Table 2. Multilevel logistic regression analysis of psychotropic drug use in the 35–65 year-old population of Malmö, 2006.
Values are odds ratios (OR) with 95% confidence interval (CI) unless stated otherwise. The intercept is not shown in the table.
| Simple logistic regression analysis | Multilevel logistic regression analysis | ||
|---|---|---|---|
| Model 1 | Model 2 | Model 3 | |
| Specific Individual Average Effects | |||
| Men vs. women | 0.61 (0.58–0.64) | 0.60 (0.58–0.63) | 0.60 (0.58–0.63) |
| Age groups | |||
| 35–39 | Reference | ||
| 40–44 | 1.35 (1.24–1.46) | 1.35 (1.25–1.45) | 1.35 (1.24–1.46) |
| 45–49 | 1.63 (1.50–1.77) | 1.64 (1.51–1.77) | 1.63 (1.51–1.77) |
| 50–54 | 1.81 (1.67–1.95) | 1.82 (1.69–1.96) | 1.82 (1.68–1.97) |
| 55–59 | 1.91 (1.76–2.07) | 1.94 (1.80–2.10) | 1.95 (1.81–2.10) |
| 60–64 | 1.95 (1.80–2.11) | 2.00 (1.85–2.16) | 2.01 (1.86–2.17) |
| Low vs. high income | 1.67 (1.60–1.74) | 1.56 (1.49–1.64) | 1.52 (1.44–1.59) |
| Specific Contextual Average Effects | |||
| Low vs. high neighbourhood income | 1.29 (1.21–1.38) | ||
| 80% IOR | 0.99–1.69 | ||
| POOR (%) | 11 | ||
| General contextual effects | |||
| Neighbourhood variance | 0.038 (0.026–0.054) | 0.022 (0.012–0.035) | |
| PCV (%) | 42 | ||
| ICC (%) | 1.1 (0.8–1.6) | 0.7 (0.4–1.1) | |
| MOR | 1.20 (1.17–1.25) | 1.16 (1.11–1.20) | |
| AUC | 0.616 (0.610–0.622) | 0.630 (0.625–0.636) | 0.629 (0.623–0.635) |
| AUC change* | 0.014 | -0.001 | |
| Goodness of fit | |||
| DIC | 48205 | 48063 | 48041 |
| DIC change* | -142 | -22 | |
IOR: interval odds ratio. POOR: proportion of opposed odds ratios. PCV: proportional change in the variance. ICC: intra-class correlation coefficient. MOR: median odds ratio. AUC: area under the receiver operating characteristic curve. DIC: Bayesian deviance information criterion.
*: Change in relation to the previous model.