Background: This study aimed to develop a machine learning (ML)–derived nutrient risk score (NRS) and examine the association of genetic susceptibility and dietary intake with colorectal cancer (CRC) risk in Korean adults.

Methods: This hospital-based case-control study included 1,420 cases and 2,840 controls. Polygenic risk scores (PRS) were used to examine genetic susceptibility. The NRS was developed by combining five algorithms, nested cross-validation (CV), and recursive feature elimination with CV. Model performance was primarily assessed using AUC. Logistic regression was used to derive the NRS and assess the association of PRS, NRS, and their combined effect on CRC risk.

Results: A total of 15 nutrients were retained to construct the NRS. In mutually adjusted analyses, both scores remained independently associated with CRC risk (IV-weighted PRS: OR = 1.50; 95% CI, 1.10-2.06; NRS: OR = 3.67; 95% CI, 2.64-5.15 for the highest versus lowest tertile). NRS showed robust discriminative performance (AUC = 0.852; 95% CI, 0.833-0.872). While no significant PRS-NRS interaction was observed, the model integrating both scores significantly enhanced individual-level risk reclassification (IDI = 0.005, Continuous NRI = 0.120, P<0.005).

Conclusion: Both PRS and NRS showed independent associations with CRC risk, and their integration significantly improved risk reclassification. NRS may serve as a useful composite metabolic marker for CRC risk stratification. Future studies incorporating longitudinal designs, larger sample sizes, and multi-omic data are warranted to further elucidate CRC risk determinants and to inform personalized risk assessment strategies.

Background

Depression has been associated with various chronic conditions, including metabolic syndrome (MetS), but findings from Japanese studies have been inconsistent and are limited to cross-sectional design. We therefore examined the longitudinal association between depressive symptoms and incidence of MetS and its components in middle-aged Japanese workers.

Methods

Data were analysed from 2969 public servants (aged 35–65, 73.1% male) in the Aichi Workers’ Cohort Study who were free of MetS at baseline (2013), with at least one follow-up assessment (between 2014 and 2018). Depressive symptoms were assessed at baseline using the 11-item Centre for Epidemiological Studies Depression scale (CES-D-11) and categorized into tertiles: first tertile (0–4), second tertile (5–7), and third tertile (≥8). MetS was defined by the harmonized criteria for Japanese, based on waist circumference (≥90 cm for men and ≥80 cm for women), triglycerides, HDL cholesterol, blood pressure, and blood glucose levels. Multivariable logistic regression estimated ORs and 95% CIs. Restricted cubic splines and piecewise logistic regression evaluated non-linearity and a threshold.

Results

During five years, 391 participants (13.2%) developed MetS, with a cumulative incidence rate of 10.2% in the first tertile, 15.8% in the second tertile, and 14.9% in the third tertile. In the fully adjusted model, participants in the second and third tertile had higher odds of incident MetS (OR: 1.89, 95% CI, 1.44–2.48; OR: 1.92, 95% CI, 1.40–2.64, respectively) and high triglycerides (OR: 1.43, 95% CI, 1.08–1.90; OR: 1.80, 95% CI, 1.31–2.48, respectively) compared with those in the first tertile. Restricted cubic spline suggested non-linearity (p < 0.001) with a threshold at CES-D-11 approximately 10. Below this threshold, each 1-SD increase in CES-D-11 was associated with higher odds of incident MetS (OR 1.59, 95% CI, 1.31–1.92).

Conclusion

Depressive symptoms were associated with incident MetS among middle-aged Japanese workers, particularly with high triglyceride component. The association appeared non-linear, with risk increasing up to a CES-D-11 score of approximately 10 and then plateauing. This potential threshold may help improve understanding of the relationship between depressive symptoms and MetS risk, although further studies are needed before it can be used for risk stratification and prevention efforts.

Background: Solitary death (kodokushi) has emerged as a public health and social issue in Japan with rapid population aging. However, recent epidemiological data on solitary deaths remain limited.

Methods: We conducted a population-based descriptive epidemiological study using records from the Tokyo Medical Examiner’s Office from 2016 to 2023. All cases that met the study definition were selected from 110,024 records and included in the analysis. Data on demographic characteristics, postmortem interval (PMI), manner and cause of death, and underlying diseases were analyzed.

Results: Among 46,313 solitary death cases, 32,345 were male (69.8%, median age 70 years) and 13,968 were female (30.2%, median age 80 years). Crude mortality rates per 100,000 persons increased from 69.3 to 102.5 in males and 29.7 to 41.4 in females between 2016 and 2023. Solitary deaths exhibited seasonal variation, with peaks in summer and winter. Most cases involved pensioners and welfare recipients. PMI was longer in males than females (5 days vs. 3 days). Cardiovascular diseases were the leading cause of death for both sexes. Heat stroke predominated in summer, whereas natural deaths were more frequent in winter. More than half of cases had underlying diseases, most commonly hypertension.

Conclusions: Solitary death in Tokyo’s 23 wards increased over the study period. Seasonal, demographic, and clinical patterns highlight the need for season-specific prevention strategies and the potential role of primary care in mitigating risk.

Background: Limited information has been shown quantitatively assessed factors associated with the risk of secondary transmission in measles index cases.

Methods: We collected data on measles index cases reported in Osaka Prefecture between November 2018 and March 2019, along with their close contacts. Secondary attack rates (SARs) in contacts of index cases were calculated and compared depending on the vaccination status, classical or modified measles and other risk factors in index cases. For the multivariate analyses, we used binary logistic regression to adjust for these potential confounding factors on the secondary transmission and calculated the adjusted odds ratio (aOR) for the SARs.

Results: A total of 105 index cases were included, involving 9,846 close contacts, with 72 secondary cases identified. A significant negative association between the number of vaccinations received by index cases and the risk of secondary infection in his/her close contacts was observed (P < 0.01). Index cases who had received one vaccine dose had an aOR of 0.22 (95% confidence interval [CI], 0.08–0.65), while those with two or more doses had an aOR of 0.25 (95% CI, 0.07–0.85). Index cases who did not exhibit all three classic measles symptoms had a significantly lower risk of causing secondary infections compared to those who presented with all three symptoms (aOR 0.12; 95% CI, 0.03–0.39).

Conclusion: The indirect protective effect of measles vaccination would present important evidence supporting the further strengthening of vaccination policies.

Background: The Physical Effort Scale (PES) assesses tendencies to approach or avoid physical effort and provides a framework for examining how automatic effort-avoidance tendencies influence physical activity (PA) behavior. To facilitate cross-cultural PA research, this study developed and validated a Japanese version of the PES (PES-JP) following cross-cultural adaptation guidelines.

Methods: An online survey was conducted among 398 Japanese aged 20–59 years.

Results: Exploratory analyses suggested a two-factor structure of the PES-JP (approach and avoidance), and both subscales demonstrated high internal consistency (α = 0.92 and 0.87, respectively). Concurrent validity was supported by associations with PA. The approach score was positively related to moderate-to-vigorous PA (MVPA; β = 0.19) and negatively to sitting time (β = −0.17). In contrast, the avoidance score showed the opposite pattern (β = −0.13 for MVPA, β = 0.20 for sitting). The global score (approach minus avoidance), which reflects the overall tendency to approach rather than avoid physical effort, showed a similar pattern to the approach dimension (β = 0.19 for MVPA, β = −0.22 for sitting). Test-retest reliability was acceptable (intraclass correlations, 0.65–0.81).

Conclusion: The PES-JP demonstrates sound psychometric properties for assessing tendencies toward physical effort in Japanese adults and provides a useful tool for research on individual differences in PA behavior. The scale may also facilitate cross-cultural investigations of how tendencies toward physical effort influence PA across diverse populations.

Background: The Area Deprivation Index (ADI) is widely used to monitor health inequalities and often used as a proxy for individual socioeconomic status (SES). However, the extent to which it reflects individual SES remains unclear. This study aimed to establish evidence linking neighborhood-level ADI with individual SES indicators in Japan and examine whether similar patterns of association are observed across independent population-based cohorts.

Methods: We analyzed baseline data from two population-based cohorts: the Japan Public Health Center-based Prospective Study for the Next Generation (JPHC-NEXT) (baseline survey, 2011–2016; ages 40–74 years; N=100,024) and the Yokohama Health Study (YHS) (baseline survey, 2023; ages 30–69 years; N=8,860). ADI was ranked into quartiles (Q1: lowest, Q4: highest deprivation) based on the distribution across all postcodes in Japan and linked to each participant by residential area. Logistic regression models estimated associations between ADI quartile and low income (<1.53 million JPY) or limited education (≤9 or ≤12 years) by gender.

Results: Higher ADI showed increasing trends of association with low income and ≤9 years of education in both cohorts and across genders, with p for trend <0.001 for all. In contrast, associations with ≤12 years of education were less consistent across cohorts and subgroups, with no significant trend observed among women in JPHC-NEXT (p for trend = 0.74).

Conclusions: While the ADI consistently reflected income-related disadvantage across cohorts, its association with educational attainment varied across contexts. This should be carefully considered when using or interpreting ADI-based measures of health inequality in Japan.

Background: Physical activity, sedentary behaviour, and sleep (collectively referred to as physical behaviours) are promising preventive factors for adverse health outcomes. However, most previous studies have relied on subjective measures assessed at a single time point, which fail to capture changes in physical behaviours over time and can introduce measurement bias.

Methods: The Tsuru Longitudinal Study is a cohort study of older adults aged ≥65 years living in Tsuru, Yamanashi, Japan. The study includes mail surveys conducted in 2016, 2018, 2019, 2022, 2024, and 2026, accelerometer-based assessments in 2024 and 2026, and linked data (Long-Term Care Insurance and Basic Resident Registration System), which enable the examination of associations of trajectories of physical behaviours and device-measured physical behaviours with various health outcomes.

Results: While the 2026 survey is currently underway, the number of respondents (response rate) to the surveys in 2018, 2019, 2022, and 2024 was 4857 (72.4%), 4941 (74.1%), 5459 (74.6%), and 4106 (63.6%), respectively. A total of 779 participants completed the accelerometer-based assessment. At baseline of the mail survey, men were more likely than women to have lower educational attainment and less favourable health characteristics. Participants who completed the physical behaviour assessment were more likely to be older, have higher socioeconomic status, be more physically active, be more socially engaged, and be less likely to have depression.

Conclusions: This cohort study is expected to provide evidence on the health effects of physical behaviours and inform strategies to promote successful ageing.

Cancer Biobank Aichi (CBBA) was established at Aichi Cancer Center in 2017 as a hospital-based prospective biobank embedded in first-visit outpatient care. With written informed consent, up to 13 mL of blood is collected, processed within 2 hours, fractionated into serum, plasma, and buffy coat, dispensed into 2D-barcoded aliquots, and stored at −80°C under continuous monitoring with distributed freezer storage. DNA is extracted from buffy coat and stored as ready-to-use aliquots. Epidemiological information is obtained using a structured questionnaire; most lifestyle exposures (physical activity, alcohol intake, smoking, and diet) refer to the year preceding the index visit. The questionnaire includes outpatient dietary assessment and separate ascertainment of electronic/heat-not-burn tobacco product use. Clinical information is updated annually through linkage with the institutional hospital-based cancer registry based on standardized registration items recommended by the National Cancer Center Japan; for approved projects, linkage to department-level clinical data is available under governance review with data minimization and pseudonymization. Among 34,574 first-visit outpatients approached or informed about CBBA, 29,470 (85.2%) provided biospecimens and/or questionnaire data and were included, including 10,075 non-cancer participants. Biospecimens were available for 25,105 participants and questionnaire data for 26,570; 22,205 provided both components. CBBA enables multiple epidemiologic and translational study designs using biospecimens collected at first presentation, one-year-referenced lifestyle data, an internal non-cancer reference group, and registry follow-up.

Background:

Few studies have specifically examined the relationship between social media use and uptake of COVID-19 booster. This study examined the association between social media use and COVID-19 booster vaccination uptake in Japan during the pandemic.

Methods:

A series of cross-sectional surveys was conducted using rapid online questionnaires among residents of Iwate Prefecture, Japan, from October 1 to 3, 2021, and from December 11 to 17, 2023 (n = 4,026). Vaccination status was assessed using a single question: “What is your vaccination status?”. Respondents were categorized into two groups: individuals with incomplete booster uptake (those who had received one or two doses; n = 226) and booster acceptors (those who had received at least three doses; n = 3,800). Social media use was defined as self-reported use of Twitter, Facebook, Instagram, YouTube, or other social networking services. Logistic regression analyses were conducted to examine the association between social media use and receipt of a COVID-19 booster vaccination.

Results:

Social media use was significantly associated with higher odds of incomplete booster uptake of COVID-19 vaccination. This association remained statistically significant after adjustment for relevant covariates (odds ratio [OR] 1.70, 95% confidence interval [CI] 1.21–2.39). Individuals with a history of COVID-19 infection prior to the initiation of booster vaccinations were less likely to receive a booster dose in an exploratory analysis (OR 3.06, 95% CI 1.28–7.32).

Conclusion:

Differential associations between media use and COVID-19 booster uptake may provide important insights for designing effective communication strategies to improve vaccine acceptance in future pandemics.

Noise-induced hearing loss (NIHL) is the second most prevalent occupational disease in China. Due to its insidious onset, patients are often diagnosed at an irreversible stage, yet high-quality prospective cohorts addressing this critical issue remain limited. We established a multicenter prospective cohort—the Study of Health Effects from Noise Generated in China (SHENG)—with the primary objectives of systematically characterizing the epidemiological features, identifying environmental and genetic risk factors, elucidating molecular mechanisms, and investigating systemic health effects of NIHL, ultimately providing a scientific foundation for evidence-based prevention and control strategies. The baseline survey of SHENG was completed in 2023, covering eight enterprises across seven provinces and enrolling 4,660 young male workers (mean age 23.5 years). Based on high-frequency pure-tone average (3-8 kHz), 27.4% of participants were classified as having hearing loss, though most of them did not report subjective hearing difficulties. Compared with the normal hearing group, the hearing loss group had significantly more noise exposure events, higher exposure rates to occupational co-hazards such as dust and vibration, and poorer hearing protection practices. SHENG is planned to conduct follow-up assessments every 1-2 years. Through longitudinal follow-up and multidimensional data integration, this cohort will establish an “environmental-clinical-genetic” risk assessment model, elucidate the dynamic progression of NIHL and its associations with systemic diseases, and provide key evidence for refining occupational noise exposure standards and developing precision prevention strategies tailored to susceptible populations.

Background: The effects of hypnotics use on the risk of adverse pregnancy outcomes across pregnancy trimesters are underexplored.

Methods: From the data of the Japan Environment and Children’s Study (JECS), we constructed cohorts of women registered between January 2011 and March 2014 who delivered a liveborn singleton child. Exposure to hypnotics during the first (Period 1) and second/third (Period 2) trimesters was assessed in interviews. We identified the outcomes, preterm birth, low birth weight, and major congenital malformations, based on medical record transcripts. For each exposure period and pregnancy outcome, we estimated the risk differences (RDs) and risk ratios (RRs) using modified least-squares and modified Poisson models, respectively, adjusting for the mother’s age, body mass index, lifestyle factors, comorbidities, and other medication use.

Results: A total of 91,370 and 80,089 women were included for Periods 1 and 2, respectively. For Period 2, an increased risk of preterm birth (RD: 0.0748, 95% confidence interval [CI] 0.0175—0.132; RR 2.57 95%CI 1.42—4.65) and low birth weight (RD 0.0661 95%CI 0.00204—0.130; RR 1.77 95%CI 1.10—2.85) were observed among exposed compared with unexposed individuals. These risks were not observed for Period 1. The risk of major congenital malformations was not increased for both Periods 1 and 2.

Conclusion: Hypnotic medication in late-stage pregnancy may increase the risks of preterm birth and low birth weight, although caution is warranted because of possible confounding by indication. The elevation of the risk of the usage of hypnotics at any stage of pregnancy on major malformations was not observed.

Background: Socioeconomic disparities influence diet quality, but it remains unclear how income-related differences in nutritional adequacy vary by household composition, particularly the presence of children. We examined income-related differences in nutritional adequacy among Japanese adults.

Methods: We analyzed data from 7,265 adults aged 20–79 years from the 2014 and 2018 National Health and Nutrition Surveys. Nutritional adequacy was assessed using a composite score based on dietary reference intakes. Household income was categorized as <2 million JPY, 2–<6 million JPY, or ≥6 million JPY. Linear mixed-effects models were used to estimate sex-specific associations, adjusting for sociodemographic, lifestyle, and health factors.

Results: Lower income was associated with lower nutritional adequacy (p for trend < 0.001), and these associations persisted after adjustment. Income-related gradients were observed in both households with and without children and were steeper among households with children, with the lowest nutritional adequacy scores observed among low-income households with children. In contrast, patterns were less apparent among older women and individuals living alone.

Conclusions: Income-related differences in nutritional adequacy were observed across Japanese adults and varied according to household composition. These findings suggest that household context, particularly the presence of children, may modify income-related patterns in dietary quality.

Background

Previous studies on foreign body airway obstruction (FBAO) have primarily focused on patients who received prehospital or emergency department care. The epidemiological features of FBAO in the broader population, including cases managed outside emergency departments, remain unknown. We aimed to determine the incidence and characteristics of clinically managed FBAO using Japanese medical claims data.

Methods

This retrospective observational study was conducted using claims data from public health insurance enrollees across 16 municipalities. FBAO cases were identified from April 2020 to March 2024. The age-stratified and age-adjusted incidence rates of FBAO were calculated using the number of publicly insured individuals in the participating municipalities as the denominator.

Results

During the four-year observation period, a total of 1,123 FBAO cases (median age: 84 years) were identified. There were 518 (46.1%) cases in cardiac arrest on arrival and 516 (45.9%) deaths in the same month as the FBAO event. Age-stratified analyses showed that FBAO incidence rates steadily increased with advancing age, with individuals aged ≥95 years exhibiting an approximately 17-fold higher incidence than those aged 65–69 years. The age-adjusted incidence rate was 6.68 per 100,000 person-years.

Conclusion

This large-scale claims-based observational study indicated that the risk of FBAO occurrence is disproportionately high among individuals of advanced age, underscoring the need for FBAO prevention and management strategies targeting highly aged populations.

Background: Global urbanization intensifies human contact with pet birds, increasing the risk of psittacosis. Using Shenzhen, China, as a case study, this research investigates knowledge and practices related to psittacosis prevention among professionals and pet bird owners, aiming to inform targeted urban zoonotic disease control strategies.

Methods: A cross-sectional survey was conducted from September 2023 to October 2024 in Shenzhen using a self-administered questionnaire. Participants included practitioners in bird/pigeon businesses and household bird keepers. Primary outcomes were knowledge and behavior scores.

Results: Among 429 participants, the median total knowledge and practice score was 47.0 (out of 80), with only 13.5% scoring above 60. Practitioners scored significantly higher than pet owners (P < 0.05). Among keepers, the lowest awareness rates were for sources of infection, clinical symptoms, and incubation period (13.2%–22.1%), and the lowest behavior formation rates were for using protective measures, waste disposal, and responding to abnormal symptoms in pigeon and bird (10.9%–15.5%). Multivariate analysis identified key factors influencing scores: among practitioners, being male, having less experience, and lacking health education were associated with lower scores (P < 0.05). Among keepers, being female, higher education, purchasing from breeding bases, keeping multiple species, and prior psittacosis diagnosis predicted higher scores, while older age (> 50) and online bird purchases predicted lower scores (P < 0.05).

Conclusion: The overall psittacosis prevention capacity among bird-keeping populations in Shenzhen is moderate. There is an urgent need for stratified, context-specific interventions including standardized guidance and systematic health education to improve prevention effectiveness in urban settings.

Background

Influenza vaccination may reduce the risk of infection-triggered ischemic stroke (IS), but evidence remains limited, particularly in Japan.

Methods

This nested case-control study examined the association between influenza vaccination and the occurrence of IS among persons aged ≥65 years in a Japanese municipality between October 2016 and September 2023. IS cases were identified from insurance claims and matched with controls based on sex, age, and month of IS occurrence. Conditional logistic regression was used to estimate the association between vaccination and IS occurrence while adjusting for covariates such as comorbidities and prescription drugs.

Results

A total of 21,476 IS cases and 107,360 matched controls were analyzed. Influenza vaccination was significantly associated with a lower odds of IS (reference: unvaccinated) when administered 0–14 days (odds ratio: 0.74, 95% confidence interval: 0.62–0.88), 15–30 days (0.51, 0.43–0.59), 31–60 days (0.50, 0.45–0.56), 61–90 days (0.79, 0.72–0.87), and 91–120 days (0.81, 0.73–0.89) before IS occurrence. In a subgroup analysis, this association was observed only during the pre-epidemic and epidemic phases. Sex- and age-stratified analyses revealed similar patterns, with some variation in the magnitude of association.

Conclusions

These findings suggest a possible short-term inverse association between influenza vaccination and IS risk in older adults, but the influence of residual confounding cannot be excluded. Further research is needed to confirm these findings and clarify the potential role of influenza vaccination in reducing IS risk.

Background: Metabolic syndrome (MetS) is a major public health concern, yet few studies have jointly applied complementary analytic frameworks to assess overall dietary structure. This study aimed to derive a dietary pattern associated with MetS by combining food-based network analysis and reduced rank regression (RRR).

Methods: Participants from the China Nutrition and Health Surveillance (2010-2022) were included. Dietary intake was assessed using three-day 24-hour recalls and household condiment weighing. Food network analysis visualized and compared dietary structures of adults with and without MetS. RRR derived a dietary pattern based on six nutrient response variables (magnesium, zinc, calcium, potassium, insoluble fiber, riboflavin) selected for their biological relevance to MetS. Logistic regression evaluated associations between dietary patterns and MetS.

Results: Among 3,884 participants (median age 55.5 years), the non-MetS group showed a denser food network than the MetS group. The first RRR-derived pattern (DP1), explaining 40.15% of the variance in nutrient responses, was characterized by higher intakes of vegetables, fungi and algae, soybean products, dairy and beef/lamb, and lower intakes of alcohol, refined grains, and edible oils. DP1 was inversely associated with MetS (OR = 0.93, 95% CI: 0.87-0.99). A significant inverse trend across DP1 quintiles was observed, with the highest quintile showing the strongest protective association (OR = 0.66, 95% CI: 0.50-0.88).

Conclusions: A protective dietary pattern for MetS was identified through RRR and contextualized using food network analysis. This nutrient-dense pattern may help prevent MetS, and the integrated approach offers broader structural insight for nutritional epidemiology.

Objectives

Adolescents remain underprioritized in tuberculosis (TB) control, with limited evidence guiding screening and preventive strategies. Leveraging a mass school-based screening program in eastern China, we evaluated the prognostic value of tuberculin skin test (TST) induration size for predicting progression to active TB disease among adolescents within a prospective cohort framework.

Methods:

From 2018 to 2021, TST-based screening was conducted among newly enrolled students across 68 high schools in Jiangsu Province, China. All students were screened at baseline and followed through record linkage. Binary logistic and Cox proportional hazards regression models identified factors associated with Mycobacterium tuberculosis (Mtb) infection and disease progression.

Results

Among 198,276 students, TST positivity at 5 mm, 10 mm, and 15 mm cutoffs was 11%, 6%, and 2% respectively. Adjusted hazard ratios (95% CI) for indurations of 5–9 mm, 10–14 mm, and ≥15 mm versus <5 mm were 1.27 (0.46–3.50), 6.65 (3.85–11.49), and 10.02 (5.23–19.20). However, sensitivity to detect incident TB at each cutoff remained low (<40%). Compared with students recruited in 2018–2019, TST positivity was lower among those recruited in 2020 (aRR = 0.73; 95% CI, 0.59–0.89) and 2021 (aRR = 0.68; 95% CI, 0.53–0.86), as was TB progression among students recruited in 2021 (aHR = 0.35; 95% CI, 0.16–0.76).

Conclusion

In this large school-based prospective cohort, larger TST indurations were associated with subsequent TB, but overall sensitivity and positive predictive value of TST screening were low. School-entry TST screening alone cannot identify most at-risk adolescents, and more targeted strategies are needed to detect and protect them.

Background: In late 2015, an automatic influenza alert system was developed in Goto City, Nagasaki Prefecture, Japan using drug-dispensing data to enable earlier provider response to increased case counts. However, changes in incidence after implementation of this system remain unassessed.

Methods: An analytical observational study was conducted on influenza surveillance data during 2007–2019. The outcome was the annual cumulative influenza incidence per sentinel site in Nagasaki Prefecture before and after alert system implementation in Goto City in the 2015–2016 season. To evaluate the effect of the system in Goto City, a Poisson generalized linear mixed model was applied. The model included variables for location (Goto City versus other districts in Nagasaki Prefecture), timing (before or after the implementation), and an interaction term between location and timing.

Results: Before 2015, the median annual cumulative incidence was 149 cases (interquartile range: 61–320) in Goto City and 284 (162–474) in other districts. Following implementation, annual incidence was 163 (102–274) in Goto and 302 (197–503) elsewhere. The effect of timing variable (before versus after implementation) on annual incidence significantly differed for Goto City sites versus other sites in Nagasaki Prefecture (p for interaction < 0.001).

Conclusions: Although annual influenza incidence trended upwards, the increase in Goto City was significantly less than in other districts of Nagasaki Prefecture following alert system implementation. This study suggests the potential effect of a real-time surveillance strategy for settings with comprehensive, computerized dispensing data.

Background

There is limited evidence on the relationship between educational attainment and health markers in Japan. This study examined the associations of education level with cardiometabolic markers and health behaviors in a large Japanese population.

Methods

This cross-sectional study used health-related data (medical claims and health checkup results) from public health insurance enrollees linked with national census data between April 2020 and March 2021 from 16 municipalities. Participants were categorized according to education level: low (junior high school graduates or lower), middle (high school graduates), and high (vocational school graduates or higher). The study outcomes were cardiometabolic markers (waist circumference, body mass index, systolic blood pressure, diastolic blood pressure, HbA1c, triglycerides, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol) and health behaviors (smoking habit, drinking habit, restful sleep, and regular exercise). Linear regression and modified Poisson regression analyses were performed to analyze the associations between education levels and the outcomes. The analyses were also conducted according to sex and age (39-64 and 65-75 years).

Results

The study cohort comprised 61,684 participants. Higher education levels were generally associated with better cardiometabolic markers, but were also linked with higher low-density lipoprotein cholesterol levels. In addition, higher education levels were associated with better health behaviors. The magnitude of association was more pronounced in women and middle-aged persons.

Conclusions

These findings provide a broad overview of the associations between educational attainment and health markers in a large Japanese population, and deepen our understanding of health disparities.

Background: Experimental studies have linked elevated serum urate (SUA) to kidney injury; however, Mendelian randomization (MR) studies and randomized trials have not consistently supported a causal effect on kidney outcomes. Japanese populations are enriched for the URAT1 loss-of-function variant rs121907892, allowing evaluation of a biologically specific question related to tubular urate transport. We examined the association of genetically predicted SUA with kidney function and, separately, the association of rs121907892 with kidney function measures in Japanese cohorts.

Methods: We performed a two-sample MR analysis using 34 genome-wide significant SUA-associated variants from BioBank Japan. Genetic associations with SUA were obtained from the Japan Multi-Institutional Collaborative Cohort Study (n=10,428), and associations with kidney function were obtained from the Tohoku Medical Megabank Organization cohort (n=47,066). Kidney function was assessed using creatinine- and cystatin C-based estimated glomerular filtration rate (eGFR). We also evaluated the association of rs121907892 with these measures.

Results: Polygenic MR showed no evidence of an association between genetically predicted SUA and kidney function, with null findings for creatinine-based eGFR (inverse-variance weighted β=−0.031; 95% confidence interval [CI], −0.138 to 0.075; P=0.56) and cystatin C-based eGFR (β=−0.020; 95% CI, −0.123 to 0.084; P=0.71). In contrast, the rs121907892 A allele was associated with higher creatinine-based eGFR (β=0.047; 95% CI, 0.005 to 0.089; P=0.03), but not with cystatin C-based eGFR (β=−0.002; 95% CI, −0.044 to 0.040; P=0.97).

Conclusions: Polygenic MR provided no evidence that genetically predicted SUA has a major causal effect on kidney function in Japanese cohorts. The rs121907892 findings indicate a biomarker-dependent, variant-specific association.

Background: Incentive-based strategies are widely used to promote physical activity. However, behavioural responses around predefined targets and their long-term persistence remain insufficiently understood. The aim of this study was to examine changes in daily step count distributions following the introduction of a non-financial incentive and their persistence over long-term application use.

Methods: This observational study used large-scale data from users of Pa-League Walk, a baseball fandom-based gamified smartphone application in Japan. Daily step-count data from March 2016 to December 2022 were analysed. A non-financial incentive in the form of a digital player photo reward for achieving 10,000 steps per day was introduced in October 2016. Changes in the proportion of days achieving the 10,000-step goal and daily step count distributions were examined using bunching estimation. Analyses were conducted according to application use duration (<1 year to ≥6 years).

Results: Among 49,008 users with 11,129,438 daily observations, excess mass at the 10,000-step threshold was not significant prior to the introduction of the incentive. However, a clear accumulation of observations around the threshold emerged after the incentive was introduced. Additionally, the proportion of days achieving the 10,000-step goal increased. These distributional features persisted across application usage durations and were sustained for more than six years among long-term users.

Conclusions: In a gamified smartphone application, the introduction of a non-financial, repeated incentive was associated with sustained behavioural adjustments toward a predefined step-count target. These findings suggest that well-designed non-financial incentives may support long-term physical activity behaviours in a subset of users.

Background: We evaluated the reliability of large language models (LLMs) for abstract screening under real-world review practices and qualitatively characterized model–human discordance to inform safe workflow integration.

Methods: We evaluated GPT-4.0, GPT-5.0, and GPT-5.0-mini on two curated systematic review datasets representing contrasting topic densities, defined by the target-to-background ratio (TBR): a core-subject dataset (TBR 69%) in which the target intervention was central, and a peripheral-subject dataset (TBR 4.4%) in which the target intervention was incidental. Final inclusion after full-text review served as the reference standard. We developed a qualitative taxonomy of disagreements, classifying false negatives as intended human leniency, gray-zone ambiguity, or true LLM misses, and false positives as implicit or additional human exclusion rules, gray-zone ambiguity, or nominal inclusions that increase workload only.

Results: GPT-5.0-mini achieved the best sensitivity–efficiency trade-off (core-subject: 91% sensitivity with 96.7% workload reduction; peripheral-subject: 83% sensitivity with 92.7% workload reduction) and negative predictive value >99% in both datasets. Disagreement was lower when relevance was central (core-subject: 1.6%, 7/430) with no true LLM misses (0/430). In the peripheral-subject dataset, disagreement was higher (10.6%, 74/696), driven mainly by intended human leniency among false negatives (52/56) and gray-zone ambiguity among false positives (12/18), while true LLM misses remained rare (0.4%, 3/696).

Conclusion: Many model–human disagreements reflect topic- and workflow-dependent screening conventions rather than intrinsic model failure. LLM-assisted screening may improve efficiency without compromising reliability when accompanied by appropriate safeguards for ambiguous records.

The Tokyo Adolescent Mental Health Study (TAMHS) is an ongoing, school-based longitudinal cohort designed to elucidate the developmental trajectories of mental health among adolescents in Japan and to inform evidence-based strategies for promoting adolescent well-being. Since its inception in 2009, TAMHS has annually tracked adolescents aged 12–18 from a Tokyo secondary school known for its tradition of enrolling twins and triplets. Participants are followed for up to 6 years. As of 2024, 2,435 adolescents (1,211 boys, 1,196 girls, 28 unspecified), including 551 twins and triplets (22%), have been enrolled. This unique sample composition allows for rigorous comparisons between singletons and multiples to examine the impact of birth plurality on neurodevelopmental and mental health trajectories. High annual retention (96.2–99.2%) ensures the integrity of longitudinal analyses. Data are collected through self-report questionnaires covering demographics, physical and mental health status, daily habits and lifestyle behaviors, social relationships, behavioral problems, healthcare access, and social and environmental determinants of health. The study has yielded important insights into associations between adolescent lifestyle behaviors and mental health outcomes.

Background: The coronavirus disease 2019 (COVID-19) pandemic disrupted access to healthcare for homebound older adults, particularly those with cancer. We compared changes in the initiation and outcomes of physician home-visit care among older Japanese patients with and without cancer during the pandemic.

Methods: We conducted a retrospective observational study using national medical claims data of patients aged ≥65 years who began physician home visits between April 2018 and December 2020 in Japan. The proportions of all-cause death, in-home death, and at least one hospitalization within 90 days after the initiation of home-visit care were compared between the pre-pandemic (April 2018 to December 2019) and pandemic periods (January 2020 to December 2020), stratified by cancer diagnosis. Differences between the cancer and non-cancer groups were estimated using linear regression models with an interaction term for cancer status and pandemic period.

Results: Among 509,718 patients (251,311 with cancer), the initiation of home-visit care increased in both groups during the pandemic. All-cause and in-home death rates increased and hospitalization rates decreased in both groups, with greater changes among patients with cancer: 1.6% (95% confidence interval [CI], 1.1–2.1) for all-cause death, 3.8% (95% CI, 3.4–4.3) for in-home death, and −2.9% (95% CI, −3.7 to −2.2) for hospitalization.

Conclusion: During the COVID-19 pandemic, physician home-visit care expanded, and patients with cancer showed a greater shift toward home-based end-of-life care. These findings highlight the need for targeted resource allocation strategies to support home-visit physicians caring for patients with cancer during public health emergencies.

Age standardization is a key statistical method used in health statistics to adjust rates, such as mortality or incidence, enabling comparisons across populations or time points with different age structures. This review traces its historical development, global and country-specific practices, and future directions. The method dates back to the 19th century, with major adoption in the 20th century through Segi and Doll’s World Standard Population. While the World Health Organization (WHO) introduced an updated standard in 2000, the International Agency for Research on Cancer (IARC) continues to use Segi and Doll’s standard in the Cancer Incidence in Five Continents series, prioritizing consistency and comparability in long-term cancer surveillance. Case studies from the IARC, the United States, Japan, and the Republic of Korea (Korea) illustrate different responses to changing demographics. The United States adopted the 2000 standard with expanded age detail for the elderly population. Japan introduced the 2015 Japan Standard Population to account for its rapidly aging society, though regional data limitations presented challenges. Korea, experiencing one of the fastest aging transitions globally, updated to a 2020 standard for more accurate national and sub-national reporting. The review also emphasizes that age standardization can obscure important age-specific trends. Methods like Joinpoint clustering help detect divergent trends by age groups. Looking forward, age standardization remains essential amid global demographic shifts. However, updates of standard populations must balance improved relevance with the need for continuity and robust data. International coordination and digital tools will support more flexible and transparent health statistics in the future.

Objectives: The rising prevalence and earlier onset of hair loss have highlighted its associations with cardiovascular risk factors (hypertension, diabetes, dyslipidemia) and comorbidities among adults. This cross-sectional study examined the associations of hair loss severity with hypertension, diabetes, dyslipidemia, and cardiometabolic multimorbidity (CMM).

Methods: Hair loss was assessed via Modified Hamilton-Norwood (males) and Savin (females) scales in the Tianning cohort (N=5020). Hypertension, diabetes, and dyslipidemia were diagnosed per Chinese guidelines, with two or more coexisting conditions defining CMM. Multivariate logistic regression analyzed associations between hair loss and these diseases, including CMM.

Results: Compared with participants without hair loss, those with mild and severe hair loss had 29% (OR=1.29, 95% CI, 1.08-1.54) and 48% (OR=1.48, 95% CI, 1.16-1.89) higher odds of hypertension, respectively. Severe hair loss was also associated with 44% higher odds of diabetes (OR=1.44, 95% CI, 1.11-1.86) and 30% higher odds of CMM (OR=1.30, 95% CI, 1.02-1.65). No significant association was observed for dyslipidemia after multivariable adjustment.

Conclusion: Greater hair loss severity was associated with higher odds of hypertension, diabetes, and CMM in this cross-sectional analysis. These findings suggest that hair loss severity may be relevant to cardiometabolic health in adults.

Background: Individuals receiving public assistance have been reported to be at increased suicide risk; however, evidence from emergency medical settings remains limited. This study investigated the population-level association between public assistance status and suicide-related cardiac arrest using events captured through emergency department presentations.

Methods: We conducted a nationwide registry-based study using the Japan Registry of Self-harm and Suicide Attempts, including patients presenting to emergency and critical care centers following self-harm or suicide attempts from December 2022 to December 2023. The primary outcome was suicide-related cardiac arrest, and non-fatal self-harm was examined as a related outcome. Adjusted incidence rate ratios (aIRRs) were estimated using Poisson regression models with the log of the population at risk included as an offset, adjusting for age and sex. National suicide statistics were analyzed for validation.

Results: Among 1,757 patients, 207 (11.8%) were receiving public assistance. Public assistance status was associated with a higher incidence of suicide-related cardiac arrest compared with non-recipients (aIRR 3.34, 95% confidence interval [CI], 1.60-6.95). A strong association was also observed with non-fatal self-harm (aIRR 13.10, 95% CI, 8.25-20.80). Similar associations were observed across sex and age groups. Analyses using national suicide statistics confirmed a consistent association (aIRR 2.21, 95% CI, 1.50-3.26).

Conclusions: In this nationwide registry-based study using events captured through emergency department presentations, public assistance status was associated with suicide-related cardiac arrest and non-fatal self-harm. At the population level, public assistance recipients showed a higher incidence of suicide-related cardiac arrest, supporting the public health relevance of this socioeconomic indicator.

Background: Dietary diversity is associated with enhanced cognitive function and reduced dementia risk among older adults; however, its relationship with structural brain ageing remains unclear. We examined cross-sectional and longitudinal associations between dietary variety score (DVS) and brain volume among older Japanese adults.

Methods: We analysed data from the Neuron to Environmental Impact across Generations (NEIGE) cohort, a population-based longitudinal study of community-dwelling Japanese adults aged 65–84 years. The analysis included 279 participants with baseline and follow-up magnetic resonance imaging (MRI) data. Dietary diversity was assessed at baseline using the DVS based on the consumption frequency of 10 food groups. Hippocampal and total grey matter volumes were measured using MRI at baseline (2017) and follow-up (2021). Cross-sectional associations were examined using multivariable linear regression models, and longitudinal associations were assessed using linear mixed-effects models adjusting for confounders.

Results: Mean follow-up was 4.1-year (SD 0.2). In cross-sectional analyses, DVS was not significantly associated with brain volume. In longitudinal mixed-effects models, the interaction between baseline DVS and duration was not significantly associated with changes in total grey matter volume (−128.11 mm³/year per 1-point increase in DVS; 95% CI, −267.86 to 11.65; p = 0.07). Similarly, no significant association was observed for hippocampal volume (-0.44 mm³/year per 1-point DVS; 95%CI, -3.52 to 2.65; p=0.78).

Conclusions: Dietary diversity was not significantly associated with brain volume or with the rate of change in brain volume over time among older Japanese adults.

Background

Previous studies have reported educational inequalities in all and site-specific cancer mortality. However, the mechanisms underlying this association remain unclear. We investigated the association between educational level and all- and site-specific cancer mortality and examined potential mediating factors.

Methods

We conducted a longitudinal study using data from the Japan Gerontological Evaluation Study (JAGES), initiated in 2010 with 7-year follow-up targeting independent adults aged ≥65 years. Self-report questionnaire data were linked to mortality records from municipal registries. Educational level was exposure (<10/10–12/>12 years for competing risk models and <10/≥10 years for causal mediation analysis). All- and site-specific cancer mortality were outcomes. Competing risk models estimated sub-distribution hazard ratios (SHRs) and 95% confidence intervals (CIs). Causal mediation analysis assessed mediating effects of smoking status, drinking status, health examination, vegetable and/or fruit intake, and walking time.

Results

Among 43,478 participants (46.6% men, mean age 73.7 years), the 7-year mortality from all cancers was 5.1%. Low educational level was associated with higher all-cancer mortality (SHR for the lowest compared with the highest 1.27; 95%CI, 1.13–1.44). Similar associations were observed for tracheal and lung, and esophageal cancer mortality. The association between educational level and all cancer mortality was significantly mediated by smoking status (5.6%), health examination (8.0%), and walking time (7.1%), but not by drinking status and vegetable and/or fruit intake.

Conclusion

Lower educational level was associated with a higher risk of all cancer, tracheal and lung cancer, and esophageal cancer mortality. Health behaviors partially explained these associations.

Background: Frailty is a growing public health concern, with emerging evidence highlighting the importance of social determinants, including external and intra-household isolation. This study examined the association between intra-household isolation and frailty among community-dwelling older adults in Japan and explored whether external social connections modified this association.

Methods: Data were obtained from the 2023 baseline survey of the Wako Cohort Study, including 6,429 participants aged ≥65 years in Wako City, Saitama Prefecture, without long-term care certification levels 3–5 (men: 45.0%; mean age: 76.5 years). Intra-household isolation was defined as living with others but having minimal interaction (talking <15 min/day and spending most time alone at home). Frailty was assessed using the 25-item Kihon Checklist. Modified Poisson regression analysis estimated the associations, adjusting for sociodemographic factors, health behaviors, health conditions, and external social connections.

Results: Intra-household isolation was observed in 5.3% of participants, while 21.0% lived alone. Frailty was observed in 28.3% of participants. Individuals with intra-household isolation were more likely to be frail than non-isolated individuals (prevalence ratio [95% confidence interval]: 1.42 [1.26–1.59]). This association remained significant when compared with individuals living alone (1.34 [1.16–1.54]) and was stronger among those aged 65–74 years than those aged ≥75 years. No significant interaction was observed between intra-household isolation and external social connections.

Conclusion: Intra-household isolation among cohabiting older adults was independently associated and showed a stronger association with frailty than living alone. These findings suggest that living with others does not necessarily have protective effects unless meaningful interactions exist.

Background: Angiopoietin-like protein 2 (ANGPTL2) is an aging-associated protein that contributes to the maintenance of tissue homeostasis. Excessive activation of ANGPTL2 disrupts this homeostatic process, leading to several adverse conditions, such as chronic inflammation, abnormal tissue remodeling, insulin resistance, and carcinogenesis. Nevertheless, only a very few studies have investigated the relationship between circulating ANGPTL2 concentrations and cause-specific mortality risk in the general population.

Methods: We conducted a prospective cohort study involving 2,912 community-dwelling Japanese adults aged ≥40 years with no prior history of cardiovascular diseases or cancer. Baseline serum ANGPTL2 concentrations (collected in 2002–2003) were quantified, and the participants were followed up for a median of 15.2 years. Hazards ratios (HRs) and 95% confidence intervals (CIs) for all-cause and cause-specific mortality were estimated using Cox proportional hazards models.

Results: During the follow-up period, 601 participants died (139 from cardiovascular diseases, 200 from cancer, 142 from infectious diseases, and 120 from other causes). Higher serum ANGPTL2 levels were significantly associated with increased risks of all-cause and cause-specific mortality (HR [95% CI] per 1-SD increment in log-transformed serum ANGPTL2 concentration: 1.34 [1.23–1.47] for all-cause mortality; 1.28 [1.06–1.55] for cardiovascular death; 1.31 [1.12–1.53] for cancer death; and 1.27 [1.05–1.54] for death from infectious diseases) after adjustment for potential confounders. These associations were modestly attenuated after additional adjustment for mediators related to insulin resistance and chronic inflammation.

Conclusions: These findings indicate that elevated serum ANGPTL2 concentrations are associated with a higher risk of mortality from multiple causes.

Sleep and physical activity play a crucial role in brain and mental health. While traditional self-reported methods and research-grade accelerometers have several limitations, consumer-grade wearable devices, such as Fitbit, allow continuous, objective, and real-life data collection. The purpose of this article is to report the study protocol and participant characteristics of a wearable device survey that primarily aims to examine the relationship between brain health and long-term patterns of daily physical activity and sleep.

Nearly 2,000 participants were recruited between September 2022 and December 2023, as part of the Tohoku Medical Megabank Brain Magnetic Resonance Imaging Study, to wear Fitbit Charge 5 devices for 1 year, continuously tracking physical activity and sleep. Additionally, home blood pressure measurements and questionnaire data on housing conditions, psychological distress, and medication were collected every 4 months for a total of four assessments.

The mean age of the participants was 58.5 years and 37.4% were men. For the first 30 days after the recruitment, mean step count was 8,910 steps/day, and a mean total sleep time was 370.4 min/night (approximately 6.2 h). The mean morning systolic blood pressure and diastolic blood pressure were 126.2 and 75.5 mmHg, respectively.

This study provides a unique opportunity to integrate longitudinal consumer-grade wearable data, brain MRI, multiomics data, and comprehensive health records from existing cohorts to advance precision medicine and help prevent mental and neurodegenerative diseases, such as dementia. Limitations include potential selection bias, and a relatively smaller sample size than larger global studies.

In epidemiological research, time-to-event outcomes, commonly referred to as survival outcomes, are a common subject of investigation. Here, we first describe the circumstances under which methods of survival analysis are necessary, review the roles of hazard functions, and discuss the limitations of hazard ratios for causal inference. Second, we explain how confounding and dependent censoring, which are common in observational studies, can be addressed by inverse probability weighting and parametric g-formula estimators. We emphasize that hazards serve as building blocks for estimating counterfactual risks. Third, we summarize recent developments in defining causal estimands in the presence of competing risks, including risk without eliminating competing events, net risk, and cumulative incidence under modified treatment. To illustrate how these challenges are addressed in practice, we revisit a recent clinical study on pharmacological interventions for the onset of dementia. We further empirically compare various estimands in the presence of competing events, including separable effects, through simulations. Our overall aim is to elucidate the need to move beyond routinely used methods of survival analysis, particularly the mere estimation of hazard ratios, if the goal is to draw causal inferences. This paper provides an overview of causal survival analysis, focusing on how confounding, dependent censoring, and competing risks can be addressed to estimate causal parameters of interest (counterfactual survival functions or counterfactual risks), which are interpretable and often meaningful for investigators.

Background: The COVID-19 pandemic severely strained public health systems. In Japan, Public Health Centers faced administrative bottlenecks in patient monitoring and allocation. To address this, Sapporo City introduced COVIMARU, a web-based self-reporting application. This study quantitatively evaluated the impact of this digital tool on patient care site allocation efficiency.

Methods: We conducted a retrospective study using a regression discontinuity design (RDD) using data from Sapporo City's comprehensive COVID-19 database from May 1, 2020, to April 25, 2022. The intervention was COVIMARU implementation for home care on November 12, 2020. The primary outcome was time in days from diagnosis to first care site allocation. We quantified the estimated effect at the cutoff to evaluate the localized impact of the intervention.

Results: The analysis included 130,500 patients. COVIMARU introduction was associated with a significant reduction in mean allocation time at the intervention threshold (estimated effect: -0.96 days (95% CI, -1.39 to -0.20; P = 0.0091). This improvement was not attributable to changes in patient volume. A sensitivity analysis excluding non-working days confirmed the finding (estimated effect: -0.87 days; 95% CI, -1.39 to -0.21; P = 0.008).

Conclusion: The COVIMARU web application significantly reduced patient allocation times, demonstrating that digital health tools can effectively mitigate administrative bottlenecks and improve pandemic response efficiency. Investing in such digital infrastructure is a critical strategy for building resilient public health systems for future emergencies.

Background

North-south differences in context and lifestyle may shape multimorbidity patterns in China. We compared multimorbidity between two northern (Liaoning, Jilin) and two southern (Guangdong, Jiangsu) CNHS provinces in eastern China using association rule mining (ARM) and sequential pattern mining (SPM).

Methods

We analyzed 19,067 adults (≥20 years) and 20 chronic conditions. ARM identified co-occurrence rules and reported support (proportion with the combination), confidence (P(consequent|antecedent)), and lift (observed/expected co-occurrence; >1 indicates positive association). SPM summarized common diagnosis-order sequences using sequence support (percentage of participants exhibiting an ordered sequence). Sensitivity analyses were restricted to self-reported physician diagnoses.

Results

Multimorbidity prevalence was 37.85% (age-standardized 32.84%), higher in northern than southern provinces (41.39% vs. 34.08%; standardized 35.41% vs. 31.04%). Hypertension, hyperuricemia, and diabetes were most common. Cardiovascular/cerebrovascular conditions were more frequent in the north, whereas anemia and kidney stones were more common in the south. ARM suggested metabolic–renal clustering in the south (e.g., CKD→gout, support=1.18%, confidence=26.14%, lift=2.60; gout+anemia→CKD, support=0.55%, confidence=38.89%, lift=3.71) and musculoskeletal–cardiovascular links in the north (e.g., osteoarthritis→spinal disease, support=0.52%, confidence=24.69%, lift=2.22; CTRD→CHD, support=0.63%, confidence=28.24%, lift=1.26). Common SPM sequences included hypertension→hyperuricemia (24.22%) and hypertension→diabetes→hyperuricemia (5.32%); hypertension→hyperuricemia→CKD was more frequent in the south (2.28%), while hypertension→CHD→hyperuricemia was more frequent in the north (5.70%). Sensitivity analyses showed broadly consistent regional contrasts.

Conclusions

Multimorbidity burden and structure differed between the northern and southern provinces, supporting province-tailored prevention and integrated care.

Background: Early childhood offers a foundational window for developmental progress, yet many children in low- and middle-income countries (LMICs) face risks of developmental delay (DD) due to inadequate or harsh caregiving environments. This study investigates the associations between child discipline practices (CDP) and suspected DD and aggressive behavior in Lao People's Democratic Republic (Lao PDR).

Methods: We analyzed nationally representative data from the 2023 Lao Social Indicator Survey III, including 5,007 children aged 24–59 months. CDP was assessed using an 11-item module categorized into non-violent discipline, psychological, and physical punishment. Suspected DD was measured using the Early Childhood Development Index (ECDI2030) tool; aggressive behavior was caregiver-reported. Poisson and multinomial logistic regression models were employed to estimate prevalence ratios (PRs) and odds ratios (ORs), each with 95% confidence intervals (CIs).

Results: The prevalence of suspected DD was 56.3%. Non-violent discipline (adjusted PR: 0.82; 95% confidence interval [CI], 0.80–0.87) and psychological punishment (adjusted PR: 0.90; 95% CI, 0.85–0.95) were associated with lower prevalence of suspected DD, although an interaction analysis suggested the latter effect was evident only when co-occurring with non-violent discipline. Physical punishment was associated with increased prevalence of suspected DD (adjusted PR: 1.07; 95% CI, 1.01–1.15) and higher odds of elevated aggressive behavior (adjusted OR: 2.15; 95% CI, 1.15–4.05).

Conclusion: Non-violent and physical discipline were differentially associated with developmental outcomes. Promoting non-violent caregiving strategies may reduce behavioral risks and support early development in Lao PDR and similar settings.

Background: Physical activity (PA) levels are inversely associated with hip fracture risk, but whether this is the case for other types of fractures, especially vertebral fractures, is unclear. The present study aimed to examine whether PA levels are associated with osteoporotic fracture risk in community-dwelling people.

Methods: This cohort study analyzed 13,558 Japanese people aged 40–74 years (6,756 women and 6,802 men). Information on demographic characteristics, body size, lifestyle, and disease history was obtained by a self-administered questionnaire conducted in the 2011–2013 baseline survey. Levels of total PA excluding sleep, sitting, and other sedentary activities (TPA) were estimated by the metabolic equivalents score (MET-h/d). Incident limb fractures (distal radius, humeral neck, and hip) and vertebral fractures were identified in medical records. A Cox proportional hazards model was used to calculate hazard ratios (HRs).

Results: During a mean follow-up period of 12.1 years, 966 fractures occurred (725 in women and 241 in men). Higher TPA levels were associated with a lower hazard of limb fracture (P for trend=0.0244), with the highest quintile having a lower HR (HR=0.67; 95%CI, 0.49–0.90) than the reference (lowest quintile). Higher LTPA levels were associated with a lower hazard of limb fracture (P for trend=0.0077), with the highest level having a lower HR (HR=0.73; 95%CI, 0.58–0.92) than the reference. No PA measures were associated with vertebral fracture.

Conclusion: Higher levels of PA of any type are associated with a low risk of limb fracture, but not clinical vertebral fracture, in Japanese adults.

Background: Physical frailty is a well-established risk factor for cognitive impairment, but the psychosocial mechanisms underlying this link remain insufficiently understood. Psychological well-being (PWB) may represent a modifiable pathway. This study examined whether PWB mediates the association between physical frailty and incident cognitive impairment, and whether this pathway varies by age or gender.

Methods: We analyzed data from 11,607 cognitively normal adults aged ≥60 years in the Chinese Longitudinal Healthy Longevity Survey (CLHLS, 2008-2018). Frailty was assessed using a deficit-based Frailty Index, PWB using validated multi-item scales, and cognition using the Mini-Mental State Examination. Associations were examined using correlation, multivariable regression, and survival analyses. Causal mediation analysis under a counterfactual framework with 1,000 bootstrap resamples estimated direct and indirect effects, with age- and gender-stratified analyses assessing moderated mediation.

Results: Over a median follow-up of 40.3 months, 19.5% of participants developed cognitive impairment. Frailty was associated with a 40% higher hazard of cognitive impairment (HR=1.40, 95%CI: 1.15-1.71), and PWB mediated 6.5% of this association. The proportion mediated was greatest among adults aged 60-79 years (10.8%) and was not modified by gender.

Conclusions: PWB partially mediates the frailty-cognition association in Chinese older adults, with stronger mediation observed in younger age groups. Enhancing psychological well-being may offer a scalable and modifiable strategy to strengthen cognitive resilience in rapidly ageing societies.

Background: Evidence on factors influencing the association between physical activity (PA) and knee osteoarthritis (KOA) is scarce. The present study aimed to examine this association and explore interactions of PA with sex and age in relation to KOA in Japanese adults.

Methods: This 11-year cohort study included 10,876 community-dwelling individuals aged 40–74 years in Murakami City (Niigata Prefecture, Japan). Data were collected using a self-administered questionnaire (2011–2013). Exposures were self-reported PA levels, and the outcome was clinically diagnosed KOA. Covariates included demographic factors, body size, lifestyle habits, and disease history. Cox proportional hazards models were used to estimate hazard ratios (HRs).

Results: The mean age of participants was 58.0 years. Over a mean follow-up period of 11.1 years, 840 cases of KOA were observed. High total PA levels were associated with a greater KOA risk (P for trend <0.0001), with the highest quartile showing a higher hazard (HR=1.52, 95%CI:1.23–1.87) than the lowest (reference). This association was observed in both males (HR=2.01, 95%CI:1.39–2.93) and females (HR=1.30, 95%CI:1.01–1.66), with a P for interaction of 0.0219. In addition, the association was more pronounced in younger participants (<55 years; HR=2.07, 95%CI:1.25–3.43) and attenuated in older age groups, with a P for interaction of 0.0278. Non-leisure-time PA showed a similar association with KOA as total PA, whereas leisure-time PA did not.

Conclusion: High total and non-leisure-time PA levels were associated with increased KOA risk, and these associations interacted with sex and age in middle-aged Japanese adults.

Evidence-based policymaking increasingly demands integrating real-world data with patient perspectives. However, a methodological gap exists: administrative claims data provide comprehensive coverage but lack subjective insights, whereas patient experience surveys frequently experience selection bias owing to the lack of reliable sampling frames. This study introduces Claims-Based Enumeration Sampling (CBES), a methodology utilizing administrative claims data as a sampling frame to substitute for clinical registries.

By linking survey responses with claims data at the individual level, CBES facilitates the calculation of sampling probabilities and the application of weighting for ensuring representativeness. We demonstrate the practical feasibility of this design through a case study conducted in Tsukuba City, Japan. Targeting National Health Insurance beneficiaries with diabetes, the survey was implemented as an insurer-commissioned administrative operation, successfully overcoming privacy- and data access-related legal hurdles. The resulting dataset revealed crucial insights into patient stigma and socioeconomic disparities, factors invisible to claims analysis alone.

This study discusses the theoretical framework, statistical advantages, and legal solutions related to CBES. We conclude that CBES provides a robust and scalable alternative to conventional methods, empowering policymakers to capture the “silent voices” of patients and advancing patient-centered healthcare policy.

Background: Kawasaki disease (KD) is a vasculitis syndrome of unknown etiology and the leading cause of acquired heart disease in children in developed countries. The Japanese Nationwide Survey of Kawasaki Disease (JNSKD) in Japan has provided valuable epidemiological data since 1970, but has inherent limitations. We aimed to evaluate the utility of the National Database of Health Insurance Claims (NDB) for describing the epidemiologic features of KD.

Methods: We conducted retrospective cohort study compared NDB data (2013-2019) against JNSKD data for the same period. We analyzed demographic characteristics, temporal and geographic distributions, treatment modalities, and cardiac complications.

Results: NDB identified 120,391 KD patients (2013-2019), with JNSKD estimated to cover 94.0% of cases. Demographic characteristics showed strong concordance between datasets (male-to-female ratios: 1.32 vs 1.33; similar age distributions). Initial IVIG administration rates were highly similar (93.74% vs 93.85%). Incomplete KD was significantly underestimated in NDB (8.07% vs 19.48%). NDB revealed adjunctive therapies including urinastatin and tracked increasing infliximab use. Cardiac complication detection in NDB showed limited sensitivity for mild abnormalities (overall 5.0% vs 7.33%). NDB showed higher recorded rates of giant aneurysms (0.53% vs 0.17%) and myocardial infarction (0.15% vs 0.01%).

Conclusions: The NDB provides a valid, complementary data source for KD epidemiological research with near-complete population coverage. While limitations exist for identifying disease subtypes and cardiac complications, its detailed treatment data offer significant advantages for nationwide surveillance and health policy development.

Background: As the global population ages, understanding factors influencing structural brain aging is crucial for dementia prevention. Social networks have emerged as important social factors; however, longitudinal evidence linking specific structural characteristics of social networks to brain volume change remains limited. This study examined whether social network homogeneity and heterogeneity were related to longitudinal changes in brain volume (hippocampus, gray matter, and prefrontal cortex) among community-dwelling older Japanese adults.

Methods: Data were obtained from the Neuron to Environmental Impact Across Generations (NEIGE) study. Participants aged 65–84 years were randomly selected from Tokamachi City, Japan. Of the 527 older adults surveyed at baseline, 279 who underwent magnetic resonance imaging at both baseline (2017) and follow-up (2021) were included. Social relationship (dis)similarities were assessed using ten items, and network homogeneity and heterogeneity domains were derived through factor analysis. Covariates included sociodemographic factors, health behaviors, health conditions, and apolipoprotein E genotype.

Results: Participants were 47.6% male, with a mean age of 73.2 years (SD = 5.1). Multiple linear regression analyses with inverse probability weighting indicated no association between homogeneous social networks and brain volume changes. However, lower network heterogeneity was significantly associated with a greater decline in hippocampal volume. This was particularly pronounced in the right hippocampus.

Conclusion: Greater social network heterogeneity may attenuate age-related hippocampal volume loss in older adults, highlighting potential benefits of social network heterogeneity for structural brain aging and cognitive aging in later life. Further studies are warranted to clarify the underlying mechanisms.

Background: Long-term trends in cancer incidence were reported until 2015 in Japan, but recent data have been limited since the launch of the National Cancer Registry (NCR) in 2016. We analyzed 30-year trends in cancer incidence and mortality, emphasizing the NCR’s establishment.

Methods: Cancer incidence data were obtained from high-quality population-based cancer registries in three prefectures (Yamagata, Fukui, and Nagasaki) from 1985 to 2015, and the NCR extracted data from these prefectures between 2016 and 2021. National cancer mortality data from 1979 to 2023 were obtained from published vital statistics. Trends in age-standardized rates (ASR) calculated using the 1985 Japan Standard Population were examined using joinpoint regression analysis.

Results: The ASR incidence in 2016 (the first year of the NCR) was the highest ever recorded. All-cancer incidence (for both sexes) increased between 1985 and 2021 (annual percentage change, 0.9%; 95% confidence interval [95% CI]: 0.8 to 1.0). For males, all-cancer incidence stabilized after increasing until 2017, whereas for females, it increased after 2002 (1.7%; 95% CI: 1.4 to 2.3). All-cancer mortality (for both sexes) declined between 1997 and 2023 (-1.5%; 95% CI: -1.5 to -1.4), mainly due to decreases in stomach, lung, and liver cancers.

Conclusions: The NCR’s launch and the sharp rise in incidence in 2016 may have influenced trend interpretation and should be considered with caution. A longer observation period is needed before clear secular trends can be established under the NCR data system.