The need for holistic assessment in psychiatry is becoming ever more pertinent. Therefore, we were pleased to read Boyer et al’s randomised controlled trial.^{Reference Boyer, Lançon, Baumstarck, Parola, Berbis and Auquier1} The team investigated whether assessing and feeding back a quality of life (QoL) measure to the patient’s care team as part of a psychiatric assessment would improve patient satisfaction when compared with both standard psychiatric assessment and with measuring QoL but not informing the care team of the results.

We were also initially pleased to read that ‘global satisfaction was significantly higher in the QoL feedback group […] compared with the standard psychiatric assessment […] and QoL assessment groups’. However, on closer inspection, it appears that this interpretation of the results is not correct.

The primary outcome was level of patient satisfaction in the QoL feedback group compared with standard psychiatric assessment. In this comparison, 29 out of 40 patients (72.5%) in the QoL feedback group were ‘very satisfied’ with their care, compared with 27 out of 40 (67.5%) in the control group. This difference of 5% is far too small to be statistically significant with the sample size used.

Indeed, when we undertook our own basic statistical analysis, we found that the 95% CIs for relative risk ratio between these two groups were –37% to 22% - clearly not significant.

The correct interpretation of these results is that the study actually provides no evidence for assessing and feeding back a QoL measure in preference to simply not measuring QoL at all. The conclusions drawn by the authors, that their findings ‘provide strong support for integrating QoL assessment and feedback’ and that ‘priority should be given to strategies to implement QoL measurements in routine practice’ seem particularly unfounded.

Although we agree that QoL measures represent a potentially highly useful clinical tool, we cannot accept that Boyer et al’s study provides evidence for this claim in any way.

We felt that the most salient finding from the trial was in fact the far lower satisfaction in the control group of patients who had their QoL assessed but had the results ignored. If we offer an assessment or intervention, we should be careful to follow-up our intentions or the result may actually be detrimental overall.