Article Text

Download PDFPDF
Inequality in provision of medical care in Sweden: a case of social epidemiological hypochondria?
  1. Jay S Kaufman
  1. Department of Epidemiology, Biostatistics, and Occupational Health, McGill University, Montreal, Quebec, Canada
  1. Correspondence to Jay S Kaufman, Department of Epidemiology, Biostatistics, and Occupational Health, McGill University, 1020 Pine Ave West, Montreal, Quebec H3A 1A2, Canada; jay.kaufman{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Socioeconomic disparities in health arise through multiple pathways, including differential exposure,1 potentiation of exposure effects by comorbidities or other vulnerabilities,2 differential access to medical intervention,3 and lower quality or intensity of medical care. The last of these is the focus of the methodologically impressive new study by Ohlsson and colleagues in this issue of JECH (see page 678).4 The authors use a remarkably comprehensive data resource, which captures 13% of Sweden's entire population in a multi-level, linked-record database that includes five decades of population surveillance.5 Using linkage to the Swedish Prescribed Drug Register, the authors were able to track all residents of the Skåne region of Sweden who received a statin prescription from a physician during a 6-month period in 2005. They investigated whether use of recommended treatment was associated with patients' social characteristics, such as marital status and income, or with clinic characteristics, such as percentage of high-income patients seen at the facility. The authors conducted the statistical analysis using sophisticated hierarchical models that not only accounted properly for the clustered nature of the data (ie, for unmeasured similarities among patients within a clinic), but which also allowed for explicit decomposition of the variance into between-clinic/area and within-clinic/area …

View Full Text


  • Linked articles 81166.

  • Competing interests None.

  • Provenance and peer review Commissioned; not externally peer reviewed.

Linked Articles