Table 3.

Logistic regression models testing the associations between receptive and participatory arts engagement (measured in 2014) and subsequent healthy aging (measured in 2016 and 2018).

	Healthy aging 2 years later					Healthy aging 4 years later
	Healthy	Unadjusted		Adjusted		Healthy	Unadjusted		Adjusted
	N (%)	OR (95% CI)	p value	OR (95% CI)	p value	N (%)	OR (95% CI)	p value	OR (95% CI)	p value
Receptive arts
Never	136 (32%)	–		–		126 (29%)	–		–
<1 a month	267 (50%)	1.34 (0.91, 1.98)	.136	1.00 (0.65, 1.53)	.993	250 (47%)	1.42 (0.94, 2.16)	.098	1.08 (0.67, 1.72)	.758
≥1 a month	147 (49%)	1.33 (0.85, 2.07)	.207	1.04 (0.64, 1.70)	.861	150 (50%)	1.80 (1.10, 2.96)	.019	1.44 (0.84, 2.46)	.179
Participatory arts
0 activities	80 (35%)	–		–		85 (38%)	–		–
1 activity	172 (43%)	1.00 (0.62, 1.63)	.985	0.99 (0.58, 1.70)	.981	158 (39%)	0.68 (0.40, 1.16)	.157	0.63 (0.35, 1.12)	.114
2 activities	169 (46%)	1.33 (0.80, 2.22)	.267	1.27 (0.72, 2.24)	.414	154 (42%)	0.83 (0.49, 1.41)	.491	0.74 (0.42, 1.31)	.303
3+ activities	129 (47%)	1.06 (0.63, 1.79)	.813	0.94 (0.52, 1.69)	.835	130 (48%)	0.87 (0.49, 1.54)	.625	0.68 (0.35, 1.30)	.238

Note. OR: odds ratio. 95% CI: 95% confidence interval. Total N = 1269. Results based on 40 multiply imputed data sets. Reference categories are indicated with dashes. All models included both receptive and participatory arts exposures and were adjusted for healthy aging status in 2014. Adjusted models were additionally adjusted for age, gender, race/ethnicity, education, marital status, employment status, household income, and neighborhood safety. Bold text indicates p < .05.