Several recent studies have highlighted the need for greater use of long-acting contraception. The most influential of these studies is the Contraceptive CHOICE Project, which was credited with substantially reducing participants’ pregnancy risk by increasing their use of long-acting methods such as intrauterine devices and subdermal implants. However, because participants’ rates of nonuse and condom use fell to zero at the outset of the intervention, it is possible that sizable pregnancy reductions could still have been achieved if enrollees had chosen shorter-acting, female-controlled methods such as oral contraception.

The objective of the study was to estimate the proportion of the CHOICE Project’s fertility impacts that could have been achieved without any increase in long-acting method use.

The FamilyScape 3.0 microsimulation model was used to estimate CHOICE’s impact on pregnancy risk and to simulate the counterfactual effect of moving all nonusers and condom users onto shorter-acting, female-controlled methods. FamilyScape models the sexual and contraceptive behaviors of women in the United States between 2006 and 2010, which is the period when CHOICE was implemented.

Nearly three quarters of the CHOICE intervention’s effects on pregnancy risk could have been achieved if participants had chosen shorter-acting, female-controlled methods over long-acting methods.

Prioritizing the adoption of long-acting contraception may not be the most advisable strategy for reducing unintended pregnancy. The most impactful interventions will likely be those that increase the use of female-controlled methods, long-acting or otherwise.