Breast cancer screening and prevention cost effective in Europe

Abstract

Breast cancer screening and prevention strategies appear to be cost effective in Europe, according to findings of a systematic review published in the European Journal of Health Economics. PubMed, Ovid Medline, Embase, EconLit, Cochrane Library and CRD database were searched up to April 2018 for economic breast cancer models assessing the cost effectiveness of breast cancer screening and prevention strategies in Europe. Primary and secondary prevention strategies were evaluated with a focus on risk-adapted strategies. In total, 32 economic evaluations of breast cancer screening and seven evaluations of primary breast cancer prevention met the selection criteria; five of the screening studies and none of the prevention studies considered riskadapted strategies. Most of the screening studies were performed in the UK (n=8), Spain (7) or the Netherlands (6). Evaluations used discrete event simulation (n=15), state-transition models (6), mathematical models (5) and other modeling types. Most models used a lifetime time horizon. Approximately half of the screening studies modelled harms due to overdiagnosis of breast cancer, but in most of these studies the harms were modelled indirectly and the magnitude was not reported. Only 25 of the screening studies reported incremental cost-effectiveness ratios (ICERs). All of these models predicted gains in life-years and/or QALYs at costs which met willingness-to-pay thresholds; in most estimates, ICERs were well below €30 000* per life-year or QALY gained. Risk-adapted screening was more efficient than conventional screening. It can be concluded that European economic models almost unanimously suggest that breast cancer screening and primary prevention are cost effective in the European setting, even in more recent studies when overdiagnosis-related harms are accounted for more explicitly. However, it also is shown that all models assume that reductions in breast cancer mortality translate into gains in life-expectancy, which has not been convincingly shown in trials yet, said the authors.