The effect of a postpartum IUD intervention on counseling and choice: Evidence from a cluster-randomized stepped-wedge trial in Sri Lanka

Citation:

Karra M, Pearson E, Pradhan E, de Silva R, Samarasekera A, Canning D, Shah I, Weerasekera D, Senanayake H. The effect of a postpartum IUD intervention on counseling and choice: Evidence from a cluster-randomized stepped-wedge trial in Sri Lanka. Trials. 2019;20 (407) :1-15.

Abstract:

Background

The International Federation of Gynaecology and Obstetrics (FIGO), in collaboration with the Sri Lankan College of Obstetrics and Gynaecologists (SLCOG), launched an initiative in 2014 to institutionalize immediate postpartum IUD (PPIUD) services as a routine part of antenatal counseling and delivery room services in Sri Lanka. In this study, we evaluate the effect of the FIGO-SLCOG PPIUD intervention in six hospitals by means of a cluster-randomized stepped-wedge trial.

Methods/design

Six hospitals were randomized into two groups of three using matched pairs. Following a 3-month baseline period, the intervention was administered to the first group, while the second group received the intervention after 9 months of baseline data collection. We collected data from 39,084 women who delivered in these hospitals between September 2015 and January 2017. We conduct an intent-to-treat (ITT) analysis to determine the impact of the intervention on PPIUD counseling and choice of PPIUD, as measured by consent to receive a PPIUD, as well as PPIUD uptake (insertion following delivery). We also investigate how factors related to counseling, such as counseling timing and quality, are linked to choice of PPIUD.

Results

We find that the intervention increased rates of counseling, from an average counseling rate of 12% in all hospitals prior to the intervention to an average rate of 51% in all hospitals after the rollout of the intervention (0.307; 95% CI 0.148–0.465). In contrast, we find the impact of the intervention on choice of PPIUD to be less robust and mixed, with 4.1% of women choosing PPIUD prior to the intervention compared to 9.8% of women choosing PPIUD after the rollout of the intervention (0.027; 95% CI 0.000–0.054).

Conclusions

This study demonstrates that incorporating PPIUD services into postpartum care is feasible and potentially effective. Taking the evidence on both counseling and choice of PPIUD together, we find that the intervention had a generally positive impact on receipt of PPIUD counseling and, to a lesser degree, on choice of the PPIUD. Nevertheless, it is clear that the intervention’s effectiveness can be improved to be able to meet the demand for postpartum family planning of women.

Publisher's Version