In their study of rehabilitation services in England, Killaspy and co-authors conclude that ‘Quality of care is linked to better clinical outcomes in people with complex and longer-term mental health problems’. ^{Reference Killaspy, Marston, Omar, Green, Harrison and Lean1} We are not persuaded that this conclusion is justified by the results reported.

The study investigated the relationship between measures of quality and clinical outcomes across a number of services at different locations in England. The scales used to measure outcomes consisted of the Resident Choice Scale (RCS), the Manchester Short Assessment of Quality of Life (MANSA), the Your Treatment and Care (YTC), the General Milieu Index (GMI) and the Global Assessment of Functioning (GAF).

Our main argument is that of these the RCS, YTC and GMI are actually measures of the service user's perception of the quality of care rather than measuring what most people would understand by the term ‘clinical outcome’. Although these measures are widely referred to as outcome measures, we argue that the term ‘clinical outcome’ should refer to factors such as symptom level, relapse rates, functioning, carer burden and quality of life. The RCS, YTC and GMI scales were reported to be associated with the objective measures of service quality. However, we would argue that this result simply demonstrates that patients are able to make a valid judgement about the quality of the services they receive.

On the other hand, we agree that the MANSA and GAF are indeed measures of clinical outcome. However, the MANSA, which does measure quality of life, did not show any association with measures of service quality and the authors did not report any results for the GAF.

Based on these findings we are unable to agree with the authors' conclusion that they have demonstrated an association between service quality and clinical outcomes. This negative finding is aligned with that of a study that they cite and which failed to demonstrate an improvement in outcomes following the implementation of the quality and outcomes framework for diabetes. ^{Reference Calvert, Shankar, McManus, Lester and Freemantle2}

It seems that it is difficult to convincingly demonstrate a real relationship between care quality and clinical outcomes. However, we would argue that attempts to do this should use outcome measures of real relevance, such as the Health of the Nation Outcome Scales (HoNOS), ^{Reference Wing, Beevor, Curtis, Park, Hadden and Burns3} GAF and MANSA.