Ascertaining minimal clinically meaningful changes in symptoms of depression rated by the 15-item Centre for Epidemiologic Studies Depression Scale

Abstract

Rationale, aims and objectives: In clinical practise and in clinical studies on depression it is important to estimate whether changes in symptomatology measured by self-rating instruments are, in fact, clinically relevant. Therefore, the aim of the study was to estimate the clinical relevance of changes on the 15-item version of the Centre for Epidemiologic Studies Depression Scale (CES-D-15) based on the concept of the minimal clinically important difference (MCID).

Methods: Data was acquired from 4781 patients with depression symptoms from a German psychosomatic hospital who have been assessed using the CES-D-15 before and after treatment. Threshold values representing the MCID were estimated on the basis of mean change scores and sensitivity/specificity analyses. Patients' global impression of change, clinical (therapists') global impression of change and change in impairment severity were used as external anchor criteria.

Results: On average, the MCID was represented by a reduction of approximately 11 points in the CES-D-15, irrespective of age, gender, type of treatment and first or secondary diagnosis. However, higher baseline scores in the CES-D-15 required larger changes of raw values to represent a clinically important difference.

Conclusions: Anchor-based values are suggested here as an estimation of the clinical relevance of changes in the CES-D-15. Thus, instead of relying solely on effect sizes, the evaluation of treatment outcomes should be supplemented by reporting the percentage of patients who have reached the MCID. Further examinations to verify our results in other patient populations and with other types of anchor criteria will be needed.

Keywords: clinical significance; depression rating scale; minimal clinically important difference; minimal detectable change; outcome measurement; psychosomatic medicine.