Weighted multivariate logistic regression for factors associated with cost-related medication underuse in the overall cohort*
| Factors | OR (95% CI) |
|---|---|
| Age, yr | |
| ≥ 65 | Reference |
| 18–64 | 2.55 (1.79–3.63) |
| Sex | |
| Female | Reference |
| Male | 0.62 (0.44–0.88) |
| Racial background | |
| White | Reference |
| Indigenous | 2.37 (1.49–3.77) |
| Other | 0.86 (0.49–1.52) |
| Marital status | |
| Married | Reference |
| Unmarried | 1.59 (1.09–2.30) |
| Self-perceived health | |
| Poor | Reference |
| Excellent | 0.36 (0.17–0.77) |
| Very good | 0.26 (0.12–0.53) |
| Good | 0.56 (0.29–1.07) |
| Fair | 0.80 (0.43–1.46) |
| Self-perceived mental health | |
| Poor | Reference |
| Excellent | 0.60 (0.25–1.42) |
| Very good | 1.11 (0.46–2.70) |
| Good | 1.04 (0.43–2.49) |
| Fair | 1.29 (0.57–2.90) |
| Total household income, $ | |
| ≥ 80 000 | Reference |
| < 20 000 | 3.08 (1.75–5.41) |
| 20 000–39 999 | 3.32 (1.95–5.64) |
| 40 000–59 999 | 2.68 (1.62–4.42) |
| 60 000–79 999 | 1.82 (0.93–3.55) |
| Education | |
| Postsecondary education | Reference |
| Less than secondary school | 0.66 (0.39–1.09) |
| Secondary school | 0.84 (0.59–1.21) |
| Insurance for prescription medications | |
| Yes | Reference |
| No | 2.49 (1.77–3.50) |
Note: CI = confidence interval, OR = odds ratio.
↵* Participants with missing variables were excluded (2.9%); 8333 participants (97.1%) were included in the logistic regression analysis.