Table 1.
Characteristics of the population 35–65 year-olds in Malmö, 2006 by neighbourhood income.
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.
Fig 1.
Areas under the receiver operating characteristic (AUC) curve for use of psychotropic drugs during 2006 in the city of Malmö, Sweden plotted separately for Model 1 which only adjusts for individual-level covariates age, sex and income (black thick line), and Model 2 which additionally adjust for neighbourhood of residence (grey dotted line).
The diagonal line represents an AUC equal to 0.50.
Fig 2.
Predicted prevalence of psychotropic drug use in each neighborhood with 95% confidence intervals versus ranking.
Predictions are for the reference individual in model 2, i.e.,female age 35–39, high income.
Fig 3.
Predicted prevalence of psychotropic drug use in each neighborhood with 95% confidence intervals versus ranking.
Predictions are for the reference individual in model 3, i.e., female, age 35–39, high income and living in a high income neighborhood (model 3).
Table 3.
Multilevel logistic regression analysis of choosing a private versus a public specialist in the 35–65 year-old population of Malmö, 2006, Values are odds ratios (OR) and 95% confidence interval (CI) unless stated otherwise.
Fig 4.
Areas under the receiver operating characteristic curve (AUC) for choosing a private vs. a public GP during 2006 in the city of Malmö, Sweden plotted separately for Model 1 which only adjusts for individual-level covariates age, gender and income (black thick line); and for Model 2 which additionally adjusts for the neighbourhood of residence (grey dotted line).
The diagonal line represents an AUC equal to 0.50.
Fig 5.
Predicted prevalence of using a private physician in each neighborhood with 95% confidence intervals versus ranking.
Predictions are for the reference individual in model 2, i.e., female, age 35–39, low income.
Fig 6.
Predicted prevalence of using a private physician in each neighborhood with 95% confidence intervals versus ranking.
Predictions are for the reference individual in model 3, i.e., female, age 35–39, low income and living in a low income neighborhood.