Background:Cervical cancer is the most common cancer for Samoan women aged 15-44, and Samoa’s estimated incidence is the sixth highest out of 26 Asia-Pacific countries. Cervical cancer is easily prevented if pre-cancers are detected and treated early. Currently, Samoa has no preventive screening programme and the population is unvaccinated against Human Papillomavirus (HPV), the causative agent of cervical cancer. This paper investigates whethera point-of-care cervical cancer screening programme would be cost-effective for Samoa.
Methods:A Markov model based on the natural history of cervical cancer progression was built using TreeAge Pro softwareto perform a cost-utility analysis of eight screening scenarios relative to the status quo of ‘no screening’, based on evaluating their costs and health outcomes in terms of quality-adjusted life years (QALYs). The eight scenarios are compared by calculating their incremental cost-effectiveness ratios (ICERs) in terms of cost per QALY gained – i.e. where the higher the cost per QALY gained, the less cost-effective the screening scenario. These ICERs are contextualised with respect to their ‘value for money’ in relation to willingness-to-pay thresholds for Samoa. The model is calibrated using Samoan 2012 ICO cervical cancer data and sensitivity analyses are performed.
Findings:Under the assumption that 50% of eligible woman participate in the screening programme, all scenarios are cost-effective. The most effective screening scenario is 5-yearly HPV testing for women 30-59 years, which is predicted to reduce cervical cancer incidence by 41%. Incidence reduces further with higher participation rates, and under this scenario an additional 12 QALYs per 1000 women are gained. Changing the Markov model’s parameters via one-way and Monte Carlo probabilistic sensitivity analyses does not affect the cost-effectiveness results of the eight screening scenarios.
Conclusions:Based on our model, a point-of-care cervical screening programme for Samoa does not affect the relative cost-effectiveness results of the eight screening scenarios.