Physical activity and chronic diseases among older people in a mid-size city in China: a longitudinal investigation of bipolar effects

Table 4 Logistic regression^a between the onset of chronic diseases and physical activity behaviors 2010 to 2015

	Physical Activity^b
	Initiate PA vs. Sedentary		Change vs. Remain		Strengthen vs. Lighten
	Odds Ratio	95% CI	Odds Ratio	95% CI	Odds Ratio	95% CI
Any onset	0.75**^c	0.57–0.97	1.23*	1.09–1.38	2.53***	2.08–3.08
Onset of the First	0.77	0.45–1.31	0.83	0.62–1.10	1.95**	1.17–3.24
Onset of the Second	0.94	0.61–1.44	1.10	0.88–1.38	1.89**	1.29–2.77
Onset of the Third	0.60	0.35–1.00	1.23	0.98–1.55	2.13*	1.42–3.19
Types of Diseases
Overweight	1.10	0.39–3.21	1.99**	1.37–2.89	2.86***	1.57–5.21
Hypertension	1.33	0.60–2.96	2.44***	1.82–3.28	3.22***	2.08–4.99
Diabetes	1.03	0.5202.04	0.73*	0.57–0.96	1.71*	1.04–2.84
Hyperlipidemia	1.11	0.59–2.08	1.59**	1.24–2.02	1.79***	1.25–2.58
Cardiovascular Diseases	0.64*	0.41–0.95	0.48***	0.39–0.58	0.89	0.63–1.25
Liver& Biliary System Diseases	0.39*	0.16–0.94	0.95	0.67–1.30	1.15	0.68–1.92
Kidney Function Disorder	0.45	0.16–1.24	0.59**	0.41–0.90	0.75	0.38–1.46

^aThree logistic regression models were used to examine the association between the change in PA and the onsets of chronic diseases controlling for age, gender, education, and income
^bFor the sedentary individuals at baseline, the response in model 1 was if he/she initiate PA in the following year after the onset of their disease; for the individuals with PA behaviors, the response in model 2 was if he/she changed or remained at their PA level in the following year; for those who did change their PA level in model 2, the response in model 3 was if they increased or decreased their PA level in the following year after the onset of a certain chronic disease
^c*p < 0.05, **p < 0.01, ***p < 0.001

ISSN: 1471-2458