for overweight and BMI 27 kg/m 2 for obesity). 34 Comorbidities were assessed by referring to the self-reported physician s diagnosis, and included hypertension, diabetes mellitus, stroke, heart disease, arthritis, and chronic obstructive pulmonary disease. Statistical analysis Descriptive statistics were used to characterize the population. All of the continuous variables are presented as the mean ± SD. Differences in continuous variables among the groups were analyzed using oneway analysis of variance (ANOVA), whereas differences in categorical variables (proportions) were analyzed using χ 2 -tests. We examined the relationship between the sarcopenia severity/stage and the functional health measures by applying linear regression analysis. We included the following covariates in the adjusted model: age, sex, education level, smoking status, obesity status, comorbidities and cohorts. We estimated the adjusted values of the summary performance score, and the number of ADL and IADL difficulties according to the sarcopenia severity/stages. In all of the analyses, differences were considered significant when P < 0.05. We calculated 95% confidence intervals (CI), and reported the CI for each parameter estimate. All of the analyses were carried out using SPSS version 19.0 (SPSS, Chicago, IL, USA). Results The mean ± SD value of the participants age was 74 ± 6 years, and 50% of the participants were women. The mean ± SD values of the sarcopenia variables were as follows: ASM/ht 2, 6.38 ± 0.82 kg/m 2 in women, and 8.05 ± 1.15 kg/m 2 in men; gait speed, 0.82 ± 0.26 m/s in women, and 0.90 ± 0.27 m/s in men; and handgrip strength, 20.0 ± 5.4 kg in women, and 32.3 ± 7.4 kg in men. The cut-off points for low muscle mass, low physical performance and low muscle strength are summarized in Table 2. After applying the EWGSOP diagnostic criteria of sarcopenia with muscle mass cut-off points derived from reference young adults, we observed a prevalence of sarcopenia of 2.5% among women and 5.4% among Cut-off points Muscle mass (ASM/ht 2 ) Men 6.76 kg/m 2 7.09 kg/m 2 Women 5.28 kg/m 2 5.70 kg/m 2 Physical performance (gait speed) Men Height 163 cm 0.67 m/s Height >163 cm 0.71 m/s Women Height 152 cm 0.57 m/s Height >152 cm 0.67 m/s Muscle strength (handgrip strength) Men BMI <22.1 kg/m 2 25.0 kg BMI 22.1 24.3 kg/m 2 26.5 kg BMI 24.4 26.3 kg/m 2 26.4 kg BMI >26.3 kg/m 2 27.2 kg Women BMI <22.3 kg/m 2 14.6 kg BMI 22.3 24.2 kg/m 2 16.1 kg BMI 24.3 26.8 kg/m 2 16.5 kg BMI >26.8 kg/m 2 16.4 kg Defined as values two standard deviations below the sex-specific means of the reference young adults aged 20 40 years. Defined as the 20th percentile of the sex-specific distribution in our study population. Defined as the 20th percentile of the sex- and height-specific distribution in our study population. Defined as the 20th percentile of the sexand body mass index (BMI)-specific distribution in our study population. ASM, appendicular muscle mass. men. The corresponding values were 6.5% in women and 8.2% in men when using the muscle mass cut-off points derived from the study population. As expected, the prevalence of sarcopenia increased significantly with age in both sexes (Table 3). Participants with sarcopenia were more likely to be older, lean and diagnosed with a chronic disease (Table 4). Older adults at a more advanced stage of sarcopenia tended to have a lower summary performance score as well as difficulties in more ADL and IADL (Table 4). After adjusting for covariates, the increasing sarcopenia level remained significantly associated with a lower summary performance score, and more ADL and IADL difficulties (Table 5). Using muscle mass cut-off points derived from the reference young adults, we found that participants with severe sarcopenia had a mean of 1.42 more ADL difficulties (P < 0.001), a mean of 1.54 more IADL difficulties (P < 0.001), and their summary performance score was a mean of 2.5 points lower (P = 0.042) compared with 2014 Japan Geriatrics Society 55

I-C Wu et al. Table 3 Prevalence of sarcopenia using European Working Group on Sarcopenia in Older People definitions All Women Men All ages Aged 65 74 years Aged 75 years All ages Aged 65 74 years Aged 75 years n (%) n (%) n (%) n (%) n (%) n (%) n (%) Muscle mass cut-off points derived from reference young population No sarcopenia 1964 (91.1) 1023 (93.0) 629 (94.6) 394 (90.6) 941 (89.2) 535 (92.7) 406 (84.9) Presarcopenia 106 (4.9) 49 (4.5) 29 (4.4) 20 (4.6) 57 (5.4) 30 (5.2) 27 (5.6) Sarcopenia 50 (2.3) 17 (1.5) 6 (0.9) 11 (2.5) 33 (3.1) 10 (1.7) 23 (4.8) Severe sarcopenia 35 (1.6) 11 (1.0) 1 (0.2) 10 (2.3) 24 (2.3) 2 (0.3) 22 (4.6) Any sarcopenia 85 (3.9) 28 (2.5) 7 (1.1) 21 (4.8) 57 (5.4) 12 (2.1) 45 (9.4) Muscle mass cut-off points derived from the study population No sarcopenia 1732 (80.4) 882 (80.2) 542 (81.5) 340 (78.2) 850 (80.6) 496 (86.0) 354 (74.1) Presarcopenia 266 (12.3) 147 (13.4) 98 (14.7) 49 (11.3) 119 (11.3) 60 (10.4) 59 (12.3) Sarcopenia 105 (4.9) 50 (4.5) 22 (3.3) 28 (4.6) 55 (5.2) 19 (3.3) 36 (7.5) Severe sarcopenia 52 (2.4) 21 (1.9) 3 (0.5) 18 (4.1) 31 (2.9) 2 (0.3) 29 (6.1) Any sarcopenia 157 (7.3) 71 (6.5) 25 (3.8) 46 (10.6) 86 (8.2) 21 (3.6) 65 (13.6) Low muscle mass only. Low muscle mass + slower walking speed or low muscle mass + low handgrip strength. Low muscle mass + slower walking speed + low handgrip strength. Sarcopenia + severe sarcopenia. those without sarcopenia. We observed a similar pattern when using the muscle mass cut-off points derived from the study population (Table 5). Discussion In the present pooled analysis of five cohort studies, we identified the cut-off points of a relative skeletal muscle mass index, usual gait speed and handgrip strength, based on the population distributions. The prevalence of sarcopenia among community-dwelling older adults was also estimated. Based on the staging scheme proposed by the EWGSOP, the sarcopenia severity/stage was found to be related to degrees of functional limitations and disabilities in a dose response manner, independent of potential confounders. The ASM/ht 2 cut-off points derived from our reference young population were 6.76 kg/m 2 for men and 5.28 kg/m 2 for women, which are values similar to those in a Japanese population (6.87 kg/m 2 for men and 5.46 kg/m 2 for women), 23 although they were slightly higher than those in a South Korean population (6.58 kg/m 2 for men and 4.59 kg/m 2 for women) 19 and a Chinese population (6.08 kg/m 2 for men and 4.79 kg/m 2 for women). 17 Slight variations in muscle mass distributions among young adults might be owing to differences in early life determinants of peak muscle mass. 35 The use of muscle mass cut-off points derived from healthy young adults is recommended; however, researchers have suggested alternative methods of defining subnormal muscle mass. 3,5,6 The ASM/ht 2 values of 7.09 kg/m 2 in men and 5.70 kg/m 2 in women corresponded to the 20th percentile of the sex-specific distribution in our study population. These values are similar to the ASM/ ht 2 values identified in other populations comprising older adults. 3,5 Newman et al. noted that the ASM/ht 2 values of 7.23 kg/m 2 in men and 5.67 kg/m 2 in women represented the 20th percentile values in a USA population of older adults aged 70 years and older. 5 In general, the muscle mass cut-off points indentified in the present study were congruent with previous studies findings. The mean gait speed in the present study was 0.90 ± 0.27 m/s in men and 0.82 ± 0.26 m/s in women. This is in agreement with the results of studies that had recruited elderly Chinese cohort populations. 12 However, these estimates were lower than those observed in Western populations. 36,37 In a study that recruited 1750 community-dwelling Caucasian older adults aged 70 years and older, men had a mean gait speed of 1.15 m/s, and women had a mean gait speed of 1.08 m/s. 37 The racial/ethnic differences in gait speed values could partly be due to differences in anthropometrics. 11 Because gait speed is known to differ significantly based on a person s sex and height, 36,37 we examined the distributions according to the sex-specific 56 2014 Japan Geriatrics Society

Sarcopenia in an elderly population in Taiwan Table 4 Characteristics of study participants according to sarcopenia status (n = 2155) Characteristics No sarcopenia Presarcopenia Sarcopenia Severe sarcopenia P Muscle mass cut-off points derived from reference young population Age (years) 74.0 (6.0) 74.3 (6.4) 78.6 (7.3) 81.7 (6.2) <.001 Men, n (%) 941 (47.9) 57 (53.8) 33 (66.0) 24 (68.6) <.001 Education.034 6 years, n (%) 1089 (55.4) 42 (39.6) 24 (48.0) 18 (51.4) Smoking, n (%) 140 (7.1) 16 (15.1) 4 (8.0) 3 (8.6).137 BMI <.001 Lean, n (%) 827 (42.2) 99 (93.4) 48 (96.0) 33 (94.3) Overweight, n (%) 673 (34.3) 5 (4.7) 1 (2.0) 1 (2.9) Obesity, n (%) 462 (23.5) 2 (1.9) 1 (2.0) 1 (2.9) Diabetes mellitus, n (%) 367 (18.8) 14 (13.2) 6 (12.0) 5 (14.3).086 Hypertension, n (%) 1034 (53.0) 34 (32.7) 24 (48.0) 18 (51.4).043 Heart disease, n (%) 342 (17.6) 23 (21.7) 21 (42.9) 11 (32.4) <.001 Stroke, n (%) 107 (5.5) 4 (3.8) 1 (2.0) 6 (17.6).197 Arthritis, n (%) 387 (20.1) 9 (8.6) 7 (14.0) 8 (23.5).181 Chronic obstructive pulmonary disease, n (%) 40 (2.1) 7 (6.6) 1 (2.0) 3 (8.6).004 No. ADL difficulties 0.12 (0.44) 0.06 (0.24) 0.16 (0.48) 1.71 (2.37) <.001 No. IADL difficulties 0.20 (0.73) 0.09 (0.38) 0.64 (1.61) 2.14 (2.17) <.001 Summary performance score 10.1 (2.5) 10.5 (1.7) 9.9 (1.8) 7.7 (2.1).010 Muscle mass cut-off points derived from the study population Age (years) 73.9 (6.0) 73.9 (5.8) 77.4 (7.0) 81.4 (6.2) <.001 Men, n (%) 850 (49.1) 119 (44.7) 55 (52.4) 31 (59.6).189 Education <.001 6 years, n (%) 1007 (58.1) 92 (34.6) 45 (42.9) 29 (55.8) Smoking, n (%) 119 (6.9) 34 (12.8) 7 (6.7) 3 (5.8).268 BMI <.001 Lean, n (%) 650 (37.6) 229 (86.1) 89 (84.8) 39 (75.0) Overweight, n (%) 623 (36.0) 33 (12.14) 13 (12.4) 11 (21.2) Obesity, n (%) 457 (26.4) 4 (1.5) 3 (2.9) 2 (3.8) Diabetes mellitus, n (%) 330 (19.2) 37 (13.9) 11 (10.5) 14 (26.9).223 Hypertension, n (%) 927 (53.8) 104 (39.7) 52 (49.5) 27 (51.9).018 Heart disease, n (%) 286 (16.7) 55 (20.7) 39 (37.5) 17 (33.3) <.001 Stroke, n (%) 97 (5.7) 7 (2.6) 7 (6.7) 7 (13.7).297 Arthritis, n (%) 347 (20.4) 34 (12.9) 18 (17.5) 12 (23.5).020 Chronic obstructive pulmonary disease, n (%) 37 (2.2) 10 (3.8) 1 (1.0) 3 (5.8).211 No. ADL difficulties 0.12 (0.45) 0.06 (0.25) 0.19 (0.49) 1.33 (2.01) <.001 No. IADL difficulties 0.20 (0.73) 0.08 (0.39) 0.54 (0.13) 1.63 (2.03) <.001 Summary performance score 10.1 (2.49) 10.8 (1.6) 9.4 (1.8) 5.33 (3.08) <.001 Data are means (SD) unless otherwise specified. Low muscle mass only (n = 106 when muscle mass cut-off points were derived from reference young population; n = 266 when muscle mass cut-off points were derived from the study population). Low muscle mass + slower walking speed or low muscle mass + low handgrip strength (n = 50 when muscle mass cut-off points were derived from reference young population; n = 105 when muscle mass cut-off points were derived from the study population). Low muscle mass + slower walking speed + low handgrip strength (n = 35 when muscle mass cut-off points were derived from reference young population; n = 52 when muscle mass cut-off points were derived from the study population). ADL, activities of daily living; BMI, body mass index; IADL, instrumental activities of daily living. median height. As expected, the gait speed cut-off points corresponding to the 20th percentile of the distributions were typically lower than those identified in Western populations, 6,28 further emphasizing the need for ethnicity- or population-specific norms for physical performance measures. In the present study, older Taiwanese men had a mean handgrip strength of 32.3 ± 7.4 kg, and women had a mean handgrip strength of 20.0 ± 5.4 kg. Both values were lower than those observed in Western populations. 13,14 This is consistent with the findings from other Asian populations. 38 To account for the 2014 Japan Geriatrics Society 57

I-C Wu et al. Table 5 Multivariate analyses testing associations of sarcopenia stages with functional health status 95% CI P for trend Sarcopenia Stages Difference between No sarcopenia Presarcopenia Sarcopenia Severe severe sarcopenia sarcopenia and no sarcopenia Muscle mass cut-off points derived from reference young population Summary performance score 10.1 10.6 9.7 7.6 2.5 4.96, 0.09 0.042 No. ADL difficulties 0.13 0.06 0.14 1.55 1.42 1.23, 1.62 <0.001 No. IADL difficulties 0.20 0.11 0.55 1.74 1.54 1.26, 1.81 <0.001 Muscle mass cut-off points derived from the study population Summary performance score 10.1 10.7 9.3 5.8 4.28 6.0, 2.6 <0.001 No. ADL difficulties 0.13 0.06 0.14 1.16 1.00 0.86, 1.19 <0.001 No. IADL difficulties 0.22 0.08 0.39 1.2 0.98 0.74, 1.21 <0.001 Data are means, and are adjusted for, age, sex, education level, smoking status, obesity, hypertension, diabetes mellitus, stroke, heart disease, arthritis, chronic obstructive pulmonary disease and cohort. Low muscle mass only. Low muscle mass + slower walking speed or low muscle mass + low handgrip strength. Low muscle mass + slower walking speed + low handgrip strength. ADL, activities of daily living; BMI, body mass index; CI, confidence interval; IADL, instrumental activities of daily living. effects of sex and BMI, 13,39 the handgrip strength distributions were stratified according to the sex-specific BMI quartiles, and we identified the 20th percentile of the distribution in each stratum as the handgrip strength cut-off point. Similarly to the gait speed, the handgrip strength cut-off points identified in the present study were lower compared with those obtained in Western studies. 6,28 According to the EWGSOP consensus definition, the prevalence of sarcopenia in our study population varied from 3.9% when using muscle mass cut-off points derived from a young population to 7.3% when using muscle mass cut-off points derived from our large sample of community-dwelling older adults. In addition, the sarcopenia prevalence increased with age, with the prevalence reaching 13.6% among older men aged 75 years and older. These findings are congruent with the results of recent studies that have applied the EWGSOP operational definition of sarcopenia in different populations. Lee et al. noted a prevalence of 7.8% in a sample of 408 elderly Taiwanese participants aged 65 years and older. 15 In an elderly UK population with a mean age of 67 years, 4.6% of men and 7.9% of women were found to have sarcopenia. 40 A recent study observed a prevalence of 5% in a USA population aged 70 79 years. 41 Legrand et al. showed that the prevalence of sarcopenia was 12.5% in a European population aged 80 years and older. 42 Our study showed that the sarcopenia severity/stage is independently associated with the summary performance score, and the number of ADL and IADL difficulties. Older adults with a higher stage of sarcopenia tended to have a lower summary performance score, and more ADL and IADL difficulties. The observed dose response relationship between the sarcopenia stages, and the indicators of disability risk and severity suggested a gradient of an increasing risk of disability development and progression through these stages. These results are in agreement with the working definition of sarcopenia, as well as the current understanding of its progression. 6,41 The present study has a number of strengths. We used a large sample of individual participant data from multiple diverse populations of community-dwelling older adults in this pooled analysis. We used techniques that are simple, economical, and practical for measuring muscle mass, muscle strength, and physical performance. This study is one of the few attempts to define the cut-off points of all three sarcopenia variables, estimate the prevalence of sarcopenia in a rapidly aging Asian population in accordance with the consensus developed by the EWGSOP, and examine the relationship between the sarcopenia stages and the functional health measures. Despite these strengths, the present results should be interpreted with caution because of several potential limitations. This was a cross-sectional 58 2014 Japan Geriatrics Society

Sarcopenia in an elderly population in Taiwan study, thus, we were unable to establish causal relationships. Residual confounding by the unmeasured variables could not be excluded completely. In assessing the disease status, we relied chiefly on self-reports regarding the presence of a disease. The participating study cohorts mainly consisted of community-dwelling older adults who might be, overall, healthier than the agematched general population. Therefore, the findings might not be generalizable to the entire elderly population. Similarly to the commonly used cut-off points for defining sarcopenia, 6,28 the gait speed and handgrip strength cut-off points in the present study were developed based on their distributions in populations consisting of older adults. It would be of great interest for future studies to examine the distributions of gait speed and handgrip strength of a young population in Taiwan, and compare the gait speed and handgrip strength cutoff points derived from healthy young adults with those derived from other reference populations. Gait speed was measured across varying distances in the present study (3 m [n = 181], 4 m [n = 734], 4.572 m [n = 533] and 5 m [n = 1012]). A total of 7.4% of the participants received gait speed measurement over a distance of 3 m. Some researchers questioned the validity of measuring gait speed in less than 4 m, particularly when using a static start. 43 More research is required to standardize gait speed measurement in our population. In conclusion, the present pooled analysis provided population-specific cut-off points for low muscle mass, low physical performance and low muscle strength. After applying the sarcopenia variable cut-off points to a large sample of community-dwelling older adults, we observed prevalence estimates for sarcopenia that were comparable with those in other populations. In addition, sarcopenia severity was positively and independently associated with the degrees of functional limitations and disabilities. We recommend that the cut-off points be used in future large-scale communitybased studies examining the predictive ability of sarcopenia for multiple adverse health outcomes in older adults, as well as international comparisons on the prevalence of sarcopenia. These efforts will foster a broader adoption of sarcopenia assessments in primary care settings, and ultimately improve healthcare for older adults. Acknowledgements We thank Ming-Shiang Wu and Yi-Fen Tsai for their assistance in data management and statistical analysis. Other members of the START project are as follows: Dr Meng-Chih Lee, Center for Education and Research on Geriatrics and Gerontology and School of Medicine, Chung Shan Medical University, Taiwan; Dr Chia-Ing Li, China Medical University; Ms Ching-I Chang, Mr Ming-Shiang Wu and Ms Yi-Fen Tsai, National Health Research Institutes. Disclosure statement The authors declare no conflict of interest. References 1 Jagger C, Gillies C, Moscone F et al. Inequalities in healthy life years in the 25 countries of the European Union in 2005: a cross-national meta-regression analysis. Lancet 2008; 372: 2124 2131. 2 Morley JE. Sarcopenia: diagnosis and treatment. J Nutr Health Aging 2008; 12: 452 456. 3 Delmonico MJ, Harris TB, Lee JS et al. Alternative definitions of sarcopenia, lower extremity performance, and functional impairment with aging in older men and women. J Am Geriatr Soc 2007; 55: 769 774. 4 Janssen I, Baumgartner RN, Ross R, Rosenberg IH, Roubenoff R. Skeletal muscle cutpoints associated with elevated physical disability risk in older men and women. Am J Epidemiol 2004; 159: 413 421. 5 Newman AB, Kupelian V, Visser M et al. Sarcopenia: alternative definitions and associations with lower extremity function. J Am Geriatr Soc 2003; 51: 1602 1609. 6 Cruz-Jentoft AJ, Baeyens JP, Bauer JM et al. Sarcopenia: European consensus on definition and diagnosis: report of the European Working Group on Sarcopenia in Older People. Age Ageing 2010; 39: 412 423. 7 Fielding RA, Vellas B, Evans WJ et al. Sarcopenia: an undiagnosed condition in older adults. Current consensus definition: prevalence, etiology, and consequences. International working group on sarcopenia. J Am Med Dir Assoc 2011; 12: 249 256. 8 Pietrobelli A, Rubiano FS, Onge MP, Heymsfield SB. New bioimpedance analysis system: improved phenotyping with whole-body analysis. Eur J Clin Nutr 2004; 58: 1479 1484. 9 Abellan van Kan G, Rolland Y, Andrieu S et al. Gait speed at usual pace as a predictor of adverse outcomes in community-dwelling older people an International Academy on Nutrition and Aging (IANA) Task Force. J Nutr Health Aging 2009; 13: 881 889. 10 WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 2004; 363: 157 163. 11 Sims EL, Keefe FJ, Kraus VB, Guilak F, Queen RM, Schmitt D. Racial differences in gait mechanics associated with knee osteoarthritis. Aging Clin Exp Res 2009; 21: 463 469. 12 Chu LW, Chiu AY, Chi I. Impact of falls on the balance, gait, and activities of daily living functioning in community-dwelling Chinese older adults. J Gerontol A Biol Sci Med Sci 2006; 61: 399 404. 13 Massy-Westropp NM, Gill TK, Taylor AW, Bohannon RW, Hill CL. Hand Grip Strength: age and gender stratified normative data in a population-based study. BMC Res Notes 2011; 4: 127. 14 Gunther CM, Burger A, Rickert M, Crispin A, Schulz CU. Grip strength in healthy caucasian adults: reference values. J Hand Surg [Am] 2008; 33: 558 565. 15 Lee WJ, Liu LK, Peng LN, Lin MH, Chen LK. Comparisons of sarcopenia defined by IWGS and EWGSOP criteria 2014 Japan Geriatrics Society 59

I-C Wu et al. among older people: results from the I-Lan longitudinal aging study. J Am Med Dir Assoc 2013; 14: 528 e1 528 e7. 16 Chang CI, Chen CY, Huang KC, Wu CH, Hsiung CA, Hsu CC. Comparison of three BIA muscle indices for sarcopenia screening in old adults. Eur Geriatr Med 2013; 4: 145 149. 17 Cheng Q, Zhu X, Zhang X et al. A cross-sectional study of loss of muscle mass corresponding to sarcopenia in healthy Chinese men and women: reference values, prevalence, and association with bone mass. J Bone Miner Metab 2013; doi: 10/1007/s00774-013-0468-3. 18 Lau EM, Lynn HS, Woo JW, Kwok TC, Melton LJ, 3rd. Prevalence of and risk factors for sarcopenia in elderly Chinese men and women. J Gerontol A Biol Sci Med Sci 2005; 60: 213 216. 19 Kim YS, Lee Y, Chung YS et al. Prevalence of sarcopenia and sarcopenic obesity in the Korean population based on the Fourth Korean National Health and Nutritional Examination Surveys. J Gerontol A Biol Sci Med Sci 2012; 67: 1107 1113. 20 Lim S, Kim JH, Yoon JW et al. Sarcopenic obesity: prevalence and association with metabolic syndrome in the Korean Longitudinal Study on Health and Aging (KLoSHA). Diabetes Care 2010; 33: 1652 1654. 21 Kim TN, Yang SJ, Yoo HJ et al. Prevalence of sarcopenia and sarcopenic obesity in Korean adults: the Korean sarcopenic obesity study. Int J Obes (Lond) 2009; 33: 885 892. 22 Tanimoto Y, Watanabe M, Sun W et al. Association between sarcopenia and higher-level functional capacity in daily living in community-dwelling elderly subjects in Japan. Arch Gerontol Geriatr 2012; 55: e9 13. 23 Sanada K, Miyachi M, Tanimoto M et al. A cross-sectional study of sarcopenia in Japanese men and women: reference values and association with cardiovascular risk factors. Eur J Appl Physiol 2010; 110: 57 65. 24 Lin CC, Li CI, Chang CK et al. Reduced health-related quality of life in elders with frailty: a cross-sectional study of community-dwelling elders in Taiwan. PLoS One 2011; 6: e21841. 25 Wang CY, Chen LY. Grip strength in older adults: testretest reliability and cutoff for subjective weakness of using the hands in heavy tasks. Arch Phys Med Rehabil 2010; 91: 1747 1751. 26 Chang CS, Chang YF, Wang MW et al. Inverse relationship between central obesity and osteoporosis in osteoporotic drug naive elderly females: the Tianliao Old People (TOP) Study. J Clin Densitom 2013; 16: 204 211. 27 Organ LW, Bradham GB, Gore DT, Lozier SL. Segmental bioelectrical impedance analysis: theory and application of a new technique. J Appl Physiol 1994; 77: 98 112. 28 Fried LP, Tangen CM, Walston J et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56: M146 M156. 29 Guralnik JM, Ferrucci L, Simonsick EM, Salive ME, Wallace RB. Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability. N Engl J Med 1995; 332: 556 561. 30 Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of illness in the aged. The index of ADL: a standardized measure of biological and psychosocial function. JAMA 1963; 185: 914 919. 31 Mahoney FI, Barthel DW. Functional evaluation: the barthel index. Md State Med J 1965; 14: 61 65. 32 Lawton MP, Brody EM. Assessment of older people: selfmaintaining and instrumental activities of daily living. Gerontologist 1969; 9: 179 186. 33 Hardy SE, Gill TM. Recovery from disability among community-dwelling older persons. JAMA 2004; 291: 1596 1602. 34 Pan WH, Flegal KM, Chang HY, Yeh WT, Yeh CJ, Lee WC. Body mass index and obesity-related metabolic disorders in Taiwanese and US whites and blacks: implications for definitions of overweight and obesity for Asians. Am J Clin Nutr 2004; 79: 31 39. 35 Sayer AA, Syddall H, Martin H, Patel H, Baylis D, Cooper C. The developmental origins of sarcopenia. J Nutr Health Aging 2008; 12: 427 432. 36 Bohannon RW. Comfortable and maximum walking speed of adults aged 20 79 years: reference values and determinants. Age Ageing 1997; 26: 15 19. 37 Hollman JH, McDade EM, Petersen RC. Normative spatiotemporal gait parameters in older adults. Gait Posture 2011; 34: 111 118. 38 Wu SW, Wu SF, Liang HW, Wu ZT, Huang S. Measuring factors affecting grip strength in a Taiwan Chinese population and a comparison with consolidated norms. Appl Ergon 2009; 40: 811 815. 39 Schlussel MM, dos Anjos LA, de Vasconcellos MT, Kac G. Reference values of handgrip dynamometry of healthy adults: a population-based study. Clin Nutr 2008; 27: 601 607. 40 Patel HP, Syddall HE, Jameson K et al. Prevalence of sarcopenia in community-dwelling older people in the UK using the European Working Group on Sarcopenia in Older People (EWGSOP) definition: findings from the Hertfordshire Cohort Study (HCS). Age Ageing 2013; 42: 378 384. 41 Murphy RA, Ip EH, Zhang Q et al. Transition to sarcopenia and determinants of transitions in older adults: a population-based study. J Gerontol A Biol Sci Med Sci 2013; doi: 10.1093/gerona/glt131. 42 Legrand D, Vaes B, Mathei C, Swine C, Degryse JM. The prevalence of sarcopenia in very old individuals according to the European consensus definition: insights from the BELFRAIL study. Age Ageing 2013; doi: 10.1093/ageing/aft/ aft128. 43 Guralnik JM, Ferrucci L, Pieper CF et al. Lower extremity function and subsequent disability: consistency across studies, predictive models, and value of gait speed alone compared with the short physical performance battery. J Gerontol A Biol Sci Med Sci 2000; 55: M221 231. 60 2014 Japan Geriatrics Society