OBJECTIVES: Health technology assessment (HTA) agencies have adopted different methodological approaches, which – at least in part – reflect different value judgments. The Federal Joint Committee (Gemeinsamer Bundesausschuss, GBA) and the Institute for Quality and Efficiency in Health Care (IQWiG) in Germany as well as the National Institute for Health and Care Excellence (NICE) in England may be regarded as exemplars for the implementation of evidence-based medicine (EbM) and the logic of cost effectiveness, respectively. The present study aims to explore the extent to which these agencies follow their own assessment criteria, and to expound exceptions that have been made.

METHODS: We extracted data from all publically available GBA appraisals and IQWiG benefit assessments between January 2011 and April 2015, as well as all NICE single technology appraisals (STAs) completed during this period. We then analyzed benefit assessment results focusing on official assessment criteria by GBA/IQWiG (clinical evidence and patient-relevant endpoints) and by NICE (clinical evidence, incremental cost effectiveness ratios (ICERs), and end of life (EoL) considerations).

RESULTS: GBA (105 appraisals comprising 226 subgroups) and IQWiG (105 assessments with 240 subgroups) benefit determinations were predominantly driven by patient-relevant endpoints (p<0.01), although orphan drugs were legally stipulated to offer added benefit. NICE STAs (88 appraisals with 125 subgroups) adhered to the ICER as a primary assessment criterion (p<0.01), even though 6 appraisals (15 subgroups) without an estimated ICER relied on other criteria. Evaluations with an ICER above GBP30,000/QALY were influenced by EoL criteria (p<0.01). Randomized controlled trials (RCTs) are considered to be most appropriate to measure clinical effectiveness by both NICE (p<0.05) and GBA/IQWiG (p<0.01); nevertheless, positive HTA outcomes presume that assessment criteria were met.

CONCLUSIONS: Overall, our results confirm that both agencies follow their own assessment criteria in a consistent manner. We identified a limited number of exceptions, which we will report in detail.