OBJECTIVES: In version 6.0 of its General Methods, IQWiG sought to address the increasing complexity of assessing minimal important differences (MIDs) for patient-reported outcomes (PROs), by stipulating that response thresholds for submitted responder analyses must be ≥ 15% of the scale range. This study investigated the impact of the General Methods update on IQWiG's assessments of PROs and how these have influenced G-BA's view on submitted PRO evidence.

METHODS: A comparison of IQWiG dossier assessments published six months before and six months after the publication of IQWiG's General Methods version 6.0 was conducted. The analysis focused on comparing benefit ratings and the impact of IQWiG's acceptance of PRO data on these. It further examined G-BA's consideration of responder analyses with thresholds < 15% of the scale range rejected by IQWiG, and the effect of submitting post hoc analyses during commenting procedures.

RESULTS: Of the 58 IQWiG assessments from the 6 months before the method update, an additional benefit was concluded in 17 (29%) assessments. Of the 27 (47%) assessments that considered PRO, results from PROs affected 14 (52%) benefit ratings. Of the 78 IQWiG assessments from the 6 months after the methods update, an additional benefit was concluded in 21 (27%) assessments. Of the 32 (41%) assessments that considered PRO, results from PROs affected 15 (47%) benefit ratings. In its assessments, G-BA mentioned that discussions were still ongoing as to whether a validated MID or a 15% response threshold is preferred. Post hoc analyses were submitted during several commenting procedures, but none changed IQWiG's benefit rating in published addenda to the dossier assessment.

CONCLUSIONS: IQWiG's General Methods update has simplified IQWiG's process for assessing responder analyses but has not had a clear impact on the proportion of dossier assessments with accepted PROs and benefit ratings affected by PROs.