[Sensitivity to change of questionnaires measuring subjective health--results of a prospective comparative study]

Questionnaires measuring patients' subjective health or health-related quality of life are indispensable tools for the evaluation of effects revealed by intervention studies in the field of medical rehabilitation. These patient-reported outcomes should appropriately reflect change over time. Unfortunately, "sensitivity to change" has so far not been adequately examined for German health-related quality of life questionnaires, especially not in a comparative way. Therefore, indices of sensitivity to change for three widespread generic assessment tools have been determined: IRES-3, SF-36, scales of the SCL-90-R.

A prospective comparative study was conducted in n = 1145 inpatients with orthopaedic/rheumatologic and cardiac diseases from 16 rehabilitation clinics. All patients received usual care. Their subjective health-status was assessed at two to four weeks before admission (t0), admission (t1), discharge (t2), and three months after discharge (t3). At each time point, they completed the IRES-3, SF-36, and relevant scales of the SCL-90-R. For the time interval t1-t2, Guyatt's responsiveness index (GRI) was calculated and compared across scales and instruments.

Virtually all GRI coefficients for scales and aggregated scores, respectively, reached statistical significance. With respect to the GRI distributions of the diagnostic groups, most coefficients were located in a middle to upper range. While the results for the scales do not clearly indicate which assessment instrument should be preferred, GRI coefficients for higher aggregated scores suggest the IRES-3 to be most sensitive to change.

These results can be helpful in selecting a health-related quality of life instrument or certain subscales for evaluation studies in the field of medical rehabilitation.