Datasets used in the study
| Turin | Stockholm | Finnish cities | |
|---|---|---|---|
| Data source* | Census linked to prescription register | Administrative registers and Swedish prescribed drug register | Administrative registers and prescription register |
| Sample coverage | 100% | 100% | 11%+80% oversample† |
| Baseline year | 2001 | 2011 | 2003 |
| Follow-up period | 2002–2007 | 2012–2017 | 2004–2009 |
| Urban Atlas data year | 2006 | 2012 | 2012 |
| Number of areas | 92 statistical zones | 233 basic areas | 339 postal codes |
| Median area population (IQR) | 6761 (2358–12 790) | 4926 (2654–8748) | 3785 (1413–6649) |
*Register linkages were authorised by national statistical or data protection authorities in each country.
†The Finnish dataset was a nationally representative 11% random sample of all Finnish residents in 2003. To increase the statistical power for small-area analysis, Statistics Finland amended the data with an 80% oversample of persons who died in 2004–2007 (probability weights were used in the analysis to account for the sampling design).