A comparison of SF-6D and EQ-5D utility scores in a study of patients with schizophrenia

Abstract

Background: Economic evaluations of healthcare interventions increasingly measure outcomes using quality-adjusted life years (QALYs). The SF-6D and the EQ-5D are alternative ways of generating utility scores for use in QALY estimations, but it is unclear which is most sensitive to change in psychiatric symptom severity. There are also limited data on the sensitivity of these measures to changes in existing clinical indicators in long-term mental health conditions like schizophrenia.

Aims of the study: To: (i) assess the relationship between SF-6D and EQ-5D utility scores for patients with schizophrenia at two points in time, (ii) assess the relationship in the change scores of these two measures, (iii) measure the sensitivity of these measures to changes in an established measure of symptomatology.

Methods: Patients with schizophrenia were recruited and the SF-36 and EQ-5D were administered at baseline and one-year follow-up and utility scores were computed and compared. Standardized response mean (SMR) scores were calculated for the SF-6D and EQ-5D and compared for patients who improved or deteriorated by at least 25% on the Brief Psychiatric Rating Scale.

Results: EQ-5D ratings were available for 394 patients at baseline, 368 at follow-up and 358 at both time points. The respective figures for the SF-6D were 383, 367 and 345. Mean utility scores were very similar at baseline (EQ-5D 0.68, SF-6D 0.67) and follow-up (EQ-5D 0.71, SF-6D 0.68). Median scores were markedly higher for the EQ-5D (0.76 v 0.66 at baseline, 0.80 v 0.68 at follow-up). The SF-6D scores followed a normal distribution whilst the EQ-5D scores were negatively skewed with a clustering at 1.00. There were few differences in sensitivity to change between the EQ-5D and SF-6D.

Discussion: From an analytical perspective the SF-6D has advantages over the EQ-5D due to its normal distribution and lack of ceiling effect. However, both measures produce similar mean utility scores. Overall the SF-6D appears more suitable as a measure of utility in this patient group.

Implications for health policies: Decisions made on the basis of cost-effectiveness results need to consider the method by which QALYs have been calculated.

Implications for further research: Further comparisons of the EQ-5D and SF-6D are required.