中込 敦士

There is a growing, but inconclusive, evidence on the beneficial influence of the Internet on the health/wellbeing of older people. This study aims to examine outcomes-wide associations between frequency of internet use and subsequent health/well-being among older individuals. We used the three-wave data (2013, 2016, and 2019) from the Japan Gerontological Evaluation Study. Our exposure was frequency of internet use (not at all/ use a few times a month/use a few times a week/use almost every day) in 2016. We assessed 34 health/wellbeing outcomes in 2019 across six dimensions. Pre-exposure characteristics and prior outcome levels in 2013 were adjusted. We included 5879 respondents for 4 outcomes (death, dementia, and functional disability) and 4232 respondents for 30 other outcomes. Using the Internet almost every day (vs. not using at all) in 2016 was associated with some outcomes in 2019 in three dimensions: physical/cognitive health (better instrumental activities of daily living), social well-being (more frequent participation in sports groups, meeting friends more frequently, and seeing more friends within a month), and health behaviors (receiving health screening) (p < 0.05 for these associations after Bonferroni correction). It was not associated with any outcomes of three other dimensions: psychological distress, subjective well-being, or pro-social/altruistic behaviors.

BACKGROUND: In-person religious service attendance has been linked to favorable health and well-being outcomes. However, little research has examined whether online religious participation improves these outcomes, especially when in-person attendance is suspended. METHODS: Using longitudinal data of 8951 UK adults, this study prospectively examined the association between frequency of online religious participation during the stringent lockdown in the UK (23 March -13 May 2020) and 21 indicators of psychological well-being, social well-being, pro-social/altruistic behaviors, psychological distress, and health behaviors. All analyses adjusted for baseline socio-demographic characteristics, pre-pandemic in-person religious service attendance, and prior values of the outcome variables whenever data were available. Bonferroni correction was used to correct for multiple testing. RESULTS: Individuals with online religious participation of ≥1/week (v. those with no participation at all) during the lockdown had a lower prevalence of thoughts of self-harm in week 20 (odds ratio 0.24; 95% CI 0.09-0.62). Online religious participation of <1/week (v. no participation) was associated with higher life satisfaction (standardized β = 0.25; 0.11-0.39) and happiness (standardized β = 0.25; 0.08-0.42). However, there was little evidence for the associations between online religious participation and all other outcomes (e.g. depressive symptoms and anxiety). CONCLUSIONS: There was evidence that online religious participation during the lockdown was associated with some subsequent health and well-being outcomes. Future studies should examine mechanisms underlying the inconsistent results for online v. in-person religious service attendance and also use data from non-pandemic situations.

Evidence on the association between internet usage and incidence of depression remains mixed. We examined the associations between different categories of internet usage and developing clinical depression. We used data from the 2013 and 2016 waves of the Japan Gerontological Evaluation Study (JAGES) comprising 12,333 physically and cognitively independent adults aged ≥65 years. Participants were engaged in seven categories of internet usage: communication with friends/family, social media, information collection about health/medicine, searching for medical facilities, purchase of drugs and vitamins, shopping, and banking. We found that internet use for communication had a protective influence on the probability of developing clinical depression defined as the Geriatric Depression Scale scores ≥5 or self-reported diagnosed depression. Our findings support the role of online communication with friends/family in preventing clinical depression among older people. Online communication could be particularly useful in the COVID-19 crisis because many families are geographically dispersed and/or socially distanced.

General health checks to detect cardiovascular risk factors form part of routine health care in many countries. Cardiovascular disease and dementia share a number of risk factors; however it remains unclear whether general health checks can reduce the incidence of dementia. We used longitudinal data from the Japan Gerontological Evaluation Study with up to 6.4 years follow-up (from 2010 to 2016). A total of 31,012 disability-free adults aged 65-74 were included. The outcome variable was dementia onset certified under the national long-term care insurance system. The treatment variable was defined as having health checks within the past 12 months prior to the baseline survey in 2010. After multiple imputation, we performed propensity score matching (PSM) to exclude off-support individuals who were the least likely to have health checks. We also performed inverse probability treatment weighting (IPTW) to estimate the effect of the treatment if everyone within the population was compliant to health checks. The hazard ratios for dementia onset among those reporting health checks within the previous year was 0.89 (95% confidence interval (CI): 0.78, 1.02) in the PSM analysis and 0.84 (95% CI: 0.75, 0.95) in the IPTW analysis. We then estimated the effect of health checks on 5-year incident dementia. The 5-year cumulative incidence difference based on the PSM analysis was -0.0046 (95%CI: -0.0101, 0.0009), while that based on the IPTW analysis was -0.0046 (95%CI, -0.0090, -0.0002). The PSM and IPTW approaches yielded similar findings that the incidence of dementia was lower among people having health checks.

OBJECTIVES: Widowhood is associated with increased risks of depression in the surviving spouse. We examined whether an increase in individual-level social capital mitigates the adverse impact of widowhood on depressive symptoms. METHODS: We used data from the 2013/2016 waves of the Japan Gerontological Evaluation Study of functionally independent adults aged 65 years or older (men: n = 20,853; women: n =16,858). Fixed-effects regression was applied to examine the potential buffering effects of changes in social capital on changes in depressive symptoms following widowhood among married people living with their spouse and/or others at baseline. RESULTS: Widowhood had a deleterious impact on depressive symptoms particularly among men who ended up living alone following their spouse's death. Fixed-effects models revealed that an increase in informal socializing and social participation might buffer the effects of spousal bereavement on depressive symptoms among men who became widowed during the first two years of follow-up and ended up living alone. DISCUSSION: An increase in structural social capital may mitigate the impact of spousal bereavement on depressive symptoms. However, the associations vary by gender, living arrangement, and time since widowhood. Intensive efforts should be directed toward connecting the vulnerable group, widowed men living alone, to sources of social capital.

Widowhood and living alone are linked to increased risk of depression. We examined prospectively whether community-level social capital can mitigate the adverse impact of widowhood and living alone on depressive symptoms. We used data of the Japan Gerontological Evaluation Study of functionally independent adults aged 65 years or older. Three waves of surveys were collected in 2010, 2013 and 2016. We conducted gender-stratified multilevel linear regression to examine the moderating effects of community-level social capital on depressive symptoms (as assessed by the 15-point Geriatric Depression Scale) associated with widowhood and living alone. Widowhood in the past 12 months in combination with living alone was associated with a marked worsening in depressive symptoms among men (beta coefficient = 1.67; 95% confidence interval: 1.38, 1.95). Community-level civic participation, but not social cohesion or reciprocity, was associated with lower depressive symptoms in men and women. In addition, community-level civic participation moderated the association between depressive symptoms and recent widowhood/living alone among men (coefficient per 1 standard deviation = -0.30; 95% confidence interval: -0.59, -0.02). We found no significant effect modification of community-level social capital on depressive symptoms associated with widowhood and living alone among women. Communities with greater civic participation appear to mitigate the onset of depressive symptoms among recently widowed men living alone. Promotion of community activities might be an effective community-level intervention to promote mental health in this vulnerable group.