Objectives
We aimed to determine observed vs. expected pregnancy risk among levonorgestrel 52 mg and copper T380A intrauterine device (IUD) emergency contraception (EC) users.
Methods
This is a secondary analysis of participants in a 1:1 randomized controlled trial comparing levonorgestrel 52 mg IUD and copper T380A IUD EC users who reported all episodes of unprotected intercourse in the two weeks prior to enrollment. We estimated number of expected pregnancies via two established approaches (Trussell, 2003 and modified Wilcox, 2015) to create a range of expected pregnancy risk among methods using risk estimates per menstrual cycle day. With both approaches, we calculated the proportion of pregnancies prevented based on those expected using both the day of most recently reported unprotected intercourse and all reported current menstrual cycle unprotected intercourse.
Results
One pregnancy occurred among the 327 participants assigned to the levonorgestrel IUD with 137 (41.9%) reported multiple episodes of unprotected intercourse. No pregnancies occurred among the 328 participants randomized to the copper IUD with 142 (43.2%) reporting multiple episodes of unprotected intercourse. In the levonorgestrel IUD group, expected pregnancies ranged from 12.4 (Trussell)-14.8 (Wilcox) using only the most recent episode of unprotected intercourse to 22.0 (T)-26.3 (W) accounting for all episodes. In the copper IUD group, there were 12.6 (T)-15.0 (W) and 24.8 (T)-29.6 (T) expected pregnancies respectively. Levonorgestrel IUD users prevented 92.0 (T)-96.2 (W)% of pregnancies and copper IUD users experienced 100% pregnancy prevention.
Conclusions
Levonorgestrel IUDs for EC prevented 92% of expected pregnancies, approaching the 100% prevention rate observed in the copper IUD group.