Herpers M, Dintsios CM. Methodological problems in the method used by IQWiG within early benefit assessment of new pharmaceuticals in Germany.
THE EUROPEAN JOURNAL OF HEALTH ECONOMICS : HEPAC : HEALTH ECONOMICS IN PREVENTION AND CARE 2019;
20:45-57. [PMID:
29696458 DOI:
10.1007/s10198-018-0981-3]
[Citation(s) in RCA: 5] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 11/07/2017] [Accepted: 04/16/2018] [Indexed: 06/08/2023]
Abstract
BACKGROUND
The decision matrix applied by the Institute for Quality and Efficiency in Health Care (IQWiG) for the quantification of added benefit within the early benefit assessment of new pharmaceuticals in Germany with its nine fields is quite complex and could be simplified. Furthermore, the method used by IQWiG is subject to manifold criticism: (1) it is implicitly weighting endpoints differently in its assessments favoring overall survival and, thereby, drug interventions in fatal diseases, (2) it is assuming that two pivotal trials are available when assessing the dossiers submitted by the pharmaceutical manufacturers, leading to far-reaching implications with respect to the quantification of added benefit, and, (3) it is basing the evaluation primarily on dichotomous endpoints and consequently leading to an information loss of usable evidence.
OBJECTIVE
To investigate if criticism is justified and to propose methodological adaptations.
METHODS
Analysis of the available dossiers up to the end of 2016 using statistical tests and multinomial logistic regression and simulations.
RESULTS
It was shown that due to power losses, the method does not ensure that results are statistically valid and outcomes of the early benefit assessment may be compromised, though evidence on favoring overall survival remains unclear. Modifications, however, of the IQWiG method are possible to address the identified problems.
CONCLUSION
By converging with the approach of approval authorities for confirmatory endpoints, the decision matrix could be simplified and the analysis method could be improved, to put the results on a more valid statistical basis.
Collapse