Health utilities of type 2 diabetes-related complications: a cross-sectional study in sweden.

Aliasghar Ahmad Kiadaliri, Ulf Gerdtham, Björn Eliasson, Soffia Gudbjörnsdottir, Ann-Marie Svensson, Katarina Steen Carlsson

Research output: Contribution to journalArticlepeer-review

19 Citations (SciVal)
172 Downloads (Pure)


This study estimates health utilities (HU) in Sweden for a range of type 2 diabetes-related complications using EQ-5D and two alternative tariffs (UK and Swedish) from 1757 patients with type 2 diabetes from the Swedish National Diabetes Register (NDR). Ordinary least squares were used for statistical analysis. Lower HU was found for female gender, younger age at diagnosis, higher BMI, and history of complications. Microvascular and macrovascular complications had the most negative effect on HU among women and men, respectively. The greatest decline in HU was associated with kidney disorders (-0.114) using the UK tariff and stroke (-0.059) using the Swedish tariff. Multiple stroke and non-acute ischaemic heart disease had higher negative effect than a single event. With the UK tariff, each year elapsed since the last microvascular/macrovascular complication was associated with 0.013 and 0.007 units higher HU, respectively. We found important heterogeneities in effects of complications on HU in terms of gender, multiple event, and time. The Swedish tariff gave smaller estimates and so may result in less cost-effective interventions than the UK tariff. These results suggest that incorporating subgroup-specific HU in cost-utility analyses might provide more insight for informed decision-making.
Original languageEnglish
Pages (from-to)4939-4952
JournalInternational Journal of Environmental Research and Public Health
Issue number5
Publication statusPublished - 2014

Subject classification (UKÄ)

  • Environmental Health and Occupational Health


Dive into the research topics of 'Health utilities of type 2 diabetes-related complications: a cross-sectional study in sweden.'. Together they form a unique fingerprint.

Cite this