MODELING THE COST-EFFECTIVENESS OF A HYPOTHETICAL FAMILY PLANNING INTERVENTION DURING A ZIKA OUTBREAK IN THE UNITED STATES

Purpose: Zika virus has spread rapidly in the Americas since 2015. Zika can cause microcephaly and fetal brain abnormalities. Improved access to effective contraception for women who do not intend to become pregnant can provide a medical countermeasure to reduce Zika-associated adverse pregnancy outcomes.

Objective: To assess from the societal perspective the cost-effectiveness of a family planning intervention that would promote highly-effective long-acting reversible contraception (LARC) to a population of women of reproductive age in areas that might experience a high incidence of Zika infection.

Method: A decision-tree simulation model to examine a hypothetical intervention to promote access to LARC, at no cost was constructed using a hypothetical cohort of 100,000 women ages 15-44 years who are fertile and do not intend to become pregnant. Effectiveness data are based on U.S. pilot projects that provided contraceptive methods at no cost to providers or participants along with contraception counseling that emphasized LARC. Outcome measures include intervention costs, numbers and costs of unintended pregnancies, numbers and cost of Zika-associated microcephaly (ZAM) cases prevented (including costs of stillbirth and terminations due to ZAM and lifetime cost among live-born infants), and cost-savings. The risk of ZAM reported during the 2013 Zika virus outbreak in French Polynesia (16 per 10,000 births) is used, and costs of care for ZAM are based on unpublished U.S. cost data for children with microcephaly associated with viral infection or other severe neurological impairments.

Result: The intervention using the hypothetical cohort of 100,000 women is projected to cost $23.3 million ($233/person) beyond the cost of existing services (no intervention group). Avoided costs are projected to total $58.8 million, with $53.2 million from avoided unintended pregnancy and $5.6 million from avoided ZAM, which is 2.5 times the additional intervention cost. Findings of net cost-savings are robust to parameter uncertainty in one-way, scenario, and probabilistic sensitivity analyses.

Conclusion: Increasing access to effective contraception for women who want to delay or avoid pregnancy during a Zika virus outbreak can result in overall reduction in the number of adverse pregnancy outcomes and lower total cost compared to no intervention. A limitation is that the numerical estimates are sensitive to uncertainty in model parameters, notably the frequency and lifetime incremental cost of ZAM; estimates may not be generalizable across settings.