[Size and quality of clinical research on non-medical interventions: a systematic analysis of IQWiG reports].

Background: A sufficient number of randomised controlled clinical trials (RCTs) with high methodological reliability are required to allow a critical assessment of therapeutic interventions. The aim of this study was to evaluate the final reports of the Institute for Quality and Efficiency in Health Care (IQWiG) addressing non-medical interventions. These reports were analysed in order to determine how many endpoint-related comparisons were evaluated by clinical studies, how many by RCTs and how many by methodologically reliable RCTs.

Methods: Seventeen out of eighteen final reports of the IQWiG department of non-medical interventions were critically analysed. The final reports had been generated from 2006 till 2009. Comparison of therapeutic procedures and patient relevant endpoints were taken from the final reports. Additional patient relevant endpoints were considered whenever these had been included in the result section. RCTs considered essential for the assessment were identified. RCTs obtaining the biometric rating no shortcomings, moderate shortcomings or low bias were considered to be of methodological reliability.

Results: A total of 537 comparisons addressing single endpoints were identified. Clinical studies were available for 50% of these comparisons (n=272). For 38% of the comparisons (n=205) RCTs could be included, but in only 18% these RCTs had reliable methods. The number of reports containing zero, one, and more than one RCT was 6, 4, and 7, respectively. Methodologically reliable RCTs were included in only 8 out of 17 final reports.

Discussion: The analysis of evidence showed that both the number and the quality of studies were insufficient. This is why better standards for the evaluation of non-medical interventions need to be established. This may be achieved by raising the regulatory requirements and by intensifying research from independent institutions.