Key summary points
Aim
To investigate the association between comorbidities associated with diabetes and higher-level functional status in older patients with type 2 diabetes mellitus.
Findings
Comorbidities of depression, low muscle strength, and physical frailty were associated with the malfunction status of higher-level functional status in older patients with type 2 diabetes mellitus. Expanding social networks reduced the impact of comorbidities on higher-level functional status deterioration.
Message
Care for depression, low muscle strength, and physical frailty, and expanding social networks are important to delay the development of future disabilities for older patients with type 2 diabetes mellitus.
Supplementary Information
The online version contains supplementary material available at 10.1007/s41999-024-00937-8.
Keywords: Depression, Frailty, Instrumental activities of daily living, Intellectual activity, Low muscle strength, Tokyo Metropolitan Institute of Gerontology Index of Competence
Abstract
Purpose
To investigate the association between comorbidities associated with diabetes and higher-level functional status as well as the relationship between comorbidities associated with diabetes and higher-level functional status in older patients with type 2 diabetes mellitus who have better social networks.
Methods
Participants were outpatients with type 2 diabetes aged ≥ 65 years, excluding individuals with severe cardiovascular or respiratory illness, hyperglycaemic crisis, type 1 diabetes, or diabetic foot. The Tokyo Metropolitan Institute of Gerontology Index of Competence (TMIG-IC) was used to evaluate the higher-level functional status. A TMIG-IC score of ≤ 9, instrumental activities of daily living (IADL) ≤ 4, intellectual activity or social role ≤ 3 were defined as decline in higher-level functional status. The comorbidities investigated included peripheral neuropathy, retinopathy, nephropathy, cognitive impairment, depression, frailty, sarcopenia, low muscle strength, stroke, heart disease, and arthritis.
Results
The analysis included 198 patients (mean age 75.9 ± 5.7 years, male 60.1%). After adjusting for potential confounders, depression was associated with TMIG-IC (Prevalence ratio (PR) 2.34, 95% confidence interval (CI) 1.44–3.82), low muscle strength was associated with IADL (PR 2.85, 95% CI 1.30–6.27), and frailty was associated with intellectual activity (PR 1.38, 95% CI 1.10–1.74). In the model with social networks added as a confounder, the relationship between depression or low muscle strength and higher-level functional status was not statistically significant.
Conclusion
Comorbidities of depression and low muscle strength for older patients with type 2 diabetes mellitus increase the risk of malfunctioning of higher-level functional status. Increased interactions with family, friends and neighbours may reduce this event.
Supplementary Information
The online version contains supplementary material available at 10.1007/s41999-024-00937-8.
Introduction
Functional status is a key health indicator for older people [1]. In order from simple to complex Lawton's seven-stage hierarchical model, functional status is characterized as the following: life maintenance, functional health, perception–cognition, physical self-maintenance, instrumental self-maintenance, effectance, and social roles [2]. In this hierarchical model, the three complex functions (instrumental self-maintenance, effectance, and social roles) are called higher-level functional status, which are necessary for the capabilities for leading an active and independent life in the community [3]. Previous studies have reported that the malfunction status of higher-level functional status is associated with an increased risk of stroke, death, and cost of medical care or nursing care [4, 5]. Therefore, maintaining higher-level functional status helps prevent adverse health outcome events in the future and prolongs healthy life expectancy.
In Japan, a super-aging society, approximately one in five adults aged 65 and older has diabetes [6]. Elderly patients with diabetes have various comorbidities [7–10], which may affect higher-level functional status. In fact, a systematic review reported that patients with diabetes have an approximately 50–80% increased risk of disability compared to patients without diabetes [11].
Comorbidities associated with the basic activities of daily living (ADLs) and instrumental activities of daily living (IADLs) have been explored in elderly patients with diabetes mellitus; however, few studies have focused on higher-level functional status. Previous studies have reported a longitudinal relationship between the malfunction status of higher-level functional status and cognitive impairment, and physical inactivity, and a cross-sectional association with low muscle strength [12, 13]. A cross-sectional study examined the relationship between comorbidities associated with diabetes and a decline in social roles, suggesting that diabetic nephropathy, depression, and frailty may affect the decline in social roles [14]. These studies suggest that higher-level functional status is based on physical and psychological/cognitive functions.
However, a previous study investigating the relationship between comorbidities associated with diabetes and higher-level functional status reported two common problems. First, confounders that influenced the relationship between comorbidities associated with diabetes and higher-level functional status (e.g., subjective health, educational condition, and socioeconomic factors) were not sufficiently considered. Among them, the social networks (relationships with friends, acquaintances, and family) have been reported to influence physical, mental, psychological, and higher-level functional status [15–17]. Therefore, social networks may influence the association between comorbidities associated with diabetes and higher-level functional status. Second, evidence regarding the relationship between musculoskeletal problems and higher-level functional status in elderly patients with diabetes is lacking. Impairment of basic ADLs and mortality risk are more closely associated with low muscle strength than with sarcopenia [18]. Low muscle strength may have more adverse effects than sarcopenia on higher-level functional status in elderly patients with diabetes mellitus. It is necessary to clarify the relationship between higher-level functional status and comorbidities associated with diabetes so as to identify comorbidities that warrant enhanced care and preventive interventions for the preservation of higher-level functional status.
Therefore, we aimed to investigate the association of comorbidities associated with diabetes including musculoskeletal disorders with higher-level functional status as well as the relationship between comorbidities associated with diabetes and higher-level functional status when the social network in older patients with type 2 diabetes mellitus is in good condition. We set up the following hypotheses: (i) among comorbidities associated with diabetes, low muscle strength, depression, and cognitive decline are associated with the malfunction status of higher-level functional status in older patients with type 2 diabetes mellitus, and (ii) elderly patients with type 2 diabetes mellitus who have better social networks attenuate the effect of comorbidities associated with diabetes on higher-level functional status.
Methods
Study design and participant
This cross-sectional study included 198 outpatients with type 2 diabetes aged 65 years or older who attended the Department of Inter Medicine or Diabetes, Endocrinology, and Metabolism, Kohnodai Hospital, National Centre for Global Health and Medicine (Ichikawa City, Chiba Prefecture, Japan) from June 2022 to March 2023. The exclusion criteria were as follows: having severe cardiovascular or respiratory illness, hyperglycaemic crisis, patients with type 1 diabetes, or diabetic foot with ulceration or gangrene. All patients who agreed to participate in the study completed a questionnaire and underwent blood, urine, physical, and cognitive function tests at the next outpatient visit after inclusion. Additionally, information on demographics and treatment status were collected at the same time. This study was approved by the ethics board of the National Centre for Global Health and Medicine (approval number: NCGM-S-004396-00) and was conducted following the principles of the Declaration of Helsinki. Written informed consent was obtained from all the patients before participating in the study.
Assessment of higher-level functional status
The self-administered questionnaire of the Tokyo Metropolitan Institute of Gerontology Index of Competence (TMIG-IC) was used to assess higher-level functional status of patients [19, 20]. The TMIG-IC is an index of higher-level functional status commonly used in Japan, which corresponds to Lawton’s hierarchical model of instrumental self-maintenance, effectance, and social roles. TMIG-IC composed of 13 questions. The response to each question was rated as “0” for no and “1” for yes. The total score for the TMIG-IC and each subscale was obtained by adding the scores of all questions (maximum score: IADL = 5; intellectual activity = 4; social role = 4; and TMIG-IC score = 13). A higher score represents good condition of higher-level functional status. Previous studies have reported that TMIG-IC has high reliability and validity [19]. In this study, patients whose total score of TMIG-IC was ≤ 9 were defined as having malfunction status of higher-level functional status [4, 21, 22]. Regarding the subscales, a patient was defined as having malfunction status of each subscale, if they reported not full score (score ≤ 4 out of 5 for IADL; ≤ 3 out of 4 for intellectual activity or social role) [4].
Assessing of comorbidities associated with diabetes
The following comorbidities associated with diabetes were investigated based on medical records, blood tests, urinalysis, questionnaires, and physical and cognitive functioning tests: diabetic peripheral neuropathy, diabetic retinopathy, diabetic nephropathy, cognitive impairment, depression, physical frailty, sarcopenia, low muscle strength, stroke, heart disease, and arthritis.
Diabetic peripheral neuropathy was defined as meeting at least two of the following three criteria: the presence of subjective symptoms (numbness, pain, and paraesthesia in both soles or toes), reduced Achilles tendon reflex, and decreased vibration sensation in the medial malleolus [23].
Diabetic retinopathy was investigated using ophthalmologists’ medical records within six months. It was classified as no diabetic retinopathy (NDR), simple retinopathy (SDR), pre-proliferative retinopathy (pre-PDR), or proliferative retinopathy (PR) according to Davis’ classification [24].
Diabetic nephropathy was classified into five stages according to the classification of Diabetic Nephropathy 2014, based on the results of urinary albumin/Cr ratio (Albumin Creatinine Ratio: ACR) and estimated glomerular filtration rate (eGFR) (Stage 1: ACR < 30 mg/gCre, Stage 2: ACR is 30–299 mg/gCre, Stage 3: ACR is ≥ 300 mg/gCre, Stage 4: eGFR is < 30 mL/min/1.73 m2 regardless of ACR value, and Stage 5: maintenance dialysis) [25]. In this study, the presence of nephropathy was defined as patients with Stage 2 or higher [14].
Cognitive function was evaluated using the Mini-Mental State Examination (MMSE) [26]. The total possible score was 30, with higher scores indicating better cognitive function. Cognitive impairment was defined as an MMSE score ≤ 23 points [27].
Depression was assessed by GDS-15 (Geriatric Depression Scale-15), with higher scores indicating a higher degree of depression [28]. In this study, a GDS score of ≥ 10 was defined as depression referring to the previous study [29].
Physical frailty was defined according to the Japanese version of the Cardiovascular Health Study (J-CHS) criteria [30]. Patients underwent grip strength measurements using a grip strength meter (GRIP-D5401; Takei Scientific Instruments Co., Ltd., Niigata, Japan) [31], a 5-m usual walking test [32], and were asked about weight loss, fatigue, and physical activity. The presence of physical frailty was defined as three or more of the following items: muscle weakness (grip strength: < 28 kg for men, < 18 kg for women), decrease in walking speed (usual walking speed < 1.0 m/s), weight loss (weight loss of ≥ 2 kg in the last 6 months), fatigue (feeling unexplainedly tired for the last 2 weeks), and inactivity (Do you do light exercise at least once a week? Do you participate in regular sports at least once a week? If they answered “no” to both questions).
Sarcopenia and low muscle strength were classified based on the AWGS2019 [33]. Sarcopenia was defined as decreased muscle strength (grip strength: < 28 kg for men, < 18 kg for women) and decreased skeletal muscle mass index (SMI) according to the Bioelectrical Impedance Analysis (BIA) method (SMI: < 7.0 kg/m2 for men, < 5.7 kg/m2 for women). Low muscle strength was defined as decreased muscle strength (grip strength: < 28 kg for men, < 18 kg for women), but normal SMI (SMI: ≥ 7.0 kg/m2 for men, ≥ 5.7 kg/m2 for women).
To investigate the presence of stroke (cerebral infarction or hemorrhage), cardiovascular disease (myocardial infarction, angina pectoris, and heart failure), and arthritis, the patients were asked, “have you ever been diagnosed or treated before?” for each disease. If they answered “yes,” they were defined as having comorbidities from each disease.
Other variables
We also collected the following information for using as covariates referencing previous studies [12, 13, 34]: age, sex, body mass index (BMI), insulin use status (using insulin or not), HbA1c level, smoking status (current smoker or not), educational status (less than high school education, high school education, or more than high school education), subjective economic conditions (poor/normal or good), subjective health (healthy or not), living status (living alone or not), physical activity, and social networks. The body mass index (BMI) was calculated by dividing the patient’s weight in kilograms by the square of their height in meters (kg/m2) and classified as underweight (BMI < 18.5), normal weight (18.5 ≤ BMI < 25.0), and obese (BMI ≥ 25.0). Physical activity was assessed by the Physical Activity Scale for the Elderly (PASE) [35]. PASE assessed the physical activities of three domains (leisure, household, and work-related activity) for the past week. Higher scores indicate better physical activity. To assess social networks, we used the Lubben Social Network Scale-short version (LSNS-6) [36]. LSNS-6 consists of six questions on interactions with family and friends. The scores range from 0 to 30, with higher scores indicating larger social networks.
Statistical analysis
Continuous variables were expressed as the mean ± standard deviations (SD), and categorical variables as ratios. To investigate the relationship between the malfunction status of higher-level functional status and comorbidities associated with diabetes, Poisson regression analysis was conducted, and the prevalence ratio (PR) and 95% confidence interval (CI) were calculated for each related factor. The malfunction status of the TMIG-IC or each subscale was used as an objective variable. Comorbidities associated with diabetes were used as explanatory variables, which were individually put into Poisson regression analysis. The use of logistic regression analysis for data with a high frequency of outcomes results in the gap between the calculated odds ratio and the true relative risk value widening. Therefore, if the frequency of the outcome (patients who had malfunction status of higher-level functional status) is > 10%, as in this study, calculation of prevalence ratios using Poisson regression analysis is recommended [37, 38]. Therefore, we conducted Poisson regression analysis. Cognitive impairment, diabetic nephropathy, depression, physical frailty, and low muscle strength were considered as comorbidities associated with diabetes. These comorbidities associated with diabetes have been reported to be associated with higher-level functional status [12, 13, 39]. A supplemental analysis was conducted to investigate the relationship between other comorbidities associated with diabetes (diabetic peripheral neuropathy, sarcopenia, stroke, heart disease, and arthritis) and higher-level functional status. Additionally, we adjusted for confounders using propensity scores, which allowed multiple confounders to be aggregated into one variable. The effectiveness of controlling for confounders using propensity scores has already been reported in several observational studies [40, 41]. The propensity scores used for the analysis in this study included the following variables: age, sex, BMI, insulin use status, HbA1c, smoking status, educational status, subjective economic condition, subjective health, living status, physical activity, and other comorbidities associated with diabetes, which were not included as explanatory variables. Thus, we calculated the propensity scores for each comorbidity associated with diabetes in this study separately. Model 1 included the abovementioned propensity scores as moderating variables. Furthermore, to investigate the impact of social networks on the association between comorbidities associated with diabetes and higher-level functional status, Model 2 included LSNS-6 scores as moderator variables in addition to Model 1.
All statistical analyses were conducted using SPSS IBM SPSS (Version 28.0; IBM Corp., Armonk, NY, USA). In this study, we investigated the association between higher-level functional status individually for each comorbidity associated with diabetes (five items); therefore, it was necessary to consider the multiplicity of the test. Thus, Bonferroni correction (0.05/5 = 0.01) was conducted, and the significance level was set at P < 0.01.
Results
The descriptive characteristics of the 198 patients who participated in this study are shown in Table 1. The mean ± SD age of the patients was 75.9 ± 5.7 years, and 119 (60.1%) were male. The mean ± SD of the TMIG-IC score was 10.7 ± 2.2 points, and 49 patients (24.9%) had TMIG-IC malfunction status. The mean ± SD of scores for IADL, intellectual activity, and social role were 4.7 ± 0.8, 3.2 ± 0.9, and 2.7 ± 1.3 points, respectively. Among the patients, 27 (13.6%), 105 (53.0%), and 126 (64.0%) had malfunctioning status of IADL, intellectual activity, and social roles, respectively.
Table 1.
Characteristics of the participants (n = 198)
| Variables | n | |
|---|---|---|
| Age (years), mean ± SD | 198 | 75.9 ± 5.7 |
| Male sex | 198 | 119 (60.1) |
| BMI (kg/m2) | 198 | |
| < 18.5 | 5 (2.5) | |
| 18.5–24.9 | 113 (57.1) | |
| ≥ 25 | 80 (40.4) | |
| Poor subjective hearing | 198 | 85 (42.9) |
| Poor subjective visual acuity | 198 | 104 (52.5) |
| Experience of falling in the past year | 198 | 58 (29.3) |
| Living alone | 198 | 45 (22.7) |
| Poor economic condition | 198 | 45 (22.7) |
| Poor self-rated health | 198 | 32 (16.2) |
| Latest academic background | 198 | |
| Junior high school | 23 (11.6) | |
| High school | 80 (40.4) | |
| University or vocational college | 95 (48.0) | |
| Smoking | 23 (11.6) | |
| Physical activity (PASE) (points), mean ± SD | 198 | 103.5 ± 66.6 |
| Social network (LSNS-6) (points), mean ± SD | 198 | 12.8 ± 5.9 |
| Use of medication | ||
| Insulin | 198 | 40 (20.2) |
| Sulfonylureas | 198 | 19 (9.6) |
| Biguanides | 198 | 111 (56.1) |
| Thiazolidines | 198 | 27 (13.6) |
| α-Glucosidase inhibitors | 198 | 28 (14.1) |
| Glinides | 198 | 17 (8.6) |
| DPP4 inhibitors | 198 | 116 (58.6) |
| SGLT2 inhibitors | 198 | 100 (50.5) |
| GLP-1 receptor agonists | 198 | 44 (22.2) |
| HbA1c (%), mean ± SD | 197 | 7.1 ± 0.9 |
| Neuropathy | 198 | 71 (35.9) |
| Retinopathy | 170 | 45 (22.7) |
| Nephropathy | 197 | 88 (44.4) |
| Stroke | 198 | 19 (9.6) |
| Cardiovascular disease | 198 | 25 (12.6) |
| Physical frailty | 198 | 49 (24.7) |
| Depression | 195 | 18 (9.1) |
| Cognitive decline | 198 | 30 (15.2) |
| Arthritis | 198 | 10 (5.1) |
| Sarcopenia | 197 | 18 (9.1) |
| Low muscle strength | 197 | 55 (27.9) |
| Higher-level competence | ||
| TMIG-IC score (points), mean ± SD | 197 | 10.7 ± 2.2 |
| IADL score (points), mean ± SD | 198 | 4.7 ± 0.8 |
| Intellectual activity score (points), mean ± SD | 198 | 3.2 ± 0.9 |
| Social role score (points), mean ± SD | 197 | 2.7 ± 1.3 |
| Malfunction status of TMIG-IC | 197 | 49 (24.9) |
| Malfunction status of IADL | 198 | 27 (13.6) |
| Malfunction status of intellectual activity | 198 | 105 (53.0) |
| Malfunction status of social role | 197 | 126 (64.0) |
Values are numbers (percentages) unless stated otherwise
SD standard deviations, BMI body mass index, PASE Physical Activity Scale for the Elderly, LSNS-6 Lubben Social Network Scale-6, HbA1c hemoglobin A1c, TMIG-IC Tokyo Metropolitan Institute Gerontology Index of Competence, IADL Instrumental activities of daily living
The association between the five comorbidities associated with diabetes and the malfunction status of the TMIG-IC are presented in Table 2. In the crude model, the prevalence of TMIG-IC malfunction status was significantly higher in patients with physical frailty, depression, cognitive decline, and low muscle strength (PR 2.46, 5.05, 2.09, and 2.36, respectively). However, these associations were weakened in Model 1, which was adjusted for propensity scores with multiple covariates, and only depression was significantly associated with the malfunctioning status of TMIG-IC (PR2.34). In Model 2, which added LSNS-6 scores to Model 1, there was no significant association between depression and the malfunction status of the TMIG-IC. Incidentally, there was no significant association between other comorbidities associated with diabetes and the malfunction status of the TMIG-IC in any model (Online Resource: Supplementary Table 1).
Table 2.
Prevalence ratio (95% confidence interval) for malfunction status of TMIG-IC: determined by Poisson regression analysis
| Variables | n | Prevalence of TMIG-IC disability, n (%) | Crude model | Model 1 | Model 2 |
|---|---|---|---|---|---|
| PR (95% CI) | PR (95% CI) | PR (95% CI) | |||
| Diabetic nephropathy | 88 | 27 (30.7) | 1.51 (0.93–2.45) | 1.28 (0.84–1.94) | 1.51 (1.00–2.26) |
| Physical frailty | 49 | 22 (44.9) | 2.46* (1.55–3.90) | 1.73 (1.14–2.64) | 1.64 (1.10–2.45) |
| Depression | 18 | 16 (88.9) | 5.05* (3.53–7.23) | 2.34* (1.44–3.82) | 1.59 (1.04–2.44) |
| Cognitive decline | 29 | 13 (44.8) | 2.09* (1.27–3.44) | 1.58 (1.02–2.45) | 1.78* (1.17–2.71) |
| Low muscle strength | 55 | 22 (40.0) | 2.36* (1.42–3.92) | 1.48 (0.94–2.33) | 1.40 (0.92–2.15) |
Model 1: adjusted for propensity score (including age, gender, use of insulin, HbA1c, BMI, self-rated health, smoking status, physical activity level, academic history, economic condition, living status, other comorbidities). Model 2: Model 1 + score of Lubben Social Network Scale-6
TMIG-IC Tokyo Metropolitan Institute Gerontology Index of Competence, PR prevalence ratio, CI confidence interval
*P < 0.01 (Bonferroni correction was used to adjusted the level of significance from 0.05 to 0.05/5 = 0.01)
The association between comorbidities associated with diabetes and IADL malfunction status are demonstrated in Table 3. In the crude model, the prevalence of malfunction status of IADL was significantly higher in those with depression, cognitive decline, and low muscle strength (PR 3.44; 3.85, and 4.55, respectively). In Model 1, depression and low muscle strength were significantly associated with IADL malfunction status (PR 2.47; 2.85). However, these comorbidities were not significantly associated with the IADL malfunction status in Model 2. There was no significant association between other comorbidities associated with diabetes and IADL malfunction status in any of the models (Online Resource: Supplementary Table 2).
Table 3.
Prevalence ratio (95% confidence interval) for malfunction status of IADL: determined by Poisson regression analysis
| Variables | n | Prevalence of IADL disability, n (%) | Crude model | Model 1 | Model 2 |
|---|---|---|---|---|---|
| PR (95% CI) | PR (95% CI) | PR (95% CI) | |||
| Diabetic nephropathy | 88 | 15 (17.0) | 1.55 (0.77–3.13) | 1.77 (0.92–3.42) | 1.78 (0.93–3.42) |
| Physical frailty | 49 | 11 (22.4) | 2.09 (1.04–4.20) | 0.88 (0.47–1.67) | 0.78 (0.42–1.45) |
| Depression | 18 | 7 (38.9) | 3.44* (1.69–7.01) | 2.47* (1.30–4.71) | 1.90 (0.89–4.05) |
| Cognitive decline | 30 | 11 (36.7) | 3.85* (1.99–7.47) | 1.31 (0.60–2.89) | 1.41 (0.65–3.05) |
| Low muscle strength | 55 | 16 (29.1) | 4.55* (2.07–9.99) | 2.85* (1.30–6.27) | 2.56 (1.15–5.68) |
Model 1: adjusted for propensity score (including age, gender, use of insulin, HbA1c, BMI, self-rated health, smoking status, physical activity level, academic history, economic condition, living status, other comorbidities). Model 2: Model 1 + score of Lubben Social Network Scale-6
IADL instrumental activities of daily living, PR prevalence ratio, CI confidence interval
*P < 0.01 (Bonferroni correction was used to adjusted the level of significance from 0.05 to 0.05/5 = 0.01)
Table 4 presents the association between comorbidities associated with diabetes and the malfunction status of intellectual activity. In the crude model, patients with depression had a significantly higher prevalence of the malfunction status intellectual disability (PR 1.70), but Models 1 and 2 did not show this association. The crude model showed no significant association between physical frailty and the malfunction status of intellectual activity; however, models 1 and 2 showed a significant association (PR 1.38; 1.35). Incidentally, there was no significant association between other comorbidities associated with diabetes and the malfunction status of intellectual activity in any model (Online Resource: Supplementary Table 3).
Table 4.
Prevalence ratio (95% confidence interval) for malfunction status of intellectual activity: determined by Poisson regression analysis
| Variables | n | Prevalence of intellectual activity disability, n (%) | Crude model | Model 1 | Model 2 |
|---|---|---|---|---|---|
| PR (95% CI) | PR (95% CI) | PR (95% CI) | |||
| Diabetic nephropathy | 88 | 44 (50.0) | 0.91 (0.69–1.19) | 0.80 (0.63–1.03) | 0.82 (0.64–1.05) |
| Physical frailty | 49 | 33 (67.3) | 1.39 (1.08–1.80) | 1.38* (1.10–1.74) | 1.35* (1.08–1.69) |
| Depression | 18 | 15 (83.3) | 1.70* (1.31–2.19) | 1.11 (0.86–1.43) | 1.02 (0.79–1.32) |
| Cognitive decline | 30 | 20 (66.7) | 1.32 (0.98–1.77) | 1.31 (1.00–1.72) | 1.30 (0.99–1.70) |
| Low muscle strength | 55 | 35 (63.6) | 1.24 (0.96–1.62) | 1.01 (0.79–1.30) | 1.00 (0.78–1.27) |
Model 1: adjusted for propensity score (including age, gender, use of insulin, HbA1c, BMI, self-rated health, smoking status, physical activity level, academic history, economic condition, living status, other comorbidities). Model 2: Model 1 + score of Lubben Social Network Scale-6
PR prevalence ratio, CI confidence interval
*P < 0.01 (Bonferroni correction was used to adjusted the level of significance from 0.05 to 0.05/5 = 0.01)
The associations between comorbidities associated with diabetes and the malfunction status of social role are shown in Table 5. In the crude model, patients with depression had a significantly higher prevalence of the malfunction status of social role (PR1.57), but Models 1 and 2 did not show this significant association. There was no significant association between other comorbidities associated with diabetes and the malfunction status of social role in any model (Online Resource: Supplementary Table 4).
Table 5.
Prevalence ratio (95% confidence interval) for malfunction status of social role: determined by Poisson regression analysis
| Variables | n | Prevalence of social role disability, n (%) | Crude model | Model 1 | Model 2 |
|---|---|---|---|---|---|
| PR (95% CI) | PR (95% CI) | PR (95% CI) | |||
| Diabetic nephropathy | 88 | 57 (64.8) | 1.03 (0.83–1.27) | 0.99 (0.80–1.21) | 1.04 (0.86–1.26) |
| Physical frailty | 49 | 34 (69.4) | 1.12 (0.89–1.40) | 1.01 (0.80–1.27) | 0.97 (0.79–1.18) |
| Depression | 18 | 17 (94.4) | 1.57* (1.33–1.85) | 1.32 (1.06–1.65) | 1.01 (0.83–1.23) |
| Cognitive decline | 29 | 23 (79.3) | 1.29 (1.04–1.61) | 1.20 (0.95–1.52) | 1.23 (0.97–1.55) |
| Low muscle strength | 55 | 39 (70.9) | 1.17 (0.94–1.46) | 1.05 (0.83–1.32) | 0.97 (0.79–1.20) |
Model 1: adjusted for propensity score (including age, gender, use of insulin, HbA1c, BMI, self-rated health, smoking status, physical activity level, academic history, economic condition, living status, other comorbidities). Model 2: Model 1 + score of Lubben Social Network Scale-6
PR prevalence ratio, CI confidence interval
*P < 0.01 (Bonferroni correction was used to adjusted the level of significance from 0.05 to 0.05/5 = 0.01)
Discussion
We investigated the cross-sectional relationship between comorbidities associated with diabetes and higher-level functional status measured using the TMIG-IC in older patients with type 2 diabetes mellitus who visited a general hospital in Japan. After adjusting for potential confounders (excluding social networks), depression was associated with the malfunction status of the TMIG-IC. The results of the subscale analysis confirmed the association between depression or low muscle strength and the malfunction status of IADL, physical frailty, and the malfunction status of intellectual activity. These results support a part of our hypotheses. Additionally, when the social network (LSNS-6 score) was added as a moderator variable, there was no association between depression or low muscle strength and higher-level functional status. As with our other hypotheses, social networks had a buffering effect on comorbidities associated with diabetes (depression and low muscle strength) and the deterioration of higher-level functional status. These findings suggest that older patients with type 2 diabetes mellitus have an increased risk of deterioration of higher-level functional status due to the comorbidity of depression, low muscle strength, and physical frailty. Furthermore, older patients with diabetes with good social networks (who often interact with others) may attenuate the effect of these comorbidities associated with diabetes on higher-level functional status.
Previous studies investigating the relationship between comorbidities associated with diabetes and higher-level functional status had limitations such as insufficient consideration of important confounders (e.g., subjective health, educational conditions, and socioeconomic factors). In this study, the propensity score was used to account for the potential confounders, which may indicate a result with higher internal validity. Additionally, we investigated the relationship between comorbidities associated with diabetes, including musculoskeletal problems, and the hierarchy of higher-level functional status in older patients with type 2 diabetes mellitus. In terms of clinical implications, these findings suggest the comorbidity for which enhanced care and preventive interventions should be recommended to maintain higher-level functional status. In other words, good control of comorbidities such as low muscle strength, frailty, and depression may contribute to preventing a decline in higher-level functional status that leads to future disability.
We have shown that comorbid depression increases the risk of the malfunction status of higher-level functional status in older patients with type 2 diabetes mellitus. Social role as a factor in higher-level functional status has been reported to be associated with depression [14]. Besides, it has been suggested that patients with diabetes with elevated depressive symptoms after the onset of diabetes may develop disability earlier [42]. Therefore, the results of this study are considered valid and support the importance of care and support for depression to prevent future disability at an early stage.
In the subscale analysis, low muscle strength was associated with a significantly higher prevalence of IADL malfunction status. Although there are differences in the definitions of low muscle strength, a previous study of community-based elderly also found an association between low muscle strength and difficulty with IADL [39]. Other previous studies have suggested that muscle strength may predict the deterioration of IADL more accurately than muscle mass [43], and that maintaining higher muscle strength contributes to the prevention of IADL disorders [44]. Therefore, measures to prevent low muscle strength (e.g. exercise and nutritional management) may be more important in older patients with type 2 diabetes mellitus. In addition, older patients with type 2 diabetes mellitus and physical frailty had a significantly higher risk of intellectual activity malfunction. A systematic review by Kojima et al. reported that physical frailty increases the risk of cerebrovascular dementia and Alzheimer's disease [45]. Although the theoretical pathway is unclear, physical frailty (easily fatigable, decreased activity) interferes with engaging in intellectual activities, which may be one of the causes of cognitive decline. In contrast, Abe et al. reported reverse causality, in which older people who engage in intellectual activities are more likely to improve from physical frailty to non-physical frailty [46]. Considering the bidirectional causal relationship between physical frailty and intellectual activity as mentioned above, encouraging not only coping with physical frailty but also intellectual activity itself may be useful in prevention of malfunction for higher-level functional status.
As hypothesized, patients with good social networks were unlikely to have depression or low muscle strength and had better higher-level functional status. In other word, these results suggest that the impact of depression or low muscle strength on higher-level functional status is attenuated in the presence of a good social network. Social interaction has been reported to have a buffering effect on the relationship between stress and physical or psychological health [47]. The present study similarly showed that social interaction mitigates the relationship of depression and low muscle strength with higher-level functional status deterioration. Contrary to our hypothesis, cognitive decline significantly increased the prevalence of higher-level functional status malfunction when social networks were included as moderators. Cognitive function greatly contributes to the maintenance of functional status [48]. Thus, patients with cognitive decline are likely to already have impaired higher-level functional status. Consequently, they may already receive support from others, and the amount of interaction may inevitably increase.
A strength of our study is its detailed investigation of the association between comorbidities associated with diabetes and each hierarchy of higher-level functional status, considering important confounders (subjective health, educational status, and social factors). However, this study has several limitations. First, the participants were limited to outpatients attending a general hospital (single centre), which may have been caused sampling bias in this study. Therefore, the results of this study may not apply to both inpatients and nursing home residents. Second, due to cross-sectional study, the causal relationship between comorbidities associated with diabetes and higher-level functional status remains unclear. Therefore, clarifying this causal relationship requires a longitudinal or intervention studies with a large sample of participants from multiple centres. Third, this study did not consider the dietary intake status. Dietary intake affects blood sugar control and various comorbidities associated with diabetes [49], and its relationship with higher-level functional status has also been reported [50]. In the future, it is necessary to consider the dietary intake status.
Conclusion
In conclusion, comorbidities of depression, low muscle strength, and physical frailty were associated with the malfunction status of higher-level functional status in elderly patients with type 2 diabetes mellitus. Expanding social networks reduced the risk of depression, low muscle strength, and physical frailty were associated with the malfunction status of higher-level functional status in older patients with type 2 diabetes mellitus. Care or prevention of these comorbidities associated with diabetes is critical to delay the development of future disabilities. Additionally, expanding social networks may reduce the impact of comorbidities such as depression and low muscle strength on higher-level functional status.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We thank the study participants and research team members for their cooperation in this study. In addition, the authors would like to thank Editage (http://www.editage.jp) for English language editing.
Author contributions
TS: conceptualization, methodology, formal analysis, investigation, writing—original draft, visualization. KK, NT: investigation. MH, HK, HY: investigation, writing—review, and editing. ST, KT, TT, TO: writing—review and editing.
Funding
This study was supported by a grant from the Japan Association for Diabetes Education and Care [grant number 2021-FND-009] and the National Center for Global Health and Medicine [23A 3002].
Data availability
The data that support the findings of this study are available from the corresponding author on reasonable request.
Declarations
Conflict of interest
The authors have no relevant financial or non-financial interests to disclose.
Ethics approval
Approval was obtained from the ethics board of the National Center for Global Health and Medicine (approval number: NCGM-S-004396-00). The procedures used in this study adhere to the tenets of the Declaration of Helsinki.
Consent to participate
Informed consent was obtained from all the patients included in the study.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.(1984) The uses of epidemiology in the study of the elderly. Report of a WHO scientific group on the epidemiology of aging. World Health Organ Tech Rep Ser 706:1–84 [PubMed]
- 2.Lawton MP (1972) Assessing the competence of older people. In: Kent DP, Kastenbaum R, Sherwood S (eds) Research, planning, and action for the elderly: the power and potential of social science. Behavioral Publications, New York, pp 122–143 [Google Scholar]
- 3.Nagata H, Miura K, Tanaka S et al (2023) Relationship of higher-level functional capacity with long-term mortality in Japanese older people: NIPPON DATA90. J Epidemiol 33:136–141. 10.2188/jea.JE20210077 10.2188/jea.JE20210077 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Murakami K, Tsubota-Utsugi M, Satoh M et al (2016) Impaired higher-level functional capacity as a predictor of stroke in community-dwelling older adults: the Ohasama study. Stroke 47:323–328. 10.1161/STROKEAHA.115.011131 10.1161/STROKEAHA.115.011131 [DOI] [PubMed] [Google Scholar]
- 5.Taniguchi Y, Kitamura A, Nofuji Y et al (2019) Association of trajectories of higher-level functional capacity with mortality and medical and long-term care costs among community-dwelling older Japanese. J Gerontol A Biol Sci Med Sci 74:211–218. 10.1093/gerona/gly024 10.1093/gerona/gly024 [DOI] [PubMed] [Google Scholar]
- 6.(2018) Essence and perspective of the JGS/JDS clinical practice guideline for the treatment of diabetes in the elderly. Nihon Ronen Igakkai Zasshi 55:1–12 (in Japanese). 10.3143/geriatrics.55.1 [DOI] [PubMed]
- 7.Ahmed HA, Alan JS (2019) Diabetes in the elderly. Medicine 47(2):119–122. 10.1016/j.mpmed.2018.11.007 10.1016/j.mpmed.2018.11.007 [DOI] [Google Scholar]
- 8.Mori H, Kuroda A, Yoshida S et al (2021) High prevalence and clinical impact of dynapenia and sarcopenia in Japanese patients with type 1 and type 2 diabetes: findings from the Impact of Diabetes Mellitus on Dynapenia study. J Diabetes Investig 12:1050–1059. 10.1111/jdi.13436 10.1111/jdi.13436 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Chen H, Huang X, Dong M, Wen S, Zhou L, Yuan X (2023) The association between sarcopenia and diabetes: from pathophysiology mechanism to therapeutic strategy. Diabetes Metab Syndr Obes 16:1541–1554. 10.2147/DMSO.S410834 10.2147/DMSO.S410834 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hofbauer LC, Busse B, Eastell R et al (2022) Bone fragility in diabetes: novel concepts and clinical implications. Lancet Diabetes Endocrinol 10:207–220. 10.1016/S2213-8587(21)00347-8 10.1016/S2213-8587(21)00347-8 [DOI] [PubMed] [Google Scholar]
- 11.Wong E, Backholer K, Gearon E et al (2013) Diabetes and risk of physical disability in adults: a systematic review and meta-analysis. Lancet Diabetes Endocrinol 1:106–114. 10.1016/S2213-8587(13)70046-9 10.1016/S2213-8587(13)70046-9 [DOI] [PubMed] [Google Scholar]
- 12.Sakurai T, Iimuro S, Sakamaki K et al (2012) Risk factors for a 6-year decline in physical disability and functional limitations among elderly people with type 2 diabetes in the Japanese Elderly Diabetes Intervention Trial. Geriatr Gerontol Int 12(suppl 1):117–126. 10.1111/j.1447-0594.2011.00819.x 10.1111/j.1447-0594.2011.00819.x [DOI] [PubMed] [Google Scholar]
- 13.Ida S, Murata K, Nakadachi D et al (2018) Association between dynapenia and decline in higher-level functional capacity in older men with diabetes. Geriatr Gerontol Int 18:1393–1397. 10.1111/ggi.13498 10.1111/ggi.13498 [DOI] [PubMed] [Google Scholar]
- 14.Ida S, Kaneko R, Imataka K, Fujiwara R, Murata K (2021) Are diabetes-related factors associated with the social roles of elderly patients with diabetes? J Diabetes Complicat 35(1):107759. 10.1016/j.jdiacomp.2020.107759 10.1016/j.jdiacomp.2020.107759 [DOI] [PubMed] [Google Scholar]
- 15.Tough H, Siegrist J, Fekete C (2017) Social relationships, mental health and wellbeing in physical disability: a systematic review. BMC Public Health 17:414. 10.1186/s12889-017-4308-6 10.1186/s12889-017-4308-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Qin W, Wang Y, Cho S (2021) Neighborhood social cohesion, physical disorder, and daily activity limitations among community-dwelling older adults. Arch Gerontol Geriatr 93:104295. 10.1016/j.archger.2020.104295 10.1016/j.archger.2020.104295 [DOI] [PubMed] [Google Scholar]
- 17.Ida S, Kaneko R, Imataka K et al (2020) Frequency of social isolation and homeboundness and their relationships with high-level functional capacity in elderly diabetic patients. Clin Interv Aging 15:1439–1447. 10.2147/CIA.S244529 10.2147/CIA.S244529 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Clark BC, Manini TM (2010) Functional consequences of sarcopenia and dynapenia in the elderly. Curr Opin Clin Nutr Metab Care 13:271–276. 10.1097/MCO.0b013e328337819e 10.1097/MCO.0b013e328337819e [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Koyano W, Shibata H, Nakazato K, Haga H, Suyama Y (1991) Measurement of competence: Reliability and validity of the TMIG Index of Competence. Arch Gerontol Geriatr 13(2):103–116. 10.1016/0167-4943(91)90053-s 10.1016/0167-4943(91)90053-s [DOI] [PubMed] [Google Scholar]
- 20.Shibata H, Sugisawa H, Watanabe S (2001) Functional capacity in elderly Japanese living in the community. Geriatr Gerontol Int 1(1–2):8–13. 10.1046/j.1444-1586.2001.00004.x 10.1046/j.1444-1586.2001.00004.x [DOI] [Google Scholar]
- 21.Fujiwara Y, Shinkai S, Amano H et al (2003) Test-retest variation in the Tokyo Metropolitan Institute of Gerontology Index of Competence in community-dwelling older people independent in daily living toward individual assessment of functional capacity. Nihon Koshu Eisei Zasshi 50:360–367 (in Japanese) [PubMed] [Google Scholar]
- 22.Tsubota-Utsugi M, Ito-Sato R, Ohkubo T et al (2011) Health behaviors as predictors for declines in higher-level functional capacity in older adults: the Ohasama study. J Am Geriatr Soc 59:1993–2000. 10.1111/j.1532-5415.2011.03633.x 10.1111/j.1532-5415.2011.03633.x [DOI] [PubMed] [Google Scholar]
- 23.Yasuda H, Sanada M, Kitada K et al (2007) Rationale and usefulness of newly devised abbreviated diagnostic criteria and staging for diabetic polyneuropathy. Diabetes Res Clin Pract 77(suppl 1):S178–S183. 10.1016/j.diabres.2007.01.053 10.1016/j.diabres.2007.01.053 [DOI] [PubMed] [Google Scholar]
- 24.Davis MD (1965) Vitreous contraction in proliferative diabetic retinopathy. Arch Ophthalmol 74:741–751. 10.1001/archopht.1965.00970040743003 10.1001/archopht.1965.00970040743003 [DOI] [PubMed] [Google Scholar]
- 25.Haneda M, Utsunomiya K, Koya D et al (2015) A new classification of Diabetic Nephropathy 2014: a report from Joint Committee on Diabetic Nephropathy. Clin Exp Nephrol 19:1–5. 10.1007/s10157-014-1057-z 10.1007/s10157-014-1057-z [DOI] [PubMed] [Google Scholar]
- 26.Folstein MF, Folstein SE, McHugh PR (1975) ‘Mini-mental state’. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12:189–198. 10.1016/0022-3956(75)90026-6 10.1016/0022-3956(75)90026-6 [DOI] [PubMed] [Google Scholar]
- 27.Tsoi KK, Chan JY, Hirai HW, Wong SY, Kwok TC (2015) Cognitive tests to detect dementia: a systematic review and meta-analysis. JAMA Intern Med 175:1450–1458. 10.1001/jamainternmed.2015.2152 10.1001/jamainternmed.2015.2152 [DOI] [PubMed] [Google Scholar]
- 28.Yesavage JA, Sheikh JI (1986) 9/Geriatric Depression Scale (GDS): recent evidence and development of a shorter version. Clin Gerontol. 10.1300/J018v05n01_09 10.1300/J018v05n01_09 [DOI] [Google Scholar]
- 29.Shin C, Park MH, Lee SH et al (2019) Usefulness of the 15-item geriatric depression scale (GDS-15) for classifying minor and major depressive disorders among community-dwelling elders. J Affect Disord 259:370–375. 10.1016/j.jad.2019.08.053 10.1016/j.jad.2019.08.053 [DOI] [PubMed] [Google Scholar]
- 30.Satake S, Arai H (2020) The revised Japanese version of the Cardiovascular Health Study criteria (revised J-CHS criteria). Geriatr Gerontol Int 20:992–993. 10.1111/ggi.14005 10.1111/ggi.14005 [DOI] [PubMed] [Google Scholar]
- 31.Wen Z, Gu J, Chen R et al (2023) Handgrip strength and muscle quality: results from the National Health and Nutrition Examination Survey database. J Clin Med 12:3184. 10.3390/jcm12093184 10.3390/jcm12093184 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Shinkai S, Watanabe S, Kumagai S et al (2000) Walking speed as a good predictor for the onset of functional dependence in a Japanese rural community population. Age Ageing 29:441–446. 10.1093/ageing/29.5.441 10.1093/ageing/29.5.441 [DOI] [PubMed] [Google Scholar]
- 33.Mori K, Murata S, Goda A et al (2022) Gait characteristics of dynapenia, sarcopenia, and presarcopenia in community-dwelling Japanese older women: a cross-sectional study. Healthcare (Basel) 10:1905. 10.3390/healthcare10101905 10.3390/healthcare10101905 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Stuck AE, Walthert JM, Nikolaus T, Büla CJ, Hohmann C, Beck JC (1999) Risk factors for functional status decline in community-living elderly people: a systematic literature review. Soc Sci Med 48:445–469. 10.1016/s0277-9536(98)00370-0 10.1016/s0277-9536(98)00370-0 [DOI] [PubMed] [Google Scholar]
- 35.Washburn RA, Smith KW, Jette AM, Janney CA (1993) The Physical Activity Scale for the Elderly (PASE): development and evaluation. J Clin Epidemiol 46:153–162. 10.1016/0895-4356(93)90053-4 10.1016/0895-4356(93)90053-4 [DOI] [PubMed] [Google Scholar]
- 36.Lubben J, Blozik E, Gillmann G et al (2006) Performance of an abbreviated version of the Lubben Social Network Scale among three European Community-dwelling older adult populations. Gerontologist 46:503–513. 10.1093/geront/46.4.503 10.1093/geront/46.4.503 [DOI] [PubMed] [Google Scholar]
- 37.Zhang J, Yu KF (1998) What’s the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA 280:1690–1691. 10.1001/jama.280.19.1690 10.1001/jama.280.19.1690 [DOI] [PubMed] [Google Scholar]
- 38.Barros AJ, Hirakata VN (2003) Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio. BMC Med Res Methodol 3:21. 10.1186/1471-2288-3-21 10.1186/1471-2288-3-21 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Iwamura M, Kanauchi M (2017) A cross-sectional study of the association between dynapenia and higher-level functional capacity in daily living in community-dwelling older adults in Japan. BMC Geriatr 17:1. 10.1186/s12877-016-0400-5 10.1186/s12877-016-0400-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Drake C (1993) Effects of misspecification of the propensity score on estimators of treatment effect. Biometrics 49:1231–1236. 10.2307/2532266 10.2307/2532266 [DOI] [Google Scholar]
- 41.Cepeda MS, Boston R, Farrar JT, Strom BL (2003) Comparison of logistic regression versus propensity score when the number of events is low and there are multiple confounders. Am J Epidemiol 158:280–287. 10.1093/aje/kwg115 10.1093/aje/kwg115 [DOI] [PubMed] [Google Scholar]
- 42.Wu CY, Terhorst L, Karp JF, Skidmore ER, Rodakowski J (2018) Trajectory of disability in older adults with newly diagnosed diabetes: role of elevated depressive symptoms. Diabetes Care 41:2072–2078. 10.2337/dc18-0007 10.2337/dc18-0007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Wang DXM, Yao J, Zirek Y, Reijnierse EM, Maier AB (2020) Muscle mass, strength, and physical performance predicting activities of daily living: a meta-analysis. J Cachexia Sarcopenia Muscle 11:3–25. 10.1002/jcsm.12502 10.1002/jcsm.12502 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.McGrath R, Robinson-Lane SG, Peterson MD, Bailey RR, Vincent BM (2018) Muscle strength and functional limitations: preserving function in older Mexican Americans. J Am Med Dir Assoc 19:391–398. 10.1016/j.jamda.2017.12.011 10.1016/j.jamda.2017.12.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Kojima G, Taniguchi Y, Iliffe S, Walters K (2016) Frailty as a predictor of Alzheimer disease, vascular dementia, and all dementia among community-dwelling older people: a systematic review and meta-analysis. J Am Med Dir Assoc 17:881–888. 10.1016/j.jamda.2016.05.013 10.1016/j.jamda.2016.05.013 [DOI] [PubMed] [Google Scholar]
- 46.Abe T, Nofuji Y, Seino S et al (2020) Healthy lifestyle behaviors and transitions in frailty status among independent community-dwelling older adults: the Yabu cohort study. Maturitas 136:54–59. 10.1016/j.maturitas.2020.04.007 10.1016/j.maturitas.2020.04.007 [DOI] [PubMed] [Google Scholar]
- 47.Thoits PA (2011) Mechanisms linking social ties and support to physical and mental health. J Health Soc Behav 52:145–161. 10.1177/0022146510395592 10.1177/0022146510395592 [DOI] [PubMed] [Google Scholar]
- 48.Lee Y, Kim JH, Lee KJ, Han G, Kim JL (2005) Association of cognitive status with functional limitation and disability in older adults. Aging Clin Exp Res 17:20–28. 10.1007/BF03337716 10.1007/BF03337716 [DOI] [PubMed] [Google Scholar]
- 49.Tamura Y, Omura T, Toyoshima K, Araki A (2020) Nutrition management in older adults with diabetes: a review on the importance of shifting prevention strategies from metabolic syndrome to frailty. Nutrients 12:3367. 10.3390/nu12113367 10.3390/nu12113367 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Kumagai S, Watanabe S, Shibata H et al (2003) Effects of dietary variety on declines in high-level functional capacity in elderly people living in a community. Nihon Koshu Eisei Zasshi 50:1117–1124 (in Japanese) [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author on reasonable request.