Table 3.

Multivariate logistic regression models for the associations between baseline sensory loss, baseline prevalence of falls and incident falls over 7 years.

	Baseline sensory loss and baseline prevalence of falls
Variables	Model 1		Model 2		Model 3		Model 4
	OR	95%CI	OR	95%CI	OR	95%CI	OR	95%CI
No sensory loss	Reference		Reference		Reference		Reference
Hearing loss	1.503***	(1.240, 1.821)	1.440***	(1.186, 1.748)	1.417***	(1.166, 1.721)	1.394***	(1.147, 1.694)
Vision loss	1.330**	(1.075, 1.646)	1.260*	(1.017, 1.562)	1.246	(1.004, 1.546)	1.21	(0.974, 1.502)
Dual sensory loss	2.061***	(1.768, 2.404)	1.899***	(1.624, 2.220)	1.846***	(1.577, 2.161)	1.801***	(1.538, 2.110)
	Baseline sensory loss and incidence of falls over 7 years
Variables	Model 1		Model 2		Model 3		Model 4
	OR	95%CI	OR	95%CI	OR	95%CI	OR	95%CI
No sensory loss	Reference		Reference		Reference		Reference
Hearing loss	1.345***	(1.161, 1.558)	1.299***	(1.119, 1.508)	1.295***	(1.115, 1.504)	1.282**	(1.104, 1.490)
Vision loss	1.174*	(0.998, 1.381)	1.103	(0.936, 1.299)	1.105	(0.938, 1.303)	1.089	(0.924, 1.285)
Dual sensory loss	1.623***	(1.442, 1.826)	1.487***	(1.318, 1.677)	1.477***	(1.307, 1.668)	1.454***	(1.287, 1.643)

The results of the logistic regression models were expressed as odds ratios (OR) and 95% confidence intervals (CI). The analytic sample size was 7,623.

Model 1: unadjusted.

Model 2: adjusted for demographic factors including age and sex.

Model 3: adjusted for factors in Model 2, as well as social-economic and life style factors including BMI, marital status, educational level, smoking, alcohol consumption and physical activity status.

Model 4: adjusted for factors in Model 3, as well as physical disorders including hypertension, diabetes, dyslipidemia, kidney diseases, emotional disorders, memory-related diseases and stroke.

^*p < 0.05;

^**p < 0.01;

^***p < 0.001.