Methodological idiosyncracies, frameworks and challenges of non-pharmaceutical and non-technical treatment interventions.

In this brief article which summarises a presentation given at the "6. Diskussionsforum zur Nutzenbewertung im Gesundheitswesen" of the German Ministry of Education and Research "Gesundheitsforschungsrat (GFR)" and the Institute for Quality and Efficiency in Healthcare (IQWiG) I will analyse some methodological idiosyncrasies of studies evaluating non-pharmacological non-technical interventions (NPNTI). I will focus on how the methodological framework of the Grading of Recommendations Assessment, Development and Evaluation (GRADE) working group may support design and appraisal of NPNTI. Specific design features that may be of particular value in NPNTI research, such as expertise-based randomised controlled trials, will be briefly described. Finally, based on an example, I will argue that - despite the methodological idiosyncrasies - there is neither a sufficient reason to accept different standards for the assessment of the confidence in the evidence from NPNTI nor for using study designs that are less rigorous compared to "simpler" interventions but that special measures have to be taken to reduce the risk of bias. The example that will be used in this article will primarily come from the field of respiratory rehabilitation, a typical multi-component or complex intervention and by definition a complex NPNTI, which has been evaluated in many randomised controlled trials. (As supplied by publisher).