Cost effectiveness of breast cancer screening and prevention: a systematic review with a focus on risk-adapted strategies

Abstract

Objectives Benefit and cost effectiveness of breast cancer screening are still matters of controversy. Risk-adapted strategies are proposed to improve its benefit-harm and cost-benefit relations. Our objective was to perform a systematic review on economic breast cancer models evaluating primary and secondary prevention strategies in the European health care setting, with specific focus on model results, model characteristics, and risk-adapted strategies. Methods Literature databases were systematically searched for economic breast cancer models evaluating the cost effectiveness of breast cancer screening and prevention strategies in the European health care context. Characteristics, methodological details and results of the identified studies are reported in evidence tables. Economic model outputs are standardized to achieve comparable costeffectiveness ratios. Results Thirty-two economic evaluations of breast cancer screening and seven evaluations of primary breast cancer prevention were included. Five screening studies and none of the prevention studies considered risk-adapted strategies. Studies differed in methodologic features. Only about half of the screening studies modeled overdiagnosis-related harms, most often indirectly and without reporting their magnitude. All models predict gains in life expectancy and/or quality-adjusted life expectancy at acceptable costs. However, risk-adapted screening was shown to be more effective and efficient than conventional screening. Conclusions Economic models suggest that breast cancer screening and prevention are cost effective in the European setting. All screening models predict gains in life expectancy, which has not yet been confirmed by trials. European modelsrisk-adapted screening strategies are rare, but suggest that risk-adapted screening is more effective and efficient than conventional screening. Keywords breast cancer screening; breast cancer prevention; cost effectiveness; decision analysis; risk stratification; overdiagnosisDutch MISCAN model. State-transition models were used by six studies [63,46,57,39,65,61] and mathematical models (e.g., equation-or regression-based models) were used in five studies [35,51,62,58,55]. The remaining studies used other types of models, including two decision trees [53,54], two life-table models [40,41],and two mixed models combining different model types [45,38].Thirteen models can be classified as population models considering the actual age 9 structure of the local target population. All but five models considered a lifetime-time horizon, appropriate to account for the long-term consequences of screening. More than half of the screening studies (18/32) and two thirds (14/21) of the screening studies, including strategies comparable to currently established breast cancer screening programs, performed a cost-utility analysis, which is the type of analysis required to account for all kinds of non-fatal health consequences, including most harms caused by overdiagnosis-and overtreatment. Sixteen studies, including 12 with...

UK Women’s Views of the Concepts of Personalised Breast Cancer Risk Assessment and Risk-Stratified Breast Screening: A Qualitative Interview Study

“…’Overall acceptors’ and women who thought ‘more is better’ appeared to be similar in age, but the latter group tended to have lower levels of educational attainment (see Table 1 ). As survey studies have demonstrated an association between lower levels of education and negative responses to reduced screening frequency for women at low-risk [ 5 , 13 ], it is possible that a lack of understanding may have influenced women’s negative attitudes towards low-risk screening options. Nevertheless, this will need to be confirmed by quantitative research with a larger and representative UK sample.…”

“…This has led to an increased interest in risk-stratified breast screening (RSBS), i.e., adapting screening recommendations to a woman’s breast cancer risk [ 4 ]. Simulation models suggest this approach has the potential to increase the benefits of screening (via early detection of cancer) while decreasing screening-related harms, e.g., false positive results and overdiagnosis [ 5 , 6 , 7 , 8 ]. However, the incorporation of RSBS within an existent screening programme will require major shifts in the way that screening is delivered and necessitate the involvement and agreement of all relevant stakeholders [ 9 , 10 ].…”

Feasibility and Acceptability of Personalized Breast Cancer Screening (DECIDO Study): A Single-Arm Proof-of-Concept Trial

“…Personalized breast cancer screening is characterized by risk-based recommendations for the ages of screening initiation and cessation, the screening frequency, the type of screening exam, consideration of preventive treatments for high-risk women, or even decisions regarding the non-screening of low-risk women. In modeling studies, personalized risk-based approaches to the early detection of breast cancer appear to be more efficient and have a better balance of benefits and harms than age-only strategies [ 1 , 2 , 3 , 4 , 5 ].…”

“…Although personalized screening will probably be the standard for early detection in the near future, the systematic reviews warn of the need for additional studies that assess risk estimation, acceptability, feasibility, and the legal and ethical aspects of personalized screening strategies [ 3 , 4 , 5 ]. To respond to these needs, several initiatives are underway.…”

An Assessment of the Predictive Performance of Current Machine Learning–Based Breast Cancer Risk Prediction Models: Systematic Review

“…The development and implementation of risk-based breast cancer control and prevention strategies can have great potential benefits and important public health implications. Moreover, risk-based breast cancer control and prevention strategy is more effective and efficient than conventional screening based on model evaluation [4,5]. A prerequisite for the implementation of personalized risk-adapted screening intervals is accurate breast cancer risk assessment [6].…”