Cost-Effectiveness of Immediate Compared With Delayed Postpartum Etonogestrel Implant Insertion

Abstract

Objective: To evaluate the cost-effectiveness of immediate compared with delayed (6 weeks) postpartum etonogestrel implant insertion in preventing future unintended pregnancy.

Methods: We constructed a decision-analytic model to examine a hypothetical population of women who request a contraceptive implant after giving birth. The timeframe for analysis was from the time of childbirth to 1 year postpartum. Model inputs were derived from a comprehensive literature review. We compared immediate (before discharge from the childbirth hospital stay) compared with delayed (at first postpartum office visit) postpartum placement of the contraceptive implant from a health care system's perspective. Implant insertion and removal, loss to follow-up at the postpartum visit, use of alternative contraceptive methods, and contraceptive failure were incorporated into the model. We calculated the incremental cost of immediate insertion for each pregnancy prevented during the first postpartum year and cost savings associated with pregnancies prevented. One-way sensitivity analyses were also performed. Cost estimates are reported in 2014 U.S. dollars.

Results: Immediate postpartum implant insertion is associated with higher expected cost than delayed insertion ($1,091/patient compared with $650/patient) but is more effective in preventing pregnancies (expected pregnancy rate: 2.4% and 21.6%, respectively). This results in an incremental cost-effectiveness ratio of $2,304 per pregnancy prevented. When taking into consideration medical costs of the resulting unintended pregnancies that could be avoided, immediate implant insertion is expected to save $1,263 per patient.

Conclusion: Immediate postpartum provision of the contraceptive implant is cost-effective in preventing unintended pregnancies and should be provided to women requesting this form of contraception.

Figures

Figure 1

Model of immediate versus delayed implant insertion. The collapsed Markov subtree in the figure models repeated menstrual cycles resulting in pregnancy or non-pregnancy states within each cycle while accounting for women’s use of contraceptives and status of exclusive breastfeeding.

Figure 2

Tornado diagram reflecting results from one-way sensitivity analysis. This figure only reports the top 15 parameters with the most impact on the incremental cost-effectiveness ratio of immediate versus delayed postpartum insertion of implant. *Costs in US dollars.