Impact of Integrating Family Planning within a Community-Based Maternal and Neonatal Health Program in Rural Bangladesh Salahuddin Ahmed1 & 2, Jaime Mungia2, Catharine McKaig2, Saifuddin Ahmed1, Amnesty LeFevre1, Peter Winch1, Ahmed Al-Kabir3, and Abdullah Baqui1 1Johns Hopkins School of Public Health; 2 Jhpiego; 3Shimantik

Study Context TFR by Divisions, Bangladesh, 2004 Selected FP indicators in Sylhet, BDHS 2007 Indicators Sylhet Unmet FP need 26% CPR (any method) 31% TFR 3.7 Birth intervals <24 months <36 months 57% 2.6 4.2 2.9 2.8 2.9 3.7

Integrated model of PPFP & MNH Evolution of MNCH packages Designed a community-based maternal and newborn care intervention package and evaluated the effectiveness of the package using a cluster randomized design A home care package which involved CHW antenatal and postnatal home visits and management of sick newborn reduced NMR by 34% (Baqui et al., Lancet, 2008) Postpartum FP counseling and contraceptive distribution Newborn care

Study Objectives To develop and test an integrated Family Planning, Maternal and Neonatal Health (FP/MNH) service delivery approach To assess the strengths and limitations of integrating FP into an ongoing community-based MNH care program To assess the impact of the intervention package on contraceptive knowledge and practices including LAM during the extended postpartum period To assess the impact of the intervention package on pregnancy spacing

Enrollment of women during <8 months of pregnancy Study Design Study sites: eight unions in two sub-districts in Sylhet district, Bangladesh Non-Random Allocation Intervention unions: four Enrolled women: 2247 Comparison unions: four Enrolled women: 2257 Enrollment of women during <8 months of pregnancy Intervention clusters: MNH plus FP during ANC and Postpartum visit Comparison clusters: MNH ONLY during ANC and Postpartum visit Follow the cohort through pregnancy to 36 months postpartum

Intervention Delivery Strategy Service Delivery: Home visits by CHWs Conduct pregnancy surveillance and provide contraceptives Counsel in antenatal and postpartum periods Conduct visits to all households every two months to identify new MWRA and pregnant women Messages on LAM and transition, return to fertility, healthy spacing, contraceptive methods Pills, condoms, and follow up doses of injectables Refer for other methods Community mobilization: Conduct meetings with husbands, mothers-in-law and community leaders to raise awareness about PPFP messages

Results

Starts in a Low Performance Area Ever Used Contraceptive Method Intervention 18.0 At baseline, it is noteworthy that we started in a geographic area where there were subtle differences between the intervention and comparison study arms in ever use of contraception. In the intervention area only 18% of women reported having ever used a contraceptive as compared to 21% in the comparison area. Despite these differences in baseline ever use of contraceptives, the intervention and comparison areas maintained comparable rates for other major demographic and socioeconomic characteristics. Control 21.1 10 20 30 40 50 60 70 80 90 100 Percent

Contraceptive Use Rate at 3, 6, 12, and 18 Months Postpartum by Study Arm As we look at the impact of postpartum family planning activities on the uptake of contraceptives over time, It is noteworthly here that we see a number of promising findings. Most notably, we see a statistically significant improvement in the intervention area over time from 36% at 3 months to 47% at 18 months postpartum. We also see a high number of new users and a trend towards increased early adoption

Contraceptive Use Rate at 3, 6, 12, and 18 Months Postpartum by Study Arm Statistically significant improvement in the contraceptive use rate in the intervention area during the high risk period of first 18 months after delivery -- 18% ever user before the index pregnancy to 47% at 18 months postpartum -- 21% ever user before the index pregnancy to 34% at 18 months postpartum High number of new users and a trend towards increased early adoption Ever use or previous use of contraceptive methods is a major factor impacting postpartum use of contraception. At baseline, it is noteworthy that we started in a geographic area where there were subtle differences between the intervention and comparison study arms in ever use of contraception. In the intervention area only 18% of women reported having ever used a contraceptive as compared to 21% in the comparison area. Despite these differences in baseline ever use of contraceptives, the intervention and comparison areas maintained comparable rates for other major demographic and socioeconomic characteristics. As we look at the impact of postpartum family planning activities on the uptake of contraceptives over time, It is noteworthly here that we see a number of promising findings. Most notably, we see a statistically significant improvement in the intervention area over time from 36% at 3 months to 47% at 18 months postpartum. We also see a high number of new users and a trend towards increased early adoption

Probability of Contraceptive Adoption by Study Arm The above slide illustrates the probability of contraceptive adoption in the 18 months following delivery. The 46% of contreceptive use – or the contraceptive prevalence rate – accounts for individuals who have discontinued use and thus masks an underlying higher trend indicating that a larger proportion of individuals adopted contraceptives at some point during the 18 months following delivery. If we look more broadly at contraceptive use in the 18 months following delivery, we see a significant difference in the probability of contraceptive adoption between the intervention and control arms. In the intervention arm – reflected here by the top blue curve – the probability of contraceptive adoption increases from slightly more than 25% to ~ 70% at 18 months postpartum. In the comparison area – depicted by the red curve-- the probability of contraceptive adoption was around 10% immediately following delivery and increased to less than 50%. Overall this figure demonstrates that project activities were associated with a significant increase in contraceptive adoption immediately following delivery and in the ensuing 18 months. ________________ Saifuddin: The probability curve shows "cumulative %" in one sense, say 70% at point of time adopted a contraceptive method, but CPR is lower because of discontinuation.

Self Reported Pregnancy Incidence As we consider the larger implications of the increased early and sustained adoption of contraceptive use that we’re seeing to date, it’s worth exploring what the impact is on Self reported pregnancy incidence. The above slide looks at the probability of pregnancy incidence during the 18 period following delivery. Findings illustrate that in the comparison area, the probability of pregnancy is significantly higher over time. If we look at the time period of 6 to 18 months we can seeing a growing differentiation between the two study areas as the probability of pregnancy incidence increases from about 3% at 9 months to 13% at 18 months in the comparison area and from 2% to slightly less than 10% over the same period in the intervention area. As the project continues and follow up data from 24, 30 and 36 months emerges, we anticipate that this trend of a higher probability of self reported pregnancy will become more pronounced.

Does integration of FP adversely affect MNH program?

CHW Home Visit Coverage   Intervention (n=1889) Comparison (n=1838) Pregnancy counseling- 1 92% 97% Pregnancy counseling- 2* 86% 90% Newborn-1 counseling 74% 80% Newborn-3 counseling 72% 76% Newborn-6 counseling* 73% 75% Newborn-9 counseling 77% Newborn-15 counseling 51% 50% One critical way to explore this issue of integration lies in looking at what impact the addition of new services might have upon the coverage of existing care. Recall that the HFS project was integrated ONTO a platform of MNCH services. It’s very promising to note that as we look at the coverage rates between the intervention and comparison arms for these critical pregnancy and newborn counseling visits carried out by the CHW in the home, that we don’t see major differences. In short, the addition of family planning doesn’t seem to negatively impact coverage for existing MNCH services. *Integrated counseling

Effect of Integration on MNH Care: Selected Newborn Care practices by Study Arm Intervention (%) Comparison (%) Drying and wrapping of newborn within 10 minutes of delivery 50.4 44.1 Initiation of Breastfeeding within 30 minutes 56.6 46.8 Simiarly if we look at immediate care for the newborn following birth, a greater number of users’ in the intervention area report “Drying and Wrapping the newbron within 5 minutes” – a action critical for hypothermia prevention– and as well immediately initiating breastfeeding (10% greater in the intervention area).

Duration of Exclusive Breastfeeding by Study Arm This slide re- enforces my last point and shows a promising trend in increasing the duration of exclusive breastfeeding by study arm. It is noteworthy that our comparison area here is again an intensive package of MNCH activities which too emphasizes exclusive breastfeeding. The added benefit of PPFP in conducting additional visits up to 4/5 months postpartum had a clear significant impact on exclusive breastfeeding. This trend remains statistically significant when we additional stratify by LAM and Non-LAM users.

Lessons Learned HFS demonstrates The feasibility of integration of FP within a community-based MNH program. The effectiveness of the model in increasing modern method use. No notable negative effect on the delivery of MNH services. The promotion of LAM had a positive effect on the duration of exclusive breastfeeding.

Study Partners Funded by USAID JHSPH ACCESS-FP /MCHIP Government of Bangladesh Shimantik and CDPA

THANK YOU