Modelling future capacity needs and spending on colonoscopy in the English bowel cancer screening programme
- Dr K E Nnoaham, Public Health Resource Unit, 4150 Chancellor Court, Oxford Business Park South, Oxford OX4 2GX, UK;
- Revised 14 April 2008
- Accepted 17 April 2008
- Published Online First 25 April 2008
Background: Bowel cancer screening using faecal occult blood testing and colonoscopy is currently being rolled out across England. Guidelines recommend that people identified by colonoscopy as having intermediate- or high-risk bowel polyps be offered periodic surveillance colonoscopy because of their elevated risk of bowel cancer. We make projections of the likely year-on-year increase in volumes and spending on colonoscopy due to the screening and surveillance programmes.
Methods: We constructed a model based on current bowel cancer screening and surveillance guidelines using screening outcome measures taken from the second round of the English bowel screening pilot. This was then used to predict colonoscopy volumes and cost for a hypothetical population.
Results: For a hypothetical population of 500 000 people, with average deprivation and 66 956 subjects aged 60–74 years, the initial screening and surveillance round would be expected to detect 34 cancers at a cost of £394 157. In the first 8 years, colonoscopy numbers will grow at a rate of 23 per year, most of which will be surveillance colonoscopies. Colonoscopy costs may grow by £11 808 yearly in the same period, representing a cost per eligible person of £2.86 initially, increasing by £0.13 every year. Sensitivity analyses suggest significant changes in these predictions if screening uptake changes by 20%.
Conclusion: The model has been used to make projections for five primary care trusts within the South Central Strategic Health Authority. Results from the volume and cost projections can inform service planning and resource allocation at local levels for the implementation of the current and future bowel cancer screening programme.
Competing interests: None.