Unjustified variation in the use of healthcare services

About the Project

The overarching theme of this thesis was geographical and socio-economic variation in the consumption of specialist healthcare services. The thesis studied three different types of healthcare services. The analyses utilised complete population data at the individual level, with a high degree of completeness and quality, from national registers (Norwegian Patient Register, Cancer Registry, and Statistics Norway).

Research Questions

The primary objectives of the thesis were to:

1. explore geographical variation in the consumption of three different healthcare services in Norway
2. explore socio-economic variation in the consumption of three different healthcare services in Norway
3. investigate whether geographical variation in the consumption of these healthcare services could be explained by differences in socio-economic status

In Article I, hospital admissions for children aged 1-16 years were studied. In Article II, care pathways for lung, colorectal, prostate, and breast cancer were examined, focusing on two different populations; i) the proportion of care pathway patients who did not have the relevant cancer type, and ii) the proportion of cancer patients included in care pathways. In Article III, patients with atrial fibrillation and the proportion treated with ablation were studied.

Significant geographical and socio-economic variation was found, and the geographical variation could not be explained by differences in socio-economic status between the geographical areas. Children with parents of low educational attainment were more likely to be admitted and had the highest number of admissions per child, while also experiencing the least severe admissions. Cancer patients in the highest income group were most likely to be included in care pathways, whereas no systematic differences were found between income and education groups for the proportion of care pathway patients who did not have the relevant cancer type. Atrial fibrillation patients with higher education and/or higher income were more frequently treated with ablation, and the effect of education was stronger with increasing age.

The variation documented in this thesis challenges the perception that the healthcare provision and medical practice in Norway are rational and evidence-based. Capacity differences may likely explain parts of the geographical variation, and differences in need may likely account for some of the socio-economic variation. To understand the causes of variation, one must also study what underlies clinical decisions, i.e., both patients' and doctors' preferences, as well as local traditions and clinical practice.