OBJECTIVES

To compare the cost-effectiveness of a cervical cancer screening algorithm using human papilloma virus (HPV) primary testing to screening with cytology and HPV co-testing.

METHODS

A Markov decision model was developed for the disease etiology of human papilloma virus (HPV) infection and the development of invasive cervical cancer. A United States payer perspective was adopted. A theoretical cohort of 1,000 women, with a starting age of 30 years, was considered for a screening period of 35 years. The CMS 2017 Medicare Physician Fee Schedule was used to determine reimbursement for outpatient screening procedures. Published literature values were used for the costs of treating cervical cancer and health utilities. A 3% discount rate was applied. An incremental cost effectiveness ratio (ICER) based on cost per quality adjusted life year was calculated using a willingness to pay threshold of US$50,000. Tornado and threshold analyses were conducted.

RESULTS

The average per patient costs for the HPV primary testing algorithm was US$1,051 and the average per patient costs for the cytology/HPV co-testing algorithm was US$1,787. The incremental cost-effectiveness ratio was -$22,577 for HPV primary testing compared to cytology/HPV co-testing. Tornado analysis showed that costs of co-testing and office visits have the largest impact on the model. Threshold analysis showed the base cost for HPV testing would need to exceed US$299 and office visits would need to increase to US$169 before co-testing was non-dominated. There was no starting age from 0-65 in which co-testing was not dominated.

CONCLUSIONS

Compared with cytology/HPV co-testing, HPV primary testing was cost-effective. The ICER represented a favorable treatment option that is both less costly and more effective than co-testing. The overall results confirmed the results of previously published literature.