File(s) under permanent embargo

Burden and health state utility values of eating disorders: results from a population-based survey

journal contribution
posted on 31.10.2019, 00:00 authored by Long Le, Cathy Mihalopoulos, Lidia Engel, S Touyz, D A González-Chica, N Stocks, P Hay
BackgroundThere are no published estimates of the health state utility values (HSUVs) for a broad range of eating disorders (EDs). HSUVs are used in economic evaluations to determine quality-adjusted life years or as a measure of disorder burden. The main objective of the current study is to present HSUVs for a broad range of EDs based on DSM-5 diagnoses.MethodsWe used pooled data of two Health Omnibus Surveys (2015 and 2016) including representative samples of individuals aged 15 + years living in South Australia. HSUVs were derived from the SF-6D (based on the SF-12 health-related quality of life questionnaire) and analysed by ED classification, ED symptoms (frequency of binge-eating or distress associated to binge eating) and weight status. Multiple linear regression models, adjusted for socio-demographics, were used to test the differences of HSUVs across ED groups.ResultsOverall, 18% of the 5609 individuals met criteria for ED threshold and subthreshold. EDs were associated with HSUV decrements, especially if they were severe disorders (compared to non-ED), binge ED: −0.16 (95% CI −0.19 to −0.13), bulimia nervosa: −0.12, (95% CI −0.16 to −0.08). There was an inverse relationship between distress related binge eating and HSUVs. HSUVs were lower among people with overweight/obese compared to those with healthy weight regardless of ED diagnosis.ConclusionsEDs were significantly associated with lower HSUVs compared to people without such disorders. This study, therefore, provides new insights into the burden of EDs. The derived HSUVs can also be used to populate future economic models.



Psychological medicine


1 - 8


Cambridge University Press


Cambridge, Eng.





Publication classification

C1 Refereed article in a scholarly journal