Purpose: Immediate post-abortal intra-uterine device (IUD) insertion has been shown to be safe and effective yet it is not commonly available. The rate of IUD expulsion after immediate insertion has been shown not to be substantially higher than with delayed insertion.

Methods: We performed a decision analysis to examine pregnancy rates at one year following abortion with immediate or delayed IUD insertion. The base case assumed an 80% 1-year IUD continuation rate for both groups, an additional 10% expulsion risk with immediate IUD insertion, a 30% risk for not returning for delayed insertion, and pregnancy rates of 8% without an IUD and 0.5% with an IUD. Pregnancy between abortion and delayed IUD insertion was not considered. One-way, two-way, and Monte Carlo sensitivity analyses were performed.

Results: In the decision model, the 12-month pregnancy rate was 21 per 1000 women in the immediate-insertion group and 35 per 1000 women in the delayed-insertion group. Sensitivity analyses show the model to be dependent on the expulsion rate in the immediate-insertion group (varied up to 30%) and the rate of not returning in the delayed-insertion group. In one-way analysis, the overall pregnancy rate was higher in the delayed-insertion group as long as the pregnancy rate in the women without IUDs was above 0.5%. In two-way analysis, immediate insertion was preferred if the rate of IUD expulsion was lower than the rate of not returning for delayed IUD insertion. Monte Carlo analysis showed that immediate insertion results in fewer pregnancies 99.4% of scenarios, with an absolute mean difference of 16 pregnancies per 1000 women.

Discussion: Despite assumptions in the model favoring delayed insertion, immediate post-abortal IUD insertion results in 14 to 16 fewer pregnancies per 1000 women than delayed insertion in the following 12 months. Future research should examine cost-effectiveness and non-pregnancy outcomes.