Potential causal effect of physical activity on reducing the risk of dementia: a 6-year cohort study from the Japan Gerontological Evaluation Study

Table 2 Cox regressions for the association between physical activity and dementia

	Conventional analysis			Instrumental variable analysis
	HR	95% CI		HR	95% CI
Year 1	0.75	0.69	0.82	0.53	0.39	0.74
Year 2	0.79	0.74	0.84	0.63	0.49	0.82
Year 3	0.86	0.81	0.91	0.75	0.58	0.96
Year 4	0.90	0.85	0.95	0.69	0.53	0.90
Year 5	0.94	0.90	0.99	0.82	0.63	1.05
Year 6	0.95	0.92	0.99	0.85	0.66	1.10
F-statistic^a	–			80.1

All models included following covariates: sex, age, educational attainment, annual equivalized household income, marital status, paid work, the existence of heart disease, stroke, and diabetes, depressive symptoms, drinking habits, smoking status, family support, contact with friends, participation in community groups, population density, average degree of slopes, annual hours of sunlight, and fixed effects of municipalities
Abbreviations: HR Hazard ratio, CI Confidence interval
^aPartial F-statistic for a snowy area at the first-stage estimation

ISSN: 1479-5868