Abstract
Objectives
Despite the recognized impact of intrinsic capacity (IC) impairment on healthy aging, international comparisons in different sociocultural contexts are scarce. This study aimed to compare IC impairment among community-dwelling older adults in Japan and Taiwan to explore the context of healthy aging in different countries.
Design
Comparative observational study.
Setting
National Institute for Longevity Sciences-Longitudinal Study of Aging (NILS-LSA) in Japan and Longitudinal Aging Study of Taipei (LAST) in Taiwan.
Participants
794 individuals (age range, 60.0–86.5 years) from NILS-LSA and 1,358 (60.0–96.7 years) from LAST.
Measurements
IC impairment was evaluated across the domains of locomotion, cognition, vitality, sensory capacity, and psychological well-being. Participants were categorized as having impaired IC or healthy. We investigated associations between IC impairment, falls, and all-cause mortality.
Results
IC impairment was present in 54.9% and 37.3% of participants in the NILS-LSA and LAST cohorts, respectively. Male NILS-LSA participants with impaired IC (odds ratio [OR]:1.50, 95% confidence interval [CI]:1.03–2.20), with hearing loss (OR:1.98, 95% CI:1.00–3.90) were more likely to fall. In LAST, impaired locomotion (OR:2.14, 95% CI:1.46–3.14) increased the risk of falls. Men with impaired IC (hazard ratio [HR]; 2.14, 95% CI.10–4.15) and visual impairment (HR:2.21, 95% CI:1.15–4.25) and women with impaired psychological well-being (HR:4.94, 95% CI:1.28–18.97) in the NILS-LSA cohort had greater risk for all-cause mortality; however, this was not shown for LAST participants.
Conclusion
The prevalence and distribution of IC impairment and associated biomarkers differed significantly between participants in Japan and Taiwan. However, the associations with adverse outcomes remained similar, emphasizing the need for tailored interventions for healthy aging.
Key words: Intrinsic capacity, healthy aging, community-dwelling older individuals, Japan, Taiwan
Introduction
The combined effects of accelerated rise in the proportion of the aging population and the upsurge in disability-adjusted life years present significant challenges to the societies. Hence, the World Health Organization (WHO) released the World Report on Health and Aging, which defines healthy aging as the ongoing process of acquiring and preserving the functional capacity to promote well-being in older adulthood (1). In 2017 and 2019, the WHO introduced the Integrated Care for Older People (ICOPE) guidelines (2), with a handbook delineating person-centered assessments and pathways within primary care settings (3). To maximize healthy life expectancy of older individuals, the ICOPE places significant emphasis on a comprehensive structural pathway encompassing person-centered screening procedures, assessments and planning, referral pathways, and caregiver support (4). This pathway begins with the evaluation of intrinsic capacity (IC), including cognition, locomotion, vitality, vision and hearing, and psychological well-being. Since its inception, studies have delved into the biological and functional characteristics of IC (5), as well as the epidemiology surrounding IC impairments and consequent clinical outcomes (6, 7).
A community-based study examining the implementation of ICOPE in France revealed a considerable prevalence of IC impairment among older adults living in the community (>90%), with most recommended care plan interventions primarily associated with locomotion, vitality, and cognition domains (8). Empirical evidence from China has demonstrated that lower IC scores are significantly associated with diminished physical and mental functions in middle-aged and older adults (9, 10). Furthermore, the ICOPE screening tool has been shown to be effective in detecting interrelated IC impairments across multiple domains among older adults in France who experienced memory declines (11).
However, the aforementioned studies predominantly employed cross-sectional designs, and only one longitudinal study has evaluated the predictive value of ICOPE screening tool for the falls over a 2-year period (12). Moreover, international comparative studies exploring IC impairment trends, associated factors, biomarkers, and clinical outcomes are lacking despite the global implementation aims of ICOPE during the United Nations Decade of Healthy Aging.
To the best of our knowledge, only one study has explored the relationship between IC impairment and health outcomes in different countries. This study included cohorts from Cuba, the Dominican Republic, Puerto Rico, Venezuela, Peru, Mexico, India, and China, and revealed a significant prevalence of impaired IC in the older population, indicating an increased risk of dependence and mortality associated with IC impairment (13). However, comparative studies among older populations in developed countries are lacking. Thus, the primary objective of this study was to perform an international comparative investigation between Japan and Taiwan, two countries with distinct sociocultural contexts, despite geographical proximity. The study included community-dwelling older adults with similar educational levels and aimed to explore potential disparities in epidemiology and clinical outcomes using two longitudinal aging cohorts. By adopting a «glocal» (global and local) approach, this research sought to contribute to the promotion of healthy aging.
Methods
Study Population
The present study utilized two prospective cohort studies, namely National Institute for Longevity Sciences-Longitudinal Study of Aging (NILS-LSA) in Japan and Longitudinal Aging Study of Taipei (LAST) in Taiwan, the comprehensive descriptions of which can be found elsewhere (14, 15). Individuals from both cohorts were subjected to comprehensive questionnaires encompassing various lifestyle factors, including medical history, physical activity, dietary habits, smoking and alcohol consumption, educational background, and marital and family statuses. Additionally, physical and cognitive function assessments, blood tests, anthropometric measurements, and body composition analyses were performed for all participants. Subsequent follow-up surveys were performed to assess health outcomes, such as falls and all-cause mortality.
In the NILS-LSA, participants were selected through age- and sex-stratified random sampling from the institute's surrounding neighborhoods in Obu and Higashiura towns, Aichi Prefecture, Japan. The initial wave of the NILS-LSA, conducted between November 1997 and April 2000, included 2,267 individuals aged 40–79. Follow-up assessments were conducted every two years, with new age- and sex-matched participants randomly recruited to replace those who were unable to attend subsequent investigations. The Committee on the Ethics of Human Research of the National Center for Geriatrics and Gerontology approved the study protocol (No. 899–6), and written informed consent was obtained from all participants.
In LAST, community residents aged 50 years and older were recruited from Taipei and New Taipei City and two waves of surveys were conducted. The first wave, spanning from May 2016 to December 2020, included 1,534 participants who were subsequently included three years later in the second wave. For all participants, follow-ups for health outcomes occurred via telephone every three months. The study protocol was approved by the Institutional Review Board of the National Yang Ming University (YM104121F-5), and all participants provided written informed consent after receiving detailed explanations of the study from the research staff.
For this study, participants from the NILS-LSA were selected from the fifth (July 2006 to July 2008) and sixth (July 2008 to July 2010) waves owing to the comprehensive survey data available in the fifth wave and a follow-up interval similar to that in LAST. For NILS-LSA and LAST, of the initial 2,419 and 1,534 baseline participants, respectively, exclusions were made for age<60 years, dropouts, and incomplete data, resulting in a final analysis of 794 individuals (407 men and 387 women) aged 60.0 to 86.5 years and 1,358 individuals (462 men and 896 women) aged 60.0-96.7 years (Figure 1), respectively.
Figure 1.
Flowchart of the comparative study for intrinsic capacity impairment of NILS-LSA and LAST
IC, intrinsic capacity; LAST, Longitudinal Aging Study of Taipei; NILS-LSA, National Institute for Longevity Sciences-Longitudinal Study of Aging
Definition of IC Impairments
According to the WHO ICOPE guidelines, impaired IC is classified within the domains of cognition, locomotion, vitality, vision and hearing, and psychological well-being. Cognitive impairment was defined using seven items from the Mini-Mental State Examination (MMSE) used in the NILS-LSA (four related to time and place orientation and three related to memory) and nine items from the Montreal Cognitive Assessment (MoCA) used in the LAST (four related to time and place orientation and five related to memory) (16). Participants in the NILS-LSA who demonstrated impairment in any of the seven MMSE items were classified with cognitive impairment, and participants in the LAST were considered to have similar cognitive impairment if they exhibited impairment in orientation or correctly recalled fewer than three words in the five-word recall test.
Impaired locomotion was operationally defined as slow gait speed in the NILS-LSA and abnormal performance in the Five Times Sit-to-Stand Test in the LAST, using the cutoff values established by the Asian Working Group for Sarcopenia (17). Participants in the NILS-LSA were classified with vitality impairment if they exhibited either a weight loss of ≥5% over a 2-year period or lack of appetite, as assessed using the Center for Epidemiologic Studies Depression (CES-D) Scale (18, 19). In LAST, vitality impairment was defined as either weight loss of more than 3 kg in the last 3 months or a lack of appetite, assessed through the question «Have you eaten less in the past three months due to poor appetite?» with response options including «severe appetite loss,» «moderate appetite loss,» and «no change in appetite,» and loss of appetite was defined as a response indicating «severe loss of appetite» and «moderate appetite loss.»
Participants in NILS-LSA were considered to have visual impairment if they rated their visual acuity as «poor» or «very poor», whereas in LAST, individuals were classified as having visual impairment if they reported that visual decline negatively affected their daily activities. In the NILS-LSA, hearing loss was assessed by measuring the air-conduction pure-tone thresholds for both ears using diagnostic audiometers in a soundproof booth. Individuals were classified as having hearing loss if the average threshold level at frequencies of 0.5, 1, 2, and 4 kHz exceeded 35 dB in the better ear. In LAST, participants were asked about their self-perception of hearing loss, and were categorized with hearing loss if they perceived that it affected their daily lives.
Both cohorts employed CES-D Scale (18, 19) to evaluate depressive symptoms within the domain of psychological well-being, with a score of ≤15 indicating normal status and ≤16 indicating significant depressive symptoms. In accordance with the aims of this investigation, individuals with impairments in any of the six domains were designated with IC impairment, whereas those without such impairments were categorized as healthy (with optimal IC).
Clinical Outcomes
Both NILS-LSA and LAST incorporated follow-up assessments to capture fall incidence and time to death among participants. In NILS-LSA, falls occurring within the past year were queried during the sixth-wave follow-up survey, and binary responses of «yes» or «no» were recorded. Mortality data encompassing the period from the baseline survey date until December 31, 2017, were obtained from a Vital Statistics database. In LAST, telephone interviews were conducted every three months for three years following the baseline survey to monitor falls. Mortality information was also tracked for three years, and the date of death was verified with the next of kin.
Other Measurements
In both baseline surveys, a comprehensive array of data was collected, encompassing medical history, lifestyle behaviors, socioeconomic status, physical examination results, body composition, physiological function, physical function, and blood parameters. Medical history (stroke, hypertension, heart disease, diabetes mellitus, kidney disease, and osteoporosis; yes or no for each), smoking status (current or non-current), alcohol consumption status (current or non-current), education level (≤9, 10–12, or ≥13 years), marital status (married or other), and living status (alone or other) were documented using a self-administered questionnaire.
In NILS-LSA, the 24-hour total physical activity was assessed by the Metabolic Equivalent of Task (MET) score (METs-h/day; continuous), which was obtained from participant interviews conducted by trained interviewers using a semi-quantitative assessment (20). In LAST, the 24-hour total physical activity was assessed using the Taiwanese version of the International Physical Activity Questionnaire (short format) (METs-h/day; continuous) (21, 22).
Body mass index (BMI) was calculated using the formula: BMI = weight (kg) / height2 (m2). Height and weight were measured using digital scales with the participants wearing light clothing and no shoes. Calf circumference (cm) was measured as the maximum of either the left or right calf circumference (depending on the participant's dominant hand) in the LAST or the maximum of both left and right calf circumference in the NILS-LSA. Appendicular skeletal muscle mass (kg) and body fat percentage (%) were estimated using dual-energy X-ray absorptiometry (DXA; QDR-4500; Hologic, Bedford, MA, USA) in NILS-LSA and InBody S10 (InBody Japan Inc., Tokyo, Japan) in LAST. The relative appendicular skeletal muscle mass (RASM) (kg/m2) was calculated as follows: RASM = appendicular skeletal muscle mass (kg)/height2 (m2).
Systolic blood pressure and heart rate (bpm) were measured using an automated sphygmomanometer (BP-203RVII, Omron Colin, Tokyo, Japan) after participants had been seated comfortably for at least 5 minutes in the NILS-LSA (23) and measured twice with at least a 1-minute interval, then averaged using an automated sphygmomanometer (OMRON HEM-7320; OMRON HEALTHCARE Co., Ltd, Kyoto, Japan) in LAST. Maximum walking speed (m/s) was measured using the fastest gait in a 10-m walk test in the NILS-LSA and a 6-m walk test in the LAST. In the NILS-LSA, hand grip strength (kg) was the maximum grip strength over four trials for both hands (two trials for each hand) using a dynamometer (Takei T.K.K.5401 and T.K.K.4301a; Takei Scientific Instruments Co., Ltd, Tokyo, Japan), while in the LAST, hand grip strength (kg) was measured using a dynamometer (TTM, Tokyo, Japan) with the elbow positioned at a 90° angle, and the maximum grip strength over three trials for either the right or left hand (depending on the participant's dominant hand) was recorded.
In the NILS-LSA, venous blood was sampled in the morning immediately after fasting for ≥12 hours using ethylenediaminetetraacetic acid (disodium salt, 50 mM) tubes; in LAST, venous blood was sampled in the morning immediately after fasting for ≥8 hours using BD Vacutainer® SST™ Tubes, BD Vacutainer® EDTA Tubes, BD Vacutainer® Fluoride Tubes, and BD Vacutainer® Specialty Tubes (BD, Franklin Lakes, NJ, USA). Blood components, including total cholesterol (mg/dL), triglycerides (mg/dL), high-density lipoprotein cholesterol (mg/dL), hemoglobin A1c (%), alanine transaminase (glutamic-pyruvic transaminase;) (U/L), albumin (g/dL), uric acid (mg/dL), hemoglobin (g/dL), 25-hydroxycholecalciferol (ng/mL), dehydroepiandrosterone sulfate (µg/dL), fasting glucose (mg/dL), insulin (mIU/L), and creatinine (mg/dL) were measured.
The homeostatic model assessment-insulin resistance was calculated as fasting glucose [mg/dL] × insulin [mIU/L]/405 (24). Estimated glomerular filtration rate (eGFR) (mL/min/1.73 m2) was calculated based on sex-age-specified serum creatinine (scr); for men, if serum creatinine (mg/dL) ≥0.9, eGFR = 142 × (scr/0.9)−0.302 × 0.9938age and if serum creatinine (mg/dL) >0.9, eGFR = 142 × (scr/0.9)−1.200 × 0.9938age; for women, if serum creatinine (mg/dL) ≥0.7, eGFR = 142 × (scr/0.7)−0.241 × 0.9938age × 1.012 and if serum creatinine (mg/dL) >0.7, eGFR = 142 × (scr/0.7)−1.200 × 0.9938age × 1.012 (25).
Statistical Analysis
Differences in mean values (continuous variables) and proportions (categorial variables) were assessed using a general linear model and the chi-square test, respectively. All subsequent analyses used a reference group of healthy participants (without IC impairment). Logistic regression models were used to estimate odds ratios (ORs) and corresponding 95% confidence intervals (CIs) for IC impairment in relation to falls over a period of 2 years in NILS-LSA and 3 years in LAST. Considering age and sex as the most influential factors, Model 1 was adjusted for age at baseline and sex. Model 2 was further adjusted for demographic and socioeconomic variables, which showed statistical differences between the impaired IC and healthy groups in the analysis of baseline variables. Model 3 was additionally adjusted for baseline physiological and biochemical indicators which showed statistical differences between the two groups. Cox proportional hazards models were employed to analyze hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) for IC impairments in relation to death over a period of 11.5 years in NILS-LSA and 3 years in LAST. These models were adjusted for the same covariates as those used in the logistic regression models. All statistical analyses were performed using R software (version 4.2.1; R Foundation for Statistical Computing, Vienna, Austria) and R Studio (version 2022.07.1 +554). Two-sided exact P-values below 0.05 were considered statistically significant.
Results
IC Impairments and Associated Characteristics at Baseline
In the baseline survey, the prevalence of IC impairment in the NILS-LSA and LAST cohorts were 54.9% and 37.3%, respectively. In the NILS-LSA cohort, the most common IC impairment was the vitality domain, followed by hearing, cognition, visual, and psychological well-being. Conversely, in the LAST cohort, IC impairment was primarily evident in cognition and locomotion with a considerably lower proportion of impairments in other IC domains. Sex differences were observed in the prevalence of impaired locomotion and hearing loss among participants in the NILS-LSA cohort, whereas sex differences were manifested specifically in cognitive impairment and hearing loss in the LAST cohort (Table 1).
Table 1.
Proportion of impairments in IC domains at baseline in NILS-LSA and LAST
| Impairments of IC domains | Proportion (%) | |||||||
|---|---|---|---|---|---|---|---|---|
| NILS-LSA | LAST | |||||||
| All participants (n = 794) | Men (n = 407) | Women (n = 387) | P-valuea | All participants (n = 1,358) | Men (n = 462) | Women (n = 896) | P-valuea | |
| Cognition | 14.5 | 16.7 | 12.1 | 0.069 | 21.9 | 26.6 | 19.5 | 0.003 |
| Locomotion | 2.0 | 1.0 | 3.1 | 0.044 | 13.0 | 10.8 | 14.2 | 0.082 |
| Vitality | 22.0 | 21.6 | 22.5 | 0.770 | 4.1 | 3.9 | 4.2 | 0.762 |
| Sensory (visual) | 12.5 | 14.3 | 10.6 | 0.120 | 1.3 | 1.1 | 1.5 | 0.574 |
| Sensory (hearing) | 17.5 | 21.6 | 13.2 | 0.002 | 3.8 | 7.1 | 2.1 | < 0.001 |
| Psychological well-being | 11.7 | 11.1 | 12.4 | 0.556 | 2.4 | 1.9 | 2.6 | 0.476 |
a. The χ2 test is used; IC, intrinsic capacity; LAST, Longitudinal Aging Study of Taipei; NILS-LSA, National Institute for Longevity Sciences-Longitudinal Study of Aging
Table 2 presents the baseline characteristics of participants in the NILS-LSA and LAST cohorts. In the NILS-LSA cohort, individuals with impaired IC were characterized by advanced age, a higher proportion of men, and a history of stroke and heart disease than healthy participants. They also exhibited a higher prevalence of current smoking, living alone, lower physical activity, lower BMI, calf circumference, and body fat percentage, slower walking speed, lower MMSE scores, lower total cholesterol levels, and lower estimated glomerular filtration rate (eGFR). Moreover, they were less likely to have completed 13 years or more of education and to be married. Similarly, in the LAST cohort, participants with IC impairment were older and more likely to be men. They had a higher prevalence of stroke, hypertension, and heart disease, lower physical activity, higher BMI, higher systolic blood pressure slower maximum walking speed, lower MoCA scores, lower serum levels of total cholesterol, high-density lipoprotein cholesterol, albumin, and dehydroepiandrosterone sulfate, higher homeostatic model assessment of insulin resistance and glycated hemoglobin, and lower eGFR. Furthermore, they were less likely to be current alcohol drinkers and to have completed 13 years or more of education.
Table 2.
Comparisons of baseline characteristics between participants in the NILS-LSA (n = 794) and LAST (n = 1,358)
| NILS-LSA | LAST | |||||
|---|---|---|---|---|---|---|
| Impaired IC (n = 436) | Healthy (n = 358) | P-valuea | Impaired IC (n = 506) | Healthy (n = 852) | P-valuea | |
| Age (years); mean ± SD | 71.1 ± 6.9 | 68.3 ± 5.7 | < 0.001 | 71.1 ± 6.2 | 68.7 ± 4.4 | < 0.001 |
| Men; % | 54.7 | 47.1 | 0.039 | 37.8 | 31.9 | 0.027 |
| Medical history (yes); % | ||||||
| Stroke | 7.3 | 3.6 | 0.027 | 5.3 | 2.9 | 0.026 |
| Hypertension | 42.0 | 37.7 | 0.223 | 42.9 | 31.3 | < 0.001 |
| Heart disease | 8.5 | 3.9 | 0.011 | 10.3 | 6.7 | 0.019 |
| Diabetes mellitus | 10.6 | 10.1 | 0.820 | 16.2 | 12.7 | 0.070 |
| Kidney disease | 3.7 | 4.7 | 0.450 | 7.3 | 5.8 | 0.253 |
| Osteoporosis | 10.8 | 7.8 | 0.158 | 23.3 | 21.0 | 0.319 |
| Current smoker; % | 13.3 | 7.8 | 0.014 | 3.0 | 3.3 | 0.743 |
| Current alcohol consumer; % | 49.8 | 52.0 | 0.540 | 54.5 | 62.1 | 0.006 |
| Total physical activity (METs-h/day); mean ± SD | 34.2 ± 3.5 | 34.8 ± 2.9 | 0.007 | 31.4 ± 24.8 | 36.3 ± 28.8 | 0.002 |
| Education level (years), % | ||||||
| ≤ 9 | 32.8 | 23.5 | 0.004 | 20.2 | 12.8 | < 0.001 |
| 10–12 | 41.1 | 45.5 | 25.9 | 23.5 | ||
| ≥ 13 | 26.1 | 31.0 | 54.0 | 63.7 | ||
| Marital status (married); % | 77.8 | 85.8 | 0.004 | 73.5 | 78.1 | 0.057 |
| Living status (alone); % | 11.9 | 7.5 | 0.042 | 13.2 | 15.0 | 0.555 |
| Physical examination & body composition | ||||||
| BMI (kg/m2); mean ± SD | 22.6 ± 2.9 | 23.1 ± 2.9 | 0.017 | 23.9 ± 3.3 | 23.5 ± 3.1 | 0.015 |
| Calf circumference (cm); mean ± SD | 33.8 ± 2.7 | 34.3 ± 2.5 | 0.005 | 34.4 ± 3.1 | 34.5 ± 3.2 | 0.615 |
| Appendicular skeletal muscle mass (kg); mean ± SD | 16.4 ± 3.7 | 16.7 ± 3.8 | 0.351 | 17.0 ± 4.2 | 16.6 ± 4.0 | 0.099 |
| Relative appendicular skeletal muscle (kg/m2)b; mean ± SD | 6.6 ± 1.0 | 6.6 ± 1.0 | 0.277 | 6.2 ± 1.9 | 6.2 ± 1.8 | 0.841 |
| Body fat percentage (%); mean ± SD | 26.6 ± 6.7 | 27.7 ± 7.0 | 0.027 | 28.1 ± 10.2 | 28.0 ± 9.6 | 0.925 |
| Physiological function assessments | ||||||
| Systolic blood pressure (mmHg); mean ± SD | 123.0 ± 18.6 | 123.0 ± 17.6 | 0.690 | 131.4 ± 18.4 | 129.2 ± 18.4 | 0.033 |
| Heart rate (bpm); mean ± SD | 69.9 ± 10.2 | 69.2 ± 10.4 | 0.347 | 69.1 ± 10.0 | 69.8 ± 10.4 | 0.257 |
| Physical function assessment | ||||||
| Maximum walking speed (m/s); mean ± SD | 1.7 ± 0.2 | 1.8 ± 0.2 | < 0.001 | 1.7 ± 0.5 | 1.9 ± 0.5 | < 0.001 |
| Handgrip strength (kg); mean ± SD | 29.8 ± 9.0 | 30.4 ± 9.3 | 0.323 | 26.1 ± 8.1 | 26.5 ± 7.8 | 0.399 |
| Cognitive function assessment | ||||||
| Mini-Mental State Examination (point); mean ± SD | 27.5 ± 2.0 | 28.4 ± 1.4 | < 0.001 | - | ||
| Montreal Cognitive Assessment (point); mean ± SD | - | 24.7 ± 3.5 | 27.5 ± 1.9 | < 0.001 | ||
| Blood test | ||||||
| Total cholesterol (mg/dL); mean ± SD | 211.0 ± 33.0 | 218.0 ± 33.8 | 0.002 | 195.2 ± 35.4 | 199.3 ± 34.5 | 0.037 |
| Triglyceride (mg/dL); mean ± SD | 112.0 ± 66.7 | 120.0 ± 90.5 | 0.152 | 115.1 ± 57.2 | 110.1 ± 69.3 | 0.174 |
| HDL-cholesterol (mg/dL); mean ± SD | 59.3 ± 15.3 | 60.5 ± 14.5 | 0.233 | 57.7 ± 15.7 | 60.0 ± 15.8 | 0.009 |
| Homeostatic model assessment-insulin resistance°; mean ± SD | 1.8 ± 3.3 | 1.6 ± 1.1 | 0.321 | 2.4 ± 1.8 | 2.0 ± 1.5 | < 0.001 |
| Hemoglobin A1c (%); mean ± SD | 5.5 ± 0.7 | 5.5 ± 0.7 | 0.286 | 5.9 ± 0.7 | 5.8 ± 0.6 | < 0.001 |
| ALT (GPT) (U/L); mean ± SD | 19.6 ± 9.2 | 20.7 ± 9.7 | 0.102 | 25.7 ± 18.3 | 24.6 ± 11.3 | 0.224 |
| Albumin (g/dL); mean ± SD | 4.5 ± 0.3 | 4.6 ± 0.3 | 0.080 | 4.5 ± 0.3 | 4.5 ± 0.2 | 0.044 |
| Uric acid (mg/dL); mean ± SD | 5.2 ± 1.3 | 5.2 ± 1.3 | 0.688 | 5.6 ± 1.3 | 5.5 ± 1.2 | 0.094 |
| Hemoglobin (g/dL); mean ± SD | 14.5 ± 1.5 | 14.6 ± 1.4 | 0.470 | 13.9 ± 1.4 | 14.0 ± 1.3 | 0.143 |
| 25-OH vitamin D (ng/mL); mean ± SD | 27.0 ± 10.0 | 26.6 ± 8.7 | 0.511 | 23.8 ± 7.1 | 23.9 ± 7.4 | 0.881 |
| DHEA-S (ng/dL); mean ± SD | 97.6 ± 62.4 | 104.0 ± 63.6 | 0.161 | 103.4 ± 61.8 | 111.7 ± 64.9 | 0.020 |
| Estimated glomerular filtration rate (mL/min/1.73m2)d; mean ± SD | 92.6 ± 10.8 | 95.0 ± 8.3 | 0.001 | 86.3 ± 19.5 | 89.2 ± 16.8 | 0.006 |
a. For continuous variables, the general linear model is used; for categorical variables, the χ2 test is used. b. Calculated as appendicular skeletal muscle mass (kg)/height (m)/height (m). c. Calculated as (fasting glucose [mg/dL × insulin [mIU/L])/405. d. Calculated as follows: for men, if serum creatinine (scr) (mg/dL) ≤ 0.9, eGFR = 142 × (scr/0.9)−0.302 × 0.9938age and if serum creatinine (mg/dL) > 0.9, eGFR = 142 × (scr/0.9)−1200 × 0.9938age; for women, if serum creatinine (mg/dL) ≤ 0.7, eGFR = 142 × (scr/0.7)−0241 × 0.9938age × 1.012 and if serum creatinine (mg/dL) > 0.7, eGFR = 142 × (scr/0.7)−1200 × 0.9938age × 1.012. 25-OH vitamin D, 25-hydroxycholecalciferol; ALT, alanine transaminase; BMI, body mass index; DHEA-S, dehydroepiandrosterone sulfate; GPT, glutamic-pyruvic transaminase; HDL, high-density lipoprotein; IC, intrinsic capacity; LAST, Longitudinal Aging Study of Taipei; METs, Metabolic Equivalents of Task; NILS-LSA, National Institute for Longevity Sciences-Longitudinal Study of Aging; SD, standard deviation
IC Impairments, Falls, and Death
A significant association was identified between IC impairments and falls among all participants in the NILS-LSA cohort over the 2-year period (Table 3). Further analysis focusing on specific IC domains revealed that hearing loss in men in the NILS-LSA cohort and impaired locomotion and hearing loss in all participants in the LAST cohort were significant predictive factors of falls (Supplementary Table 1). Nevertheless, in the fully adjusted model, a positive association between IC impairments and mortality was observed only in men in the NILS-LSA cohort (Table 4). In the analysis of specific IC domains, visual impairment in men and impaired psychological well-being in women in the NILS-LSA cohort emerged as predictive factors that were significantly associated with mortality (Supplementary Table 2).
Table 3.
Association between IC impairments and falls over 2 years in the NILS-LSA (n = 794) and 3 years in the LAST (n = 1,358)a
| NILS-LSA (2 years) | LAST (3 years) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| No. of participants | Fall event (%) | OR | 95% CI | No. of participants | Fall event (%) | OR | 95% CI | ||||
| All | All | ||||||||||
| Healthy | 358 | 16.5 | 1.00 | Ref. | Healthy | 852 | 20.3 | 1.00 | Ref. | ||
| Impaired IC | 436 | 23.4 | Impaired IC | 506 | 22.3 | ||||||
| Model 1b | 1.49 | 1.03 | 2.16 | Model 1b | 1.10 | 0.83 | 1.46 | ||||
| Model 2c | 1.46 | 1.00 | 2.13 | Model 4e | 1.12 | 0.84 | 1.49 | ||||
| Model 3d | 1.50 | 1.03 | 2.20 | Model 5f | 1.13 | 0.85 | 1.51 | ||||
| Men | Men | ||||||||||
| Healthy | 169 | 13.0 | 1.00 | Ref. | Healthy | 272 | 12.9 | 1.00 | Ref. | ||
| Impaired IC | 238 | 20.2 | Impaired IC | 190 | 10.5 | ||||||
| Model 1b | 1.70 | 0.98 | 3.02 | Model 1b | 0.76 | 0.42 | 1.40 | ||||
| Model 2c | 1.59 | 0.90 | 2.85 | Model 4e | 0.79 | 0.43 | 1.46 | ||||
| Model 3d | 1.63 | 0.92 | 2.96 | Model 5f | 0.73 | 0.38 | 1.40 | ||||
| Women | Women | ||||||||||
| Healthy | 189 | 19.6 | 1.00 | Ref. | Healthy | 580 | 23.6 | 1.00 | Ref. | ||
| Impaired IC | 198 | 27.3 | Impaired IC | 316 | 29.4 | ||||||
| Model 1b | 1.33 | 0.81 | 2.19 | Model 1b | 1.22 | 0.89 | 1.68 | ||||
| Model 2c | 1.35 | 0.81 | 2.24 | Model 4e | 1.25 | 0.90 | 1.74 | ||||
| Model 3d | 1.36 | 0.81 | 2.28 | Model 5f | 1.29 | 0.93 | 1.80 | ||||
a. Analyzed by logistic regression model. b. Adjusted for age at baseline (continuous; years) and sex. c. Adjusted for model 1 plus baseline information on medical history (stroke and heart disease; yes or no for each), smoking status (current or non-current), total physical activity (Metabolic Equivalents of Task [METs]-h/day; tertile groups), education level (≤ 9, 10–12, or ≥ 13 years), marital status (married or other), and living status (alone or other). d. Adjusted for model 2 plus baseline information on body mass index (BMI, < 18.5, 18.5-< 25, or ≥ 25; kg/m2), calf circumference (continuous; cm), body fat percentage (continuous; %), total cholesterol (continuous; mg/dL), and estimated glomerular filtration rate (eGFR, continuous; mL/min/1.73m2). e. Adjusted for model 1 plus baseline information on medical history (stroke, hypertension, and heart disease; yes or no for each), alcohol consump-tion (current or non-current), total physical activity (METs-h/day; tertile groups), and education level (≤ 9, 10–12, or ≥ 13 years). f. Adjusted for model 4 plus baseline information on BMI (< 18.5, 18.5–< 25, or ≥ 25; kg/m2), systolic blood pressure (continuous; mmHg), total cholesterol (continuous; mg/dL), high-density lipoprotein (HDL)-cholesterol (continuous; mg/dL), homeostatic model assessment-insulin resistance (HOMA-IR, continuous), hemoglobin A1c (continuous; %), albumin (continuous; g/dL) eGFR, (continuous; mL/min/1.73m2), and dehydroepi-androsterone sulfate (DHEA-S) (continuous; µg/dL). CI, confidence interval; OR, odds ratio; IC, intrinsic capacity; LAST, Longitudinal Aging Study of Taipei; NILS-LSA, National Institute for Longevi-ty Sciences-Longitudinal Study of Aging; Ref., reference
Table 4.
Association of impairments of IC with death up to 11.5 years in the NILS-LSA (n = 794) and 3 years in the LAST (n = 1,358)a
| No. of participants | Death event (%) | Death/1,000 person-years | Model 1b | Model 2c | Model 3d | Model 4e | Model 5 | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HR | 95% CI | HR | 95% CI | HR | 95% CI | HR | 95% CI | HR | 95% CI | ||||
| NILS-LSA | |||||||||||||
| All | |||||||||||||
| Healthy | 358 | 5.3 | 5.2 | 1.00 | Ref. | 1.00 | Ref. | 1.00 | Ref. | ||||
| Impaired IC | 436 | 14.7 | 15.0 | 1.81 | 1.07 3.06 | 1.57 | 0.92 2.69 | 1.55 | 0.90 | ||||
| Men | |||||||||||||
| Healthy | 169 | 7.1 | 7.0 | 1.00 | Ref. | 1.00 | Ref. | 1.00 | Ref. | ||||
| Impaired IC | 238 | 22.3 | 23.4 | 2.33 | 1.23 4.41 | 2.02 | 1.05 3.87 | 2.14 | 1.10 4.15 | ||||
| Women | |||||||||||||
| Healthy | 189 | 3.7 | 3.6 | 1.00 | Ref. | 1.00 | Ref. | 1.00 | Ref. | ||||
| Impaired IC | 198 | 5.6 | 5.5 | 0.86 | 0.32 2.31 | 0.90 | 0.33 2.47 | 0.85 | 0.29 2.52 | ||||
| LAST (3 years) | |||||||||||||
| All | |||||||||||||
| Healthy | 852 | 1.2 | 4.0 | 1.00 | Ref. | 1.00 | Ref. | 1.00 | Ref. | ||||
| Impaired IC | 506 | 1.8 | 6.1 | 1.14 | 0.44 2.94 | 1.16 | 0.44 3.08 | 1.07 | 0.39 2.96 | ||||
| Men | |||||||||||||
| Healthy | 272 | 2.6 | 8.9 | 1.00 | Ref. | 1.00 | Ref. | 1.00 | Ref. | ||||
| Impaired IC | 190 | 3.2 | 11.0 | 0.92 | 0.29 2.94 | 0.95 | 0.28 3.26 | 0.77 | 0.19 3.22 | ||||
| Women | |||||||||||||
| Healthy | 580 | 0.5 | 1.8 | 1.00 | Ref. | 1.00 | Ref. | 1.00 | Ref. | ||||
| Impaired IC | 316 | 0.9 | 3.2 | 1.74 | 0.34 8.98 | 1.68 | 0.32 8.94 | 0.76 | 0.09 6.49 | ||||
a. Analyzed by Cox proportional hazards model; b. Adjusted for age at baseline (continuous; years) and sex. c. Adjusted for model 1 plus baseline information on medical history (stroke and heart disease; yes or no for each), smoking status (current or non-current), total physical activity (METs-h/day; tertile groups), education level (≤ 9, 10–12, or ≥ 13 years), marital status (married or other), and living status (alone or other). d. Adjusted for model 2 plus baseline information on BMI (< 18.5, 18.5–< 25, or ≥ 25; kg/m2), calf circumference (continuous; cm), body fat percent-age (continuous; %), total cholesterol (continuous; mg/dL), and eGFR (continuous; mL/min/1.73m2). e. Adjusted for model 1 plus baseline information on medical histoiy (stroke, hypertension, and heart disease; yes or no for each), alcohol consump-tion (current or non-current), total physical activity (METs-h/day; tertile groups), and education level (≤ 9, 10–12, or ≥ 13 years). f. Adjusted for model 4 plus baseline information on BMI (< 18.5, 18.5–< 25, or ≥ 25; kg/m2), systolic blood pressure (continuous; mmHg), total cholesterol (continuous; mg/dL), HDL-cholesterol (continuous; mg/dL), HOMA-IR (continuous), hemoglobin A1c (continuous; %), albumin (con-tinuous; g/dL), eGFR (continuous; mL/min/1.73m2), and DHEA-S (continuous; µg/dL). CI, confidence interval; HR, hazard ratio; LAST, Longitudinal Aging Study of Taipei; NILS-LSA, National Institute for Longevity Sciences-Longitudinal Study of Aging; Ref., reference
Discussion
In this comparative study, we used NILS-LSA (Japan) and LAST (Taiwan) data to compare IC impairment characteristics as well as their predictive capacity for subsequent falls and mortality. Our findings revealed significant variations in the frequency of IC domain impairments and associated biomarkers between two cohorts. Furthermore, we observed that IC impairment was a significant 2-year predictor of falls among older individuals in Japan and was linked to increased long-term mortality risk, specifically in older Japanese men. However, IC impairment was not predictive of falls or mortality among the older Taiwanese population during a relatively short follow-up period. To the best of our knowledge, this study is the first to explore the epidemiology of IC impairment and its clinical outcomes using longitudinal cohort data for international comparisons, highlighting the significance of functional capacity in promoting healthy aging (26).
The baseline prevalence of IC impairment in the NILS-LSA and LAST cohorts exhibited significant variation, potentially attributable to differences in the sampling strategies of two studies. NILS-LSA employed age- and sex-stratified sampling, whereas LAST employed relatively convenient random sampling from communities, which may have introduced a larger selection bias. To date, epidemiological studies that have employed the WHO ICOPE screening tool to assess IC impairments have revealed noteworthy domain variations. A previous investigation conducted in a middle-aged and older Chinese population reported the prevalence of the top three impaired domains: cognition (46.8%), locomotion (25.3%), and vitality (16.2%) (9). In a separate study in an older Chinese population, the respective proportions of the top three impaired domains were locomotion (37.8%), psychological well-being (35.2%), and cognition (24.3%) (10). Conversely, among older French individuals with memory complaints, the corresponding proportions were hearing loss (56.2%), cognitive impairment (52.2%), and impaired psychological well-being (39%) (11).
The current study revealed notable differences in the proportion of IC domain impairments between the NILS-LSA and LAST cohorts. Although variations in the population characteristics within each cohort may have contributed to these differences, differences in the assessment methods employed for each IC domain may also be a significant factor. For instance, the evaluation of hearing loss in the NILS-LSA utilizes air-conduction pure-tone threshold tests, whereas LAST relied on subjective reporting. Post-hoc analysis of the NILS-LSA data revealed that only 59.6% of the participants with hearing loss perceived their hearing to be consistently poor, whereas 14.0% of the participants without hearing loss considered their hearing to be persistently poor (data not shown). These findings suggest that subjective questionnaires do not accurately assess hearing loss. However, the LAST questionnaire did not directly assess hearing loss but rather focused on the impact of hearing loss on daily functioning. This discrepancy in questionnaire design may account for the varying proportions of individuals with hearing loss in the two cohorts. Therefore, the adverse effects of hearing loss in the LAST cohort may have been underestimated in our study.
The proportion of participants with impaired vitality was another noteworthy discrepancy between the two groups. Apart from the influence of distinct operational definitions, the composition of the two sub-items within the vitality domain, namely weight loss and lack of appetite, also contributed to these differences. Among participants identified with vitality impairment in the NILS-LSA cohort, 38.9% experienced weight loss alone, 50.9% experienced a lack of appetite alone, and 10.3% experienced both concurrently. In contrast, the corresponding proportions in the LAST cohort were 30.4%, 62.5%, and 7.1%, respectively (data not shown).
We identified a significant association between IC impairment and an elevated risk of falls among participants in the NILS-LSA cohort. Subsequent analysis investigating specific IC domains and their relationship with falls revealed that men in the NILS-LSA cohort and all participants with hearing loss in the LAST cohort exhibited a higher risk of falls. This finding aligns with the conclusions of a systematic review that demonstrated a notable positive correlation between hearing loss and an increased risk of falls in older adults (27). This association could be attributed to the presence of common underlying pathological mechanisms between hearing loss and falls, including concurrent vestibular dysfunction and cerebrovascular disease as well as diminished awareness of the auditory environment. Furthermore, our findings indicated that both men and women with impaired locomotion in the LAST cohort exhibited an increased risk of falls. This observation is consistent with those of previous studies that demonstrated a significant association between impaired performance on the chair rise test and a heightened risk of falls (12).
Our study further revealed a significant association between IC impairments, particularly visual impairment, and mortality in men as well as an association between impaired psychological well-being and mortality in women in the NILS-LSA cohort. Studies have provided compelling evidence that visual impairment (28, 29) and depression (30, 31) are predictive of mortality. These associations can be attributed to various factors, including sociodemographic factors, the presence of clinical and subclinical diseases, and health risk factors. However, the underlying factors contributing to the sex-specific associations observed in the present study remain unclear and require further investigation. Conversely, the lack of significant findings in the LAST cohort could potentially be attributed to the relatively short follow-up period and extremely low mortality rates. Kaplan-Meier survival curves derived from the NILS-LSA dataset demonstrated minimal divergence in survival probability between individuals with IC impairments and their healthy counterparts during the initial 3-year follow-up period, consistent with the outcomes observed in the LAST analysis (Supplementary Figure 1). This suggests that impaired IC has limited predictive power for short-term mortality.
To the best of our knowledge, as the first study to compare the epidemiology, associated factors, and prognosis of IC between Japan and Taiwan, our findings revealed the significant predictive value of IC impairments for adverse outcomes, albeit with variations in the prevalence of specific IC impairments and outcome prediction across the two cohorts. The strength of this study is its demonstration of the feasibility of promoting healthy aging on an international scale. However, addressing these challenges, devising strategies to promote healthy aging, and exploring their economic benefits (32) necessitate culturally and socially sensitive approaches (33) within the framework of the United Nations Decade of Healthy Aging.
Despite the strengths of this study, some limitations should be acknowledged. First, differences in sampling strategies between the NILS-LSA and LAST cohorts may have contributed to divergent research findings. Nevertheless, each cohort study has its own distinct design and characteristic hypothesis. Second, the relatively shorter follow-up period of the LAST cohort and the predominantly healthy state of the participants restricted our ability to establish associations with adverse outcomes. Nonetheless, this limited timeframe offered an opportunity to explore the predictive capacity of IC for short-term mortality in the LAST cohort compared to the long-term mortality risk observed in the NILS-LSA cohort. Third, the inclusion of an extended follow-up period and multiple measurements is essential in both cohorts to investigate supplementary outcome indicators linked to the concept of healthy aging beyond the assessment of falls and all-cause mortality. Finally, although most questions and tests were similar in both cohorts, the present inconsistencies could also be a reason for discrepancies in results between the two cohorts.
In summary, our study revealed substantial differences in the prevalence of IC impairments among community-dwelling older adults in Japan and Taiwan, which may be attributed to differences in the study design or variations in their baseline health characteristics. Despite these variations, the predictive capacity for adverse clinical outcomes was similar between the two cohorts. Future research should focus on investigating the trajectory of longitudinal changes in IC impairment and the potential factors associated with the reversal of IC impairment across different countries, thereby facilitating the development of socially and culturally sensitive approaches aimed at promoting healthy aging.
Author Contributions:
LK C and HA contributed to the study conception. RO, YN, and HA contributed to the design of NILSLSA. LN P and LK C contributed to the design of LAST. SZ, LN P, RO, and LK C obtained funding. RO, YN, LN P, and LK C collected the data. SZ and LN P drafted the plans for the data analyses. SZ and LN P conducted data analysis. SZ and LN P drafted the manuscript. All authors were involved in interpretation of the results and revision of the manuscript, and all approved the final version of the manuscript. SZ, LN P, and LK C are guarantors. The corresponding author attests that all the listed authors meet the authorship criteria and that no others meeting the criteria have been omitted.
Funding Statement:
This study was supported in part by Research Funding for Longevity Sciences from the National Center for Geriatrics and Gerontology, Japan (grant number 21-18); Grant-in-Aid for Early-Career Scientists of Japan Society for the Promotion of Science KAKENHI (Grant Number JP 22K17847); National Science and Technology Council, Taiwan (NSTC 112-2321-B-A49-006; NSTC 112-2321-B-075-004); and Interdisciplinary Research Center for Healthy Longevity of National Yang Ming Chiao Tung University from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education in Taiwan.
Data Availability:
The datasets analyzed during the current study are not publicly available due to privacy reasons but are available from the corresponding author on reasonable request.
Declaration of Interests:
None.
Ethical standards:
The Committee on the Ethics of Human Research of the National Center for Geriatrics and Gerontology approved the study protocol (No. 899-6), and written informed consent was obtained from all participants in NILS-LSA. The Institutional Review Board of the National Yang Ming University approved the study (YM104121F-5), and all participants provided written informed consent.
Electronic Supplementary Material
Supplementary material is available in the online version of this article at https://doi.org/10.1007/s12603-023-2020-z.
Appendix
References
- 1.Beard JR, Officer A, de Carvalho IA, Sadana R, Pot AM, Michel JP, Lloyd-Sherlock P, Epping-Jordan JE, Peeters GMEEG, Mahanani WR, Thiyagarajan JA, Chatterji S. The World report on ageing and health: a policy framework for healthy ageing. Lancet. 2016;387:2145–2154. doi: 10.1016/S0140-6736(15)00516-4. 10.1016/S0140-6736(15)00516-4 PubMed PMID: 26520231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.World Health Organization . Integrated care for older people: guidelines on community-level interventions to manage declines in intrinsic capacity. World Health Organization; Geneva: 2017. [PubMed] [Google Scholar]
- 3.World Health Organization . Integrated care for older people (ICOPE): guidance for person-centred assessment and pathways in primary care. World Health Organization; Geneva: 2019. CC BY-NC-SA 3.0 IGO. [Google Scholar]
- 4.Chen LK. Gan-Dau healthy longevity plan: the community model for healthy aging. Aging Med Healthc. 2022;13:98–101. 10.33879/AMH.133.2022.09088 [Google Scholar]
- 5.Meng LC, Huang ST, Peng LN, Chen LK, Hsiao FY. Biological features of the outcome-based intrinsic capacity composite scores from a population-based cohort study: pas de deux of biological and functional aging. Front Med (Lausanne) 2022;9:851882. doi: 10.3389/fmed.2022.851882. 10.3389/fmed.2022.851882 PubMed PMID: 35308493. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Meng LC, Hsiao FY, Huang ST, Lu WH, Peng LN, Chen LK. Intrinsic capacity impairment patterns and their associations with unfavorable medication utilization: a nationwide population-based study of 37,993 community-dwelling older adults. J Nutr Health Aging. 2022;26:918–925. doi: 10.1007/s12603-022-1847-z. 10.1007/s12603-022-1847-z PubMed PMID: 36259580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lee WJ, Peng LN, Lin MH, Loh CH, Hsiao FY, Chen LK. Intrinsic capacity differs from functional ability in predicting 10-year mortality and biological features in healthy aging: results from the I-Lan longitudinal aging study. Aging (Albany NY) 2023;15:748–764. doi: 10.18632/aging.204508. 10.18632/aging.204508 PubMed PMID: 36750172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Tavassoli N, de Souto Barreto P, Berbon C, Mathieu C, de Kerimel J, Lafont C, Takeda C, Carrie I, Piau A, Jouffrey T, Andrieu S, Nourhashemi F, Beard JR, Soto Martin ME, Vellas B. Implementation of the WHO integrated care for older people (ICOPE) programme in clinical practice: a prospective study. Lancet Healthy Longev. 2022;3:e394–e404. doi: 10.1016/S2666-7568(22)00097-6. 10.1016/S2666-7568(22)00097-6 PubMed PMID: 36098317. [DOI] [PubMed] [Google Scholar]
- 9.Ma L, Chhetri JK, Zhang Y, Liu P, Chen Y, Li Y, Chan P. Integrated care for older people screening tool for measuring intrinsic capacity: preliminary findings from ICOPE pilot in China. Front Med (Lausanne) 2020;7:576079. doi: 10.3389/fmed.2020.576079. 10.3389/fmed.2020.576079 PubMed PMID: 33330532. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Leung AYM, Su JJ, Lee ESH, Fung JTS, Molassiotis A. Intrinsic capacity of older people in the community using WHO Integrated Care for Older People (ICOPE) framework: a cross-sectional study. BMC Geriatr. 2022;22:304. doi: 10.1186/s12877-022-02980-1. 10.1186/s12877-022-02980-1 PubMed PMID: 35395736; PMCID 8993034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.González-Bautista E, de Souto Barreto P, Virecoulon Giudici K, Andrieu S, Rolland Y, Vellas B. Frequency of conditions associated with declines in intrinsic capacity according to a screening tool in the context of integrated care for older people. J Frailty Aging. 2021;10:94–102. doi: 10.14283/jfa.2020.42. PubMed PMID: 33575697. [DOI] [PubMed] [Google Scholar]
- 12.Liu S, Yu X, Wang X, Li J, Jiang S, Kang L, Liu X. Intrinsic Capacity predicts adverse outcomes using Integrated Care for Older People screening tool in a senior community in Beijing. Arch Gerontol Geriatr. 2021;94:104358. doi: 10.1016/j.archger.2021.104358. 10.1016/j.archger.2021.104358 PubMed PMID: 33548677. [DOI] [PubMed] [Google Scholar]
- 13.Prince MJ, Acosta D, Guerra M, Huang Y, Jacob KS, Jimenez-Velazquez IZ, Jotheeswaran AT, Llibre Rodriguez JJ, Salas A, Sosa AL, Acosta I, Mayston R, Liu Z, Llibre-Guerra JJ, Prina AM, Valhuerdi A. Intrinsic capacity and its associations with incident dependence and mortality in 10/66 Dementia Research Group studies in Latin America, India, and China: A population-based cohort study. PLoS Med. 2021;18:e1003097. doi: 10.1371/journal.pmed.1003097. 10.1371/journal.pmed.1003097 PubMed PMID: 34520466; PMCID 8439485. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Shimokata H, Ando F, Niino N. A new comprehensive study on aging—the National Institute for Longevity Sciences, Longitudinal Study of Aging (NILS-LSA) J Epidemiol. 2000;10:S1–S9. doi: 10.2188/jea.10.1sup_1. 10.2188/jea.10.1sup_1 PubMed PMID: 10835822. [DOI] [PubMed] [Google Scholar]
- 15.Yu PC, Hsu CC, Lee WJ, Liang CK, Chou MY, Lin MH, Hsiao FY, Peng LN, Chen LK. Muscle-to-fat ratio identifies functional impairments and cardiometabolic risk and predicts outcomes: biomarkers of sarcopenic obesity. J Cachexia Sarcopenia Muscle. 2022;13:368–376. doi: 10.1002/jcsm.12877. 10.1002/jcsm.12877 PubMed PMID: 34866342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Markwick A, Zamboni G, de Jager CA. Profiles of cognitive subtest impairment in the Montreal Cognitive Assessment (MoCA) in a research cohort with normal MiniMental State Examination (MMSE) scores. J Clin Exp Neuropsychol. 2012;34:750–757. doi: 10.1080/13803395.2012.672966. 10.1080/13803395.2012.672966 PubMed PMID: 22468719. [DOI] [PubMed] [Google Scholar]
- 17.Chen LK, Woo J, Assantachai P, et al. Asian Working Group for Sarcopenia: 2019 consensus update on sarcopenia diagnosis and treatment. J Am Med Dir Assoc. 2020;21:300–307.e2. doi: 10.1016/j.jamda.2019.12.012. 10.1016/j.jamda.2019.12.012 PubMed PMID: 32033882. [DOI] [PubMed] [Google Scholar]
- 18.Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401. 10.1177/014662167700100306 [Google Scholar]
- 19.Shima S. New self-rating scale for depression. Clin Psychiatry. 1985;27:717–723. [Google Scholar]
- 20.Reuter M, Schmansky NJ, Rosas HD, Fischl B. Within-subject template estimation for unbiased longitudinal image analysis. Neuroimage. 2012;61:1402–1418. doi: 10.1016/j.neuroimage.2012.02.084. 10.1016/j.neuroimage.2012.02.084 PubMed PMID: 22430496. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, Pratt M, Ekelund U, Yngve A, Sallis JF, Oja P. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35:1381–1395. doi: 10.1249/01.MSS.0000078924.61453.FB. 10.1249/01.MSS.0000078924.61453.FB PubMed PMID: 12900694. [DOI] [PubMed] [Google Scholar]
- 22.Health Promotion Administration MoHaW (2018) Taiwanese version of the International Physical Activity Questionnaire Short Form. https://www.hpa.gov.tw/Pages/Detail.aspx?nodeid=876&pid=4900.
- 23.Otsuka R, Kato Y, Imai T, Ando F, Shimokata H. Decreased salt intake in Japanese men aged 40 to 70 years and women aged 70 to 79 years: an 8-year longitudinal study. J Am Diet Assoc. 2011;111:844–850. doi: 10.1016/j.jada.2011.03.020. 10.1016/j.jada.2011.03.020 PubMed PMID: 21616196. [DOI] [PubMed] [Google Scholar]
- 24.Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28:412–419. doi: 10.1007/BF00280883. 10.1007/BF00280883 PubMed PMID: 3899825. [DOI] [PubMed] [Google Scholar]
- 25.Inker LA, Eneanya ND, Coresh J, et al. New creatinine- and cystatin C-based equations to estimate GFR without race. N Engl J Med. 2021;385:1737–1749. doi: 10.1056/NEJMoa2102953. 10.1056/NEJMoa2102953 PubMed PMID: 34554658; PMCID 8822996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Kim HJ, Min KB, Min JY. Healthy aging predicts mortality risks: results from the Korean longitudinal study of aging (2006–2014) Arch Gerontol Geriatr. 2021;94:104333. doi: 10.1016/j.archger.2020.104333. 10.1016/j.archger.2020.104333 PubMed PMID: 33516975. [DOI] [PubMed] [Google Scholar]
- 27.Jiam NT, Li C, Agrawal Y. Hearing loss and falls: a systematic review and meta-analysis. Laryngoscope. 2016;126:2587–2596. doi: 10.1002/lary.25927. 10.1002/lary.25927 PubMed PMID: 27010669. [DOI] [PubMed] [Google Scholar]
- 28.Karpa MJ, Mitchell P, Beath K, Rochtchina E, Cumming RG, Wang JJ. Direct and indirect effects of visual impairment on mortality risk in older persons. Arch Ophthalmol. 2009;127:1347–1353. doi: 10.1001/archophthalmol.2009.240. 10.1001/archophthalmol.2009.240 PubMed PMID: 19822852. [DOI] [PubMed] [Google Scholar]
- 29.McCarty CA, Nanjan MB, Taylor HR. Vision impairment predicts 5-year mortality. Br J Ophthalmol. 2001;85:322–326. doi: 10.1136/bjo.85.3.322. 10.1136/bjo.85.3.322 PubMed PMID: 11222339; PMCID 1723877. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Schulz R, Martire LM, Beach SR, Scheier MF. Depression and mortality in the elderly. Curr Dir Psychol Sci. 2000;9:204–208. 10.1111/1467-8721.00095 [Google Scholar]
- 31.Schulz R, Beach SR, Ives DG, Martire LM, Ariyo AA, Kop WJ. Association between depression and mortality in older adults: the Cardiovascular Health Study. Arch Intern Med. 2000;160:1761–1768. doi: 10.1001/archinte.160.12.1761. 10.1001/archinte.160.12.1761 PubMed PMID: 10871968. [DOI] [PubMed] [Google Scholar]
- 32.Pagès A, Costa N, González-Bautista E, Mounié M, Juillard-Condat B, Molinier L, Cestac P, Rolland Y, Vellas B, De Souto Barreto P. Screening for deficits on intrinsic capacity domains and associated healthcare costs. Arch Gerontol Geriatr. 2022;100:104654. doi: 10.1016/j.archger.2022.104654. 10.1016/j.archger.2022.104654 PubMed PMID: 35193041. [DOI] [PubMed] [Google Scholar]
- 33.Chen LK, Iijima K, Shimada H, Arai H. Community re-designs for healthy longevity: Japan and Taiwan examples. Arch Gerontol Geriatr. 2022;104:104875. doi: 10.1016/j.archger.2022.104875. 10.1016/j.archger.2022.104875 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Appendix
Data Availability Statement
The datasets analyzed during the current study are not publicly available due to privacy reasons but are available from the corresponding author on reasonable request.