Impact of contraceptive counselling training among counsellors participating in the FIGO postpartum intrauterine device initiative in Bangladesh

To evaluate the impact of structured training given to dedicated family planning counsellors on postpartum intrauterine device (PPIUD) services across six tertiary hospitals in Bangladesh.


| INTRODUCTION
Birth spacing is an effective way to reduce maternal, perinatal, and child morbidity and mortality, and WHO recommends a period of at least 24 months between delivery and conception to reduce the adverse risks of pregnancy 1 Encouraging use of effective contraception can ensure sufficient spacing between pregnancies, as well as avoid instances of unsafe abortion-another contributor to maternal morbidity and mortality. In Bangladesh, the government has committed to ending preventable child and maternal deaths by 2030; however, despite various family planning initiatives, unmet need for contraception remains high with unintended pregnancies accounting for approximately one-third of all pregnancies. 2 Women who use traditional or temporary contraceptive methods are more likely to experience unintended pregnancies than those using long-acting contraceptive methods. In addition to availability of reliable methods, unmet need for contraception is also dependent on behaviors linked to adherence to the method. In Bangladesh, approximately 36% of married women aged between 15 and 49 years discontinue contraception within one year. 3 Figure 1 shows the trends in unintended pregnancy, contraceptive prevalence rate (CPR), and method discontinuation in Bangladesh from 1993-2014; however, it does not indicate the proportion of women switching methods or discontinuing use in order to conceive.
In the 2014 Bangladesh Demographic and Health Survey (BDHS) it was reported that 12% of married women aged 15-49 years in Bangladesh had an unmet need for contraception, with 54% of married women using a modern method of contraception. 4 Of these women, 33% reported discontinuation of the method after a year, and just 8% of married women used either a long-acting reversible contraceptive or a permanent method. These factors are major contributors to the stagnating total fertility rate (TFR) in Bangladesh between 2004 and 2014. 4 Although Figure 1 demonstrates some improvements from 2007, with a drop in discontinuation rates and modest improvements in unintended pregnancies and CPRs, it is evident that to achieve Bangladesh's target of 1.7 TFR by 2021, unintended pregnancies and unmet need for contraception need further urgent attention.
The immediate postpartum period is an ideal time to provide contraception in settings where women are unlikely to return to a health facility following delivery. Furthermore, the postpartum period is a particularly vulnerable time for women should they become pregnant. Nevertheless, in Bangladesh, contraceptive services have not been a priority for service providers in hospitals during maternity care, and it is rarely highlighted in prenatal check-ups. The only method that is advocated in hospitals is bilateral tubal ligation among multiparous patients undergoing cesarean delivery both in the public and private sector.
One of the most effective methods of contraception that can be provided during the immediate postpartum period is the copper intrauterine device (IUD), a nonhormonal long-acting reversible method.
The IUD can be inserted in the immediate postpartum period and up to 48 hours following delivery of the placenta (after which insertion is not recommended until 4-6 weeks after delivery); it is effective for up to 10 years but can be removed at any point. In Bangladesh, use of IUDs (both interval and postpartum) is limited to just 0.6% of married women aged 15-49 years. 4 Women and their families in Bangladesh lack accurate knowledge about the IUD, and a combination of myths regarding IUDs as well as religious beliefs deter women from using postpartum IUDs (PPIUDs).
WHO has introduced counselling interventions as a key element in family planning care to improve contraceptive use and compliance to prevent unintended pregnancy. 5 Therefore, to increase demand for PPIUD in Bangladesh, sensitization and counselling of women and their families in the prenatal, intrapartum, and postpartum periods are essential. Healthcare providers should be well equipped to offer accurate and balanced counselling to women and their families. This will ensure that they receive sufficient information about contraceptive methods regarding effectiveness, correct use, common adverse effects, health risks and benefits, and signs and symptoms that would necessitate return to the clinic. [6][7][8] To address unmet need for contraception as well as reduce mater- Since January 2015 the initiative has been implemented in six tertiary level teaching hospitals across four major cities in Bangladesh: Dhaka, Sylhet, Chittagong, and Khulna. Table 1 shows the unmet need in these divisions, and the participating hospitals with delivery and interview rates. With high delivery rates there was a concern that doctors and nurses would have limited time capacity to provide effective PPFP counselling to women. Therefore, dedicated lay female counsellors were recruited to provide additional in-depth counselling in PPFP. Although there were initial concerns about introduction of family planning counsellors (never previously employed in Bangladesh) and sustainability aspects, the strategy had been described in other low-resource countries, 9 which assisted with its implementation in Bangladesh.
Once recruited, counsellors received training on the different con- so that a higher standard of PPFP counselling services could be offered. Therefore, after 18 months of implementation a formal fourday refresher training course was organized for the 28 lay counsellors.
The aim of the present study was to assess the impact of the indepth refresher counselling course on the performance of 28 lay counsellors in family planning. The measures are indirect and asses PPFP counselling rates, PPIUD consent rates, insertion rates, and removal rates at six weeks postpartum across the six facilities.

| MATERIALS AND METHODS
A total of 28 counsellors received refresher training over the fourday course (July 17-20, 2017). The training was delivered by an experienced public health specialist from a partner organization in India, who had developed the course and successfully trained their own local family planning counsellors. Given the similarity of contexts, this was felt to be highly appropriate. Training comprised a comprehensive and structured course including counselling methodology as well as factual information on contraception and PPIUD. It All women delivering in the hospitals who lived in Bangladesh were eligible for inclusion, and all those approached were asked for their consent to participate in the data collection and given the opportunity to refuse or retract at any time. In Bangladesh, studies concerned with data collection for project monitoring are exempt from institutional approval. Table 1 shows the interview rates for the selected time period for each facility. Women were interviewed using a structured questionnaire specifically developed for the initiative, which asked questions regarding their counselling experiences, consent for PPIUD, and PPIUD insertion. These data were collected following delivery before the women were discharged from hospital. Follow-up information regarding satisfaction with counselling services, continuation of the method if one had been chosen, and complications was collected by DCOs either in person when the women returned for a postnatal check-up at 4-6 weeks, or over the phone. The date of delivery was used to assign women to either the before or after training group.
Where delivery date was unavailable, the date of the interview was used as a proxy.
The primary statistical analysis was to investigate if there was a change in the proportion of women consenting to PPIUD (among those who had been counselled) and in the proportion of women who had one inserted (among those who had consented). Initial analysis compared proportions without adjustment for other factors, using either a chi-squared test or Fisher's exact test when numbers were low. Further analysis was performed using logistic regression in Stata version 14.2 (StataCorp, College Station, TX, USA). The effect of training is reported as an odds ratio (OR) or adjusted odds ratio (aOR) with a 95% confidence interval (CI).
As secondary analyses, we compared the proportion of women who were counselled and the proportion who had their PPIUD removed at follow-up 4-6 weeks after delivery, as well as women's self-reported satisfaction with the service. We also conducted sensitivity analyses T A B L E 1 Unmet need, delivery, and interview rates in participating hospitals in Bangladesh.

| RESULTS
The tests conducted before and after the refresher counselling  (Table 2).
At 4-6-week follow-up, 1050 women were interviewed (39.9% [n=2633] of those with a PPIUD inserted), by either telephone or face-to-face interview. There was a higher follow-up rate in the before training period (56.9% vs 27.6%). Table 3 shows the outcomes in relation to PPIUD consent and insertion. After training there were fewer removals and an increase in the proportion of women who reported being "very satisfied" with counselling (from 9.9% to 13.6%).
There was a slight decrease in both primary outcomes between the before training and after training time periods (

| DISCUSSION
It is important to point out some limitations of this analysis. As the data used for analysis was regular monitoring data, it was not designed explicitly for this study. Therefore, certain variables that T A B L E 3 Effect of training on key outcome measures in relation to PPIUD consent and insertion before and after training time periods. may have shown more of the effects of the training were not captured. This included timing of counselling (prenatal, intrapartum, postpartum), number of counselling episodes, or whether the women left using any modern method of contraception other than just PPIUD. Another limitation is that the follow-up rate of women who had a PPIUD inserted was approximately 40%. As no incentives were given, this reflects the reality of postpartum follow-up, and in a way exemplifies why immediate PPFP is so valuable in this context. A one-stop approach means a woman's single visit to hospital can result in maximum benefit. The results analyzed could not therefore include the remaining 60% of women with a PPIUD who did not return for follow-up.
As well as counselling, there are other factors that would have influenced the outcomes that were analyzed. Access to service provision played a role in this study as many women who were counselled and consented during prenatal check-ups in the participating facilities did not go on to deliver at these facilities, and therefore data on these women are not available regarding their counselling experiences and their decisions with regard to PPIUD. Furthermore, the translation of insertions for those women who consented to have a PPIUD was not dependent on counsellors but rather on the availability of trained providers to perform the insertions once consent had been given, contraindications at delivery, or withdrawal of consent at the time of insertion. This explains why only 88.9% of women who consented went on to actually have a PPIUD inserted.
This was therefore independent of the counsellors training. It is important to remember that there are standard contraindications to insertion and so even if women do consent, sometimes it is not possible to perform the insertion (for example after prolonged rupture of membranes).
The outcomes measured were part of a complex interaction of many factors of which counselling was only a part, albeit an important one. Effective counselling is paramount when discussing contraception, particularly when sociocultural beliefs have a big influence on women's decision making. In Bangladesh there are many barriers to accepting contraception. Women often cited concerns that their religion would mean that they would not be allowed to be buried with a foreign body in situ were they to die. Similarly, misperceptions with regard to where the IUD could migrate once in the body included the heart and brain and constituted a further barrier to uptake. Close relatives such as a woman's husband and mother-in-law also created another layer of complexity during counselling where often the woman alone was not the decision maker with regard to her contraceptive choices.
Quality in-depth individual counselling in a country such as Bangladesh is of major importance. In the six teaching hospitals counsellors played a key role in dispelling myths and fully informing women of their contraceptive choices. Although ratification by the doctors of new methods was an important component, counsellors counselled nearly 70% of women in the two study groups. Women of reproductive age were purposefully chosen for the counselling role because it was felt that they would understand women's concerns and be better able to create a bond with the patient. Although initial training was conducted, this study does suggest that more comprehensive, structured training resulted in more confident and informed counsellors who went on to conduct not only more counselling sessions but importantly of a higher quality. While the data do not show an impact on the uptake of PPIUD before and after the training, the increased counselling rate indicates how the train-

| CONCLUSION
While the structured counselling training does not appear to have had a significant impact on PPIUD uptake, this may have been due to a multitude of other factors that influence a woman's decision and ability to access PPIUD. The reduction in removal rates and the increase in counselling rates indicate the value of having well-trained dedicated counsellors to provide comprehensive counselling to women about their postpartum contraceptive options. We believe there is great value in having competent dedicated counsellors available to provide | 55 Fatima Et aL.
in-depth counselling that may otherwise not occur given high workloads of healthcare providers in these large teaching hospitals.

AUTHOR CONTRIBUTIONS
PF, AHA, FD, and MS wrote the manuscript with help from SN. SN cleaned, analyzed, and presented the data with support from MS. All authors reviewed the manuscript before submission. MS worked on the project while employed by FIGO.

ACKNOWLEDGMENTS
We wish to acknowledge the members of the PPIUD team at OGSB and within the participating facilities who have worked hard to make this initiative possible. We would like to thank Dr Thelma Sequiera and Avni Health Foundation for their design and delivery of the counsellor training program. We would like to thank the government of Bangladesh for their support and involvement in the initiative. Finally, we would also like to thank our anonymous donors for their generous grant and continued support.

CONFLICTS OF INTEREST
The authors have no conflicts of interest to declare.