Structural physical activity restriction, cerebrovascular and diabetes mortality, and long-term care intensity in Japan: An ecological panel study across 47 prefectures (2013−2022) with a COVID-19 counterfactual analysis
Dialogues in Health · 2026-12 · 9 · 100320
DOI: 10.1016/j.dialog.2026.100320 ↗Kurz gefasst
Fachartikel in der Zeitschrift Dialogues in Health, 2026-12. Verfasst von Akira Kimura. Beteiligte Einrichtung: Division of Public Health, Graduate School of Health Sciences, Gunma Paz University, Takasaki, Gunma, Japan. Die Arbeit nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Themen: Langzeitpflege, Diabetes mellitus
Abstract (Originalsprache)
Background
Physical inactivity is an established cardiovascular risk factor at the individual level, but the population-level mortality cost attributable to modifiable, built-environment constraints on daily physical activity - rather than to individual behaviour - has lacked an empirical benchmark, leaving investment in walkable urban design without a clear health-economic case. We asked whether structurally constrained population physical activity is associated with cause-specific mortality, and used the 2020 COVID-19 mobility restriction - a society-wide curtailment that could never be imposed experimentally - to estimate the effect size of an acute collective PA shock on cardiovascular and metabolic deaths.
Methods
10-year ecological panel of Japan's 47 prefectures (2013-2022; N = 470). We modelled cerebrovascular (CVD), diabetes (DM), and heat-illness deaths with age-sex-standardized CRE Mundlak Poisson regression and the severe long-term-care ratio (SHR) with a fractional logit, applying stepwise confounder adjustment (M0-M5: medical resources, income, education, age structure, birth rate). 2020 excess was estimated against a counterfactual trend trained on 2013-2019.
Results
Higher car-only commuting (range 8.5-79.0%) was associated with greater CVD (IRR per SD = 1.064; 95% CI 1.037-1.093; +29% across the observed range) and DM mortality (IRR = 1.030; 1.000-1.060), and with greater long-term-care severity (SHR OR = 1.39; 1.18-1.64). CVD persisted across all five adjustment models; DM attenuated to non-significance after consumption-expenditure adjustment, consistent with a correlated socioeconomic cluster rather than an independent driver. In 2020, the most car-dependent tertile recorded +8.3% excess CVD deaths (95% CI +2.8% to +14.0%) versus near-zero excess elsewhere - a quantifiable effect of acute collective PA restriction on cardiovascular mortality.
Conclusion
A modifiable feature of modern life - car dependence - carries a measurable population-level cardiovascular mortality cost that amplifies when behavioural PA is acutely curtailed. The effect size justifies treating walkable urban design and public transit as primary NCD prevention in super-ageing societies, not as ancillary transport policy. Funding: None declared.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Division of Public Health, Graduate School of Health Sciences, Gunma Paz University, Takasaki, Gunma, Japan.
Zitieren
APA
Kimura, A. (2026). Structural physical activity restriction, cerebrovascular and diabetes mortality, and long-term care intensity in Japan: An ecological panel study across 47 prefectures (2013−2022) with a COVID-19 counterfactual analysis. Dialogues in Health, 9, 100320. https://doi.org/10.1016/j.dialog.2026.100320
Vancouver
Kimura A. Structural physical activity restriction, cerebrovascular and diabetes mortality, and long-term care intensity in Japan: An ecological panel study across 47 prefectures (2013−2022) with a COVID-19 counterfactual analysis. Dialogues in Health. 2026;9:100320. doi:10.1016/j.dialog.2026.100320
BibTeX
@article{Kimura2026,
author = {Kimura, Akira},
title = {Structural physical activity restriction, cerebrovascular and diabetes mortality, and long-term care intensity in Japan: An ecological panel study across 47 prefectures (2013−2022) with a COVID-19 counterfactual analysis},
journal = {Dialogues in Health},
year = {2026},
volume = {9},
pages = {100320},
doi = {10.1016/j.dialog.2026.100320}
}
EUVIMED