Operationalising the capability approach for outcome measurement in mental health research

Abstract

Amartya Sen's multidimensional capability approach focuses on the importance of freedoms to be or do things people have reason to value. It is an alternative to standard utilitarian welfarism, the theoretical approach to quality-adjusted life years (QALYs) and cost-utility analyses. Despite the limitations of the utility approach in capturing non-health benefits and broader welfare inequalities, there have been very limited applications of the capability approach in the mental health context where these issues are imperative.We report the development and application of a multidimensional instrument, the OxCAP-MH, which aims to operationalise the capability approach for outcome measurement in mental health research. The study was carried out as part of an ongoing programme on community coercion experienced by service users with severe and enduring mental illness being treated using Community Treatment Orders. Capabilities data were collected at baseline in the OCTET RCT for 333 ‘revolving door’ mental health service users who were in involuntary hospital treatment at the time of recruitment in England (2008–2011). The research focused on the identification of capabilities domains most affected by mental illness and their association with socio-demographic and clinical factors and other measures of well-being such as the EQ-5D and Global Assessment of Functioning (GAF) scales.The OxCAP-MH item response rate was 90%–68%. There were significant correlations between service users' overall capability scores and the GAF, EQ-5D VAS and EQ-5D-3L utilities (corr = 0.249, 0.514, 0.415, respectively). The most affected capability domains were: ‘Daily activities’, ‘Influencing local decisions’, ‘Enjoying recreation’, ‘Planning one's life’ and ‘Discrimination’. Age had a mixed effect, while female service users and those with a primary diagnosis of schizophrenia or longer illness duration reported significantly lower capability scores. The results support the feasibility and validity of directly measuring human capabilities for the mentally ill and the potential for applying the approach to outcome measurement.