Anthony J Hatswell1,2, Nick Freemantle3, Gianluca Baio4. 1. Department of Statistical Science, University College London, Gower Street, London, WC1E 6BT, UK. ahatswell@bresmed.com. 2. BresMed, 84 Queen Street, Sheffield, S1 2DW, UK. ahatswell@bresmed.com. 3. Department of Primary Care and Population Health, University College London, UKGower Street, London, WC1E 6BT, UK. 4. Department of Statistical Science, University College London, Gower Street, London, WC1E 6BT, UK.
Abstract
BACKGROUND: Pharmaceuticals are usually granted a marketing authorisation on the basis of randomised controlled trials (RCTs). Occasionally the efficacy of a treatment is assessed without a randomised comparator group (either active or placebo). OBJECTIVE: To identify and develop a taxonomic account of economic modelling approaches for pharmaceuticals licensed without RCT data. METHODS: We searched PubMed, the websites of UK health technology assessment bodies and the International Society for Pharmacoeconomics and Outcomes Research Scientific Presentations Database for assessments of treatments granted a marketing authorisation by the US Food and Drug Administration or European Medicines Agency from January 1999 to May 2014 without RCT data (74 indications). The outcome of interest was the approach to modelling efficacy data. RESULTS: Fifty-one unique models were identified in 29 peer-reviewed articles, 30 health technology appraisals, and 15 International Society for Pharmacoeconomics and Outcomes Research abstracts concerning 30 indications (44 indications had not been modelled). We noted the high rate of non-submission to health technology assessment agencies (28/98). The majority of models (43/51) were based on 'historical controls'-comparisons to previous meta-analysis or pooling of trials (5), individual trials (16), registries/case series (15), or expert opinion (7). Other approaches used the patient as their own control, performed threshold analysis, assumed time on treatment was added to overall survival, or performed cost-minimisation analysis. CONCLUSIONS: There is considerable variation in the quality and approach of models constructed for drugs granted a marketing authorisation without a RCT. The most common approach is of a naive comparison to historical data (using other trials/registry data as a control group), which has considerable scope for bias.
BACKGROUND: Pharmaceuticals are usually granted a marketing authorisation on the basis of randomised controlled trials (RCTs). Occasionally the efficacy of a treatment is assessed without a randomised comparator group (either active or placebo). OBJECTIVE: To identify and develop a taxonomic account of economic modelling approaches for pharmaceuticals licensed without RCT data. METHODS: We searched PubMed, the websites of UK health technology assessment bodies and the International Society for Pharmacoeconomics and Outcomes Research Scientific Presentations Database for assessments of treatments granted a marketing authorisation by the US Food and Drug Administration or European Medicines Agency from January 1999 to May 2014 without RCT data (74 indications). The outcome of interest was the approach to modelling efficacy data. RESULTS: Fifty-one unique models were identified in 29 peer-reviewed articles, 30 health technology appraisals, and 15 International Society for Pharmacoeconomics and Outcomes Research abstracts concerning 30 indications (44 indications had not been modelled). We noted the high rate of non-submission to health technology assessment agencies (28/98). The majority of models (43/51) were based on 'historical controls'-comparisons to previous meta-analysis or pooling of trials (5), individual trials (16), registries/case series (15), or expert opinion (7). Other approaches used the patient as their own control, performed threshold analysis, assumed time on treatment was added to overall survival, or performed cost-minimisation analysis. CONCLUSIONS: There is considerable variation in the quality and approach of models constructed for drugs granted a marketing authorisation without a RCT. The most common approach is of a naive comparison to historical data (using other trials/registry data as a control group), which has considerable scope for bias.
Authors: James E Signorovitch; Vanja Sikirica; M Haim Erder; Jipan Xie; Mei Lu; Paul S Hodgkins; Keith A Betts; Eric Q Wu Journal: Value Health Date: 2012 Sep-Oct Impact factor: 5.725
Authors: Tim A Kanters; Iris Hoogenboom-Plug; Maureen P M H Rutten-Van Mölken; W Ken Redekop; Ans T van der Ploeg; Leona Hakkaart Journal: Orphanet J Rare Dis Date: 2014-05-16 Impact factor: 4.123