ABSTRACT
Background: Bipolar disorder is among the top 10 causes of disability worldwide. The Short-Form Six-Dimension (SF-6D) is a frequently used measure of preference-based health-related quality of life (HRQOL). However, this measure's psychometric performance has not been tested in outpatient patients with bipolar disorder.
Aims of the study: This study assessed the psychometric properties of the SF-6D, including convergent validity, known-groups validity, and responsiveness.
Methods: We examined convergent validity between the SF-6D and four condition-specific measures of functioning (LIFE-RIFT), life satisfaction (QLESQ), depressive symptoms (MADRS), and manic symptoms (YMRS). We used known-groups validity tests to compare the SF-6D health utility values estimated for patients in different clinical states, including depression, mania, hypomania, and recovered. We assessed the responsiveness of the SF-6D by comparing the sensitivity of the SF-6D utility values to longitudinal changes in the four condition-specific measures during the same period of time. We conducted all analyses using data from 2,627 participants in the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) clinical trial.
Results: The SF-6D demonstrated moderate (0.3-0.7) convergence with the LIFE-RIFT, QLESQ, and MADRS measures. Convergence with the manic symptoms measure (YMRS) was weak (<0.3). For known-groups validity, the SF-6D distinguished the recovered state from the three symptomatic clinical states. For responsiveness, the measure did not show floor or ceiling effects. The SF-6D utility value increased when mental health improved, with a small ES of 0.3 over the 1-year period, which was comparable to the four condition-specific measures.
Discussion: The SF-6D demonstrated moderate convergent validity, moderate responsiveness, and it can distinguish the differences between known-groups that had been identified in literature. The SF-6D may be a suitable measure of preference-based HRQOL for patients with bipolar disorder, but caution is needed due to its lower convergence with the YMRS mania scale.
Limitations: The subsample of patients in manic episode was small, which may reduce the reliability of study findings regarding this specific clinical state. In terms of generalizability, the STEP-BD study sample is based on patients who received treatment in bipolar specialty clinics affiliated with academic medical centers, which may be different from other outpatient clinics.
Implications for health care provision and use: The mean health utility value for patients with hypomania is significantly lower than the mean value for recovered patients. This finding emphasizes the importance of treating hypomania.
Implication for health policies: This study validates an existing approach toward generating health utility values for bipolar disorder. These utility values can be used to create quality-adjusted life years (QALYs), which are the most commonly used measure of health benefit in cost-effectiveness studies.
Implications for further research: Studies with larger samples of patients with mania are needed to study measures of health utility in this patient population.