BACKGROUND: Japanese people sleep less compared to other countries around the world. Using a large nationally representative survey in 2019 and 2022, we investigated whether sleep duration and nonrestorative sleep (NRS) among Japanese people have improved or worsened due to the COVID-19 pandemic. METHODS: Data were drawn from the Comprehensive Survey of Living Conditions, a nationwide cross-sectional sample based on self-administered questionnaires. We analyzed 426,510 people in 2019 and 375,578 people in 2022 aged ≥20 living in the community. The generalized estimating equations of the multivariable Poisson regression models were used to estimate adjusted prevalence of NRS by survey year. Potential confounders included gender, age, marital status, family size, housing tenure, equivalent household expenditures, education, employment status, illness under treatment, lifestyle behaviors (i.e., smoking, drinking, dietary, and fitness habits), mental health, and sleep duration. RESULTS: Among the study participants, 35.7% slept less than 6 hours and 20.9% had NRS. Regarding sleep duration, the prevalence of sleep duration of less than 6 hours was significantly lower in 2022 than in 2019 for both men and women. By gender and age, the prevalence of short sleep duration (<6 hours) significantly decreased for both men and women under the age of 49, but increased significantly for men aged ≥50 and women aged ≥75. Regarding NRS, the prevalence of NRS was significantly lower in 2022 than in 2019 regardless of gender and age: Prevalence among men was 21.4% in 2019 and 18.8% in 2022, and prevalence among women was 23.7% in 2019 and 21.2% in 2022. After adjustment for potential confounders, the difference between the 2022 NRS prevalence and the 2019 NRS prevalence was minus 1.64 percent point (pp) (95% confidence interval minus 1.82 pp to minus 1.46 pp, P < 0.001), showing a significant decrease in the 2022 NRS prevalence. A significant improvement of NRS was independent of the prevalence of short sleep duration, age, gender, and employment status. CONCLUSIONS The prevalence of NRS among the general population in Japan was significantly reduced during the COVID-19 pandemic compared to before the COVID-19 pandemic. We need to monitor whether this decline continues or returns to pre-pandemic levels.
Adult ; Aged ; Female ; Humans ; Male ; Middle Aged ; Young Adult ; COVID-19/epidemiology* ; Cross-Sectional Studies ; Japan/epidemiology* ; Pandemics ; Prevalence ; SARS-CoV-2 ; Sleep ; Surveys and Questionnaires

BACKGROUND: Older adults in Japan have the highest drowning mortality rate globally due to frequent bathing practices. Low outdoor temperatures have been linked to bath-related deaths; however, previous studies employed limited statistical models and focused on a single prefecture. Given Japan's aging population, preventing bath-related deaths is a public health priority. This study aimed to analyze the association between outdoor temperature and bath-related drowning deaths across Japan from 1995 to 2020 (n = 110,938), examining regional variations and identifying contributing prefectural characteristics. METHODS: Daily counts of bath-related drowning deaths per prefecture were matched with daily mean temperature data from the Japan Meteorological Agency. Prefecture-level demographic and environmental data were obtained from Japan's Official Statistics. We applied a generalized additive mixed model to examine the association between daily mean temperature and bath-related drowning death risk. Meta-regression was used to identify prefecture-level modifiers. RESULTS: Bath-related drowning death risk peaked at a daily mean temperature of 1.8 °C (relative risk [RR] 9.7, 95% confidence interval [CI]: 9.5-9.9), compared with the lowest risk at 30.3 °C. The association was stronger at mid-range temperatures, particularly among males and individuals aged ≥65 years. Among prefectures, Kagoshima-the southernmost prefecture on Japan's main islands-had the highest maximum RR at 19.6 (95% CI: 16.2-23.6), while Hokkaido-the northernmost prefecture-had the lowest at 3.8 (95% CI: 3.4-4.3). Prefecture-level factors that strengthened this relationship included a lower prevalence of double-pane windows as a proxy of housing insulation and higher annual mean temperatures with ratio of RR change per one standard deviation increase of 0.76 (95% CI: 0.69-0.83) and 1.27 (95% CI: 1.18-1.37), respectively. CONCLUSIONS Warmer prefectures in southern regions exhibited greater maximum-to-minimum risk ratios compared to cooler northern prefectures. This paradoxical finding underscores the importance of region-specific interventions to reduce bath-related deaths.
Japan/epidemiology* ; Humans ; Drowning/epidemiology* ; Male ; Aged ; Female ; Baths/statistics & numerical data* ; Middle Aged ; Temperature ; Aged, 80 and over ; Adult ; Young Adult ; Risk Factors ; Adolescent

BACKGROUND: Globally, Japan has the highest drowning mortality among older adults, largely because of bathing customs. Although this mortality rate peaks in winter, the nationwide impact of outdoor temperature has not been quantified, and whether specific days carry greater risks for bathtub drowning deaths remains unclear. Therefore, we aimed to address these gaps using nationwide data from Japan. METHODS: We collected daily data on outdoor temperatures and bathtub drowning deaths (from death certificates), along with population data, across 47 prefectures from 1995-2020. A time series regression model incorporating a cyclic spline for day-of-year and a cross-basis function for outdoor temperature was used to estimate seasonality and temperature attributable fractions (AFs). Prefecture-specific estimates were pooled using meta-analysis. National holidays were defined by the Act on National Holidays. RESULTS: During the study period, 99,930 home bathtub drowning deaths were recorded. The AF for seasonality modelled with a cyclic spline for day-of-year was 77.8% (empirical confidence interval [eCI]: 76.7-78.8%), which decreased to 15.3% (eCI: 13.1-18.0%) after adjusting for outdoor temperature, indicating that outdoor temperature accounted for 80.3% of the seasonal effect. Elevated risks were observed on Sundays (relative risk = 1.16, 95% CI: 1.12-1.20), holidays (1.12, 95% CI: 1.08-1.16), New Year's Day (1.72, 95% CI: 1.61-1.84), and New Year's Eve (1.63, 95% CI: 1.52-1.74) in the adjusted model, which included a cyclic spline for day-of-year and a cross-basis function for outdoor temperature. CONCLUSION Our findings highlight the importance of mitigating the impact of outdoor temperature on bath-related death risk. Identifying high-risk days can be used to help develop targeted preventive strategies.
Seasons ; Humans ; Drowning/epidemiology* ; Japan/epidemiology* ; Temperature ; Aged ; Male ; Adult ; Female ; Middle Aged ; Child ; Young Adult ; Adolescent ; Aged, 80 and over ; Child, Preschool

BACKGROUND: The relationship between leisure activities (LA) in old age and prevention of disability has not been fully investigated, and age and gender differences of these relationships are unknown. This study aimed to investigate whether physical and cognitive LA predicted incident disability among community-dwelling older adults by age and gender. METHODS: We prospectively observed 8,275 residents aged 65 or above without disability at baseline for 3 years. Incident disability was defined as a new certification of the public long-term care insurance system. LA were classified into two types: physical LA and cognitive LA. The frequency of LA was categorized into frequent (i.e., once a week or more), moderate (i.e., monthly or yearly), and non-engagement. Covariates included age, gender, family number, education, perceived economic situation, body mass index, chronic medical conditions, alcohol consumption, smoking status, regular dental visits, depression, cognitive functioning, and social participation. Multivariable Poisson regression models were used to estimate adjusted cumulative incidence ratio (CIR) and 95% confidence interval (CI) for incident disability. We performed stratified analyses by age groups (i.e., the young-old aged 65-74 and the old-old aged 75-97) and gender (i.e., men and women). RESULTS: The 3-year cumulative incidence of disability was 7.5%. After adjustment for covariates and mutual adjustment for both types of LA, a significant dose-response relationship between more frequent LA and lower risk of incident disability was found in young-old physical LA (P-trend < 0.001), in old-old cognitive LA (P-trend = 0.012), in male cognitive LA (P-trend = 0.006), and in female physical LA (P-trend = 0.030). Compared with people without LA, adjusted CIR (95% CI) of frequent LA was 0.47 (0.30-0.74) in young-old physical, 0.75 (0.58-0.96) in old-old cognitive, 0.65 (0.46-0.89) in male cognitive, and 0.70 (0.52-0.95) in female physical. Regarding the effect modification according to age and gender, only interaction between age and physical LA significantly prevented incident disability (P for interaction = 0.019). CONCLUSION We found age differences in the association of physical LA with incident disability among community-dwelling older adults. An effective measure to prevent long-term care in the community would be to recommend frequent physical LA for the young-old.
Aged ; Disabled Persons ; Exercise ; Female ; Humans ; Independent Living ; Leisure Activities ; Male ; Middle Aged ; Social Participation

BACKGROUND: Community health activities by public health nurses (PHNs) are known to improve lifestyle habits of local residents, and may encourage the practice of infectious disease prevention behaviors during the COVID-19 pandemic. We investigated the association between prefecture-level COVID-19 incidence rate and the number of PHNs per population in Japan, by the COVID-19 variant type. METHODS: Our data were based on government surveys where prefectural-level data are accessible to the public. The outcome variable was the COVID-19 incidence rate (i.e., the cumulative number of COVID-19 cases per 100,000 population for each variant type in 47 prefectures). The explanatory variable was the number of PHNs per 100,000 population by prefecture. Covariates included socioeconomic factors, regional characteristics, healthcare resources, and health behaviors. The generalized estimating equations of the multivariable Poisson regression models were used to estimate adjusted incidence rate ratio (IRR) and 95% confidence interval (CI) for the COVID-19 cases. We performed stratified analyses by variant type (i.e., wild type, alpha variant, and delta variant). RESULTS: A total of 1,705,224 confirmed COVID-19 cases (1351.6 per 100,000 population) in Japan were reported as of September 30, 2021. The number of PHNs per 100,000 population in Japan was 41.9. Multivariable Poisson regression models showed that a lower number of PHNs per population was associated with higher IRR of COVID-19. Among all COVID-19 cases, compared to the highest quintile group of the number of PHNs per population, the adjusted IRR of the lowest quintile group was consistently significant in the models adjusting for socioeconomic factors (IRR: 3.76, 95% CI: 2.55-5.54), regional characteristics (1.73, 1.28-2.34), healthcare resources (3.88, 2.45-6.16), and health behaviors (2.17, 1.39-3.37). These significant associations were unaffected by the variant type of COVID-19. CONCLUSION We found that the COVID-19 incidence rate was higher in prefectures with fewer PHNs per population, regardless of the COVID-19 variant type. By increasing the number of PHNs, it may be possible to contain the spread of COVID-19 in Japan and provide an effective human resource to combat emerging infectious diseases in the future.
COVID-19/epidemiology* ; Humans ; Incidence ; Japan/epidemiology* ; Nurses, Public Health ; Pandemics ; SARS-CoV-2

BACKGROUND: We aimed to examine the cross-sectional association between occupational class and self-reported low back pain (LBP) in a representative sample of the Japanese general population. METHODS: We used anonymized data from a nationwide survey (31,443 men and 35,870 women aged ≥ 20). Occupational class variables included working status, occupation, employment status, and company size (number of employees). Covariates included age, socio-economic status, lifestyle, and comorbidities. Poisson regression models stratified by gender were used to estimate adjusted prevalence ratio (APR) and 95% confidence interval (CI) for self-reported LBP. RESULTS: The prevalence of self-reported LBP was 11.7% in men and 14.2% in women. After adjustment for covariates and mutual adjustment for all occupational class variables, among both genders, agricultural/forestry/fishery workers and non-workers had a significantly higher prevalence of self-reported LBP: APR (95% CI) of agricultural/forestry/fishery was 1.36 (1.08-1.70) in men and 1.59 (1.30-1.93) in women; that of non-workers was 1.42 (1.18-1.70) in men and 1.23 (1.08-1.40) in women. Among men, non-regular employees were more likely to have self-reported LBP than regular employees: APR (95% CI) was 1.25 (1.07-1.46) in part-timers and casual staff and 1.18 (1.03-1.35) in other types of non-regular employees. Moreover, compared to men working at companies with ≥ 100 employees, men working at companies with 30-99 employees had a significantly higher prevalence of self-reported LBP (APR 1.17; 95% CI, 1.03-1.34). Among women, professionals and technicians (1.26; 1.11-1.43) and sales workers (1.22; 1.04-1.43) had a significantly higher prevalence of self-reported LBP than clerks. Neither employment status nor company size was associated with self-reported LBP in women. After stratified analyses by age group, similar patterns were observed in participants aged 20-64, but not in those aged ≥ 65. CONCLUSION Our results suggest that self-reported LBP is highly prevalent among agricultural/forestry/fishery workers and the unemployed, regardless of gender, and that there are also gender differences in the association of occupational class factors with self-reported LBP. It is necessary, therefore, to take preventive measures against LBP based on gender and occupational class factors in Japan.
Adult ; Aged ; Cross-Sectional Studies ; Employment/statistics & numerical data* ; Female ; Humans ; Japan/epidemiology* ; Low Back Pain/epidemiology* ; Male ; Middle Aged ; Occupational Diseases/epidemiology* ; Occupations/statistics & numerical data* ; Prevalence ; Self Report ; Sex Distribution

BACKGROUND: Maintenance of instrumental activities of daily living (IADL) and social role (SR) is crucial to keep independent life because the decline in SR and IADL was a significant predictor of dependence in basic ADL in later. The independent effect of physical and cultural leisure activities and their effect modification on the IADL remains unknown. METHODS: We prospectively observed 3241 elderly with intact IADL at baseline for 5 years. Higher level functional capacity such as IADL and SR was assessed using the Tokyo Metropolitan Institute of Gerontology Index of competence (TMIG index). RESULTS: The mean age of the participants was 72.3 years (standard deviation 5.1), and 46.9% were male, and 90.9% of them received a follow-up assessment. Of the participants, 10.4% developed an IADL decline. Engagement in leisure physical activity was associated with a significantly lower risk of IADL decline (adjusted risk ratio, 0.73; 95% confidence interval [CI], 0.60 to 0.89), and cultural leisure activity was also associated with lower risk of IADL decline (adjusted risk ratio, 0.77; 95% CI, 0.63 to 0.95) independent of potential confounders. We also found significant and positive interaction between physical and cultural leisure activities at risk for IADL decline (P = 0.024) and SR decline (P = 0.004). CONCLUSIONS We found an independent association of physical and cultural leisure activities with a lower risk for functional decline in IADL and SR with positive interaction. Combined engagement in physical and cultural activities may effectively prevent from IADL decline and SR decline.