Search publications
Title: Long-term evaluation of benefits, harms, and cost-effectiveness of the National Bowel Cancer Screening Program in Australia: a modelling study
Authors: Lew JB; St John DJB; Xu XM; Greuter MJE; Caruana M; Cenin DR; He E; Saville M; Grogan P; Coupé VMH; Canfell K
Categories: Cancer Type - Bowel & Colorectal Cancer
Early Detection, Diagnosis, and Prognosis - Technology and/or Marker Testing in a Clinical Setting
Year: 2017
Journal Title: The Lancet Public Health
Volume: 2
Issue: 7
Page number start: 331
Page number end: 340
Abstract: BACKGROUND: No assessment of the National Bowel Screening Program (NBCSP) in Australia, which considers all downstream benefits, costs, and harms, has been done. We aimed to use a comprehensive natural history model and the most recent information about cancer treatment costs to estimate long-term benefits, costs, and harms of the NBCSP (2 yearly immunochemical faecal occult blood testing screening at age 50-74 years) and evaluate the incremental effect of improved screening participation under different scenarios. METHODS: In this modelling study, a microsimulation model, Policy1-Bowel, which simulates the development of colorectal cancer via both the conventional adenoma-carcinoma and serrated pathways was used to simulate the NBCSP in 2006-40, taking into account the gradual rollout of NBCSP in 2006-20. The base-case scenario assumed 40% screening participation (currently observed behaviour) and two alternative scenarios assuming 50% and 60% participation by 2020 were modelled. Aggregate year-by-year screening, diagnosis, treatment and surveillance-related costs, resource utilisation (number of screening tests and colonoscopies), and health outcomes (incident colorectal cancer cases and colorectal cancer deaths) were estimated, as was the cost-effectiveness of the NBCSP. FINDINGS: With current levels of participation (40%), the NBCSP is expected to prevent 92 200 cancer cases and 59 000 deaths over the period 2015-40; an additional 24 300 and 37 300 cases and 16 800 and 24 800 deaths would be prevented if participation was increased to 50% and 60%, respectively. In 2020, an estimated 101 000 programme-related colonoscopies will be done, associated with about 270 adverse events; an additional 32 500 and 49 800 colonoscopies and 88 and 134 adverse events would occur if participation was increased to 50% and 60%, respectively. The overall number needed to screen (NNS) is 647-788 per death prevented, with 52-59 colonoscopies per death prevented. The programme is cost-effective due to the cancer treatment costs averted (cost-effectiveness ratio compared with no screening at current participation, AUS$3014 [95% uncertainty interval 1807-5583] per life-year saved) in the cost-effectiveness analysis. In the budget impact analysis, reduced annual expenditure on colorectal cancer control is expected by 2030, with expenditure reduced by a cumulative AUS$1·7 billion, AUS$2·0 billion, and AUS$2·1 billion (2015 prices) between 2030 and 2040, at participation rates of 40%, 50%, and 60%, respectively. INTERPRETATION: The NBCSP has potential to save 83 800 lives over the period 2015-40 if coverage rates can be increased to 60%. By contrast, the associated harms, although an important consideration, are at a smaller magnitude at the population level. The programme is highly cost-effective and within a decade of full roll-out, there will be reduced annual health systems expenditure on colorectal cancer control due to the impact of screening.
Division: Cancer Research Division
Funding Body: Australia Postgraduate Award PhD Scholarship, Translational Cancer Research Network Top-up scholarship (supported by Cancer Institute NSW) and Cancer Council NSW. KC was supported by a NHMRC Fellowship (GNT1082989)
DOI: doi: 10.1016/S2468-2667(17)30105-6
URI: http://researchpubs.cancercouncil.com.au/cancercounciljspui/handle/1/1944
Appears in Collections:Research Articles

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.