Analytical approach:
The authors developed a Markov model to synthesise epidemiological and cost data collected from the NHS Sentinel Sites (Bristol, Norwich, Liverpool, Manchester, Northwick Park, and Sheffield) with a range of estimates from studies in the published literature. The time horizon was 10 years. The authors stated that a health services perspective was adopted.
Effectiveness data:
The effectiveness evidence came from a range of sources including observational data from the NHS Sentinel Sites Study (available in an online appendix), a published systematic review, a review conducted by the authors, a meta-analysis, and a selection of known recent relevant studies. The main clinical estimates were test accuracy and compliance, treatment success, and risk of recurrent disease (which culminated in cases of cervical intraepithelial neoplasia 3+ averted).
Monetary benefit and utility valuations:
Not relevant.
Measure of benefit:
The primary measure of benefit was the cost per case of cervical intraepithelial neoplasia 3+ averted.
Cost data:
The costs included: the direct costs of consumables and capital equipment for human papillomavirus testing; staff time; consultation; treatment; and costs per event of cancer. The sources of resource use and prices were the NHS Sentinel Study Sites, manufacturer’s prices, and estimates from the published literature. The price year was 2009. Future costs were discounted at a rate of 3.5%. Costs were adjusted to 2009 prices using the Hospital and Community Health Services Index.
Analysis of uncertainty:
One-way and partial multi-way sensitivity analyses were conducted. The results were presented in graphs and fully described in the text.