Table 4. Final results of the multivariate analysis of the characteristics associated with cost-related underuse of medications. Brazilian Longitudinal Study of Aging (ELSI-Brazil), 2015–2016.
| Variable | Adjusted PRa | 95%CI |
|---|---|---|
| Sex (ref: male) | ||
| Female | 1.39 | 1.16–1.67b |
| Age (ref: 50–59 years) | ||
| 60–79 | 0.75 | 0.63–0.89c |
| 80 and over | 0.43 | 0.28–0.67b |
| Socioeconomic status in quartiles (ref: Q1) | ||
| Q2 | 0.70 | 0.57–0.87c |
| Q3 | 0.79 | 0.64–0.97c |
| Q4 | 0.60 | 0.44–0.77b |
| Sufficiency of monthly family income for expenses (ref: always) | ||
| Sometimes | 1.74 | 1.28–2.35b |
| Never | 2.94 | 2.29–3.79b |
| Health insurance coverage (ref: no) | ||
| Yes | 0.79 | 0.63–0.98c |
| Frequency with which the physician explains about the disease and treatment (ref: frequently) | ||
| Rarely/Never | 1.31 | 1.09–1.57c |
| Number of medications used (ref: 1) | ||
| 2–4 | 1.39 | 1.11–1.74c |
| 5 or more | 1.53 | 1.16–2.02c |
| Self-rated health (ref: very good/good) | ||
| Fair | 2.02 | 1.60–2.57b |
| Poor/Very poor | 2.92 | 2.23–3.81b |
| Depression (ref: no) | ||
| Yes | 1.69 | 1.41–2.02b |
a Adjusted PR (95%CI): prevalence ratio estimated by Poisson regression model with robust variance, adjusted for all variables described in the table.
b p < 0.001
c p < 0.05