Budget Impact Analysis of Cancer Screening: A Methodological Review

Background: Budget impact analyses (BIAs) describe changes in intervention- and disease-related costs of new technologies. Evidence on the quality of BIAs for cancer screening is lacking. Objectives: We systematically reviewed the literature and methods to assess how closely BIA guidelines are followed when BIAs are performed for cancer-screening programs. Data sources: Systematic searches were conducted in MEDLINE, EMBASE, EconLit, CRD (Centre for Reviews and Dissemination, University of York), and CEA registry of the Tufts Medical Center. Study eligibility criteria: Eligible studies were BIAs evaluating cancer-screening programs published in English, 2010–2018. Synthesis methods: Standardized evidence tables were generated to extract and compare study characteristics outlined by the ISPOR BIA Task Force. Results: Nineteen studies were identified evaluating screening for breast (5), colorectal (6), cervical (3), lung (1), prostate (3), and skin (1) cancers. Model designs included decision-analytic models (13) and simple cost calculators (6). From all studies, only 53% reported costs for a minimum of 3 years, 58% compared to a mix of screening options, 42% reported model validation, and 37% reported uncertainty analysis for participation rates. The quality of studies appeared to be independent of cancer site. Limitations: “Gray” literature was not searched, misinterpretation is possible due to limited information in publications, and focus was on international methodological guidelines rather than regional guidelines. Conclusions: Our review highlights considerable variability in the extent to which BIAs evaluating cancer-screening programs followed recommended guidelines. The annual budget impact at least over the next 3–5 years should be estimated. Validation and uncertainty analysis should always be conducted. Continued dissemination efforts of existing best-practice guidelines are necessary to ensure high-quality analyses.