The testing of new and modified tools for PPFP is part of a larger quasi-experimental study exploring how to maximize all contacts a woman has with the health system during pregnancy, childbirth, and the first year postpartum (ClinicalTrials.gov registration number NCT03585361, posted July 13, 2018). The study was conducted in Hetosa and Lode Hetosa districts in Arsi Zone, Oromia Region. In each district, one primary health care unit (PHCU) - a health center with five satellite health posts - was randomly assigned to an intervention arm or comparison arm. Since multiple health posts report to the same health center, randomizing individual health posts would risk contamination.

Women were enrolled in their 2 ^nd and 3 ^rd trimester of pregnancy in February–March 2017. Study staff obtained consent and conducted enrollment interviews to collect demographic information, birth history, contraceptive knowledge, contraceptive use prior to pregnancy, and intention to use contraception after birth. Women were re-interviewed in May 2018 to collect endline information on contacts with the health system during and after pregnancy, information received on family planning, use of contraception since delivery, and infant feeding and immunization. The result revealed that postpartum women in the intervention arm were more likely to adopt contraception compared to women in the comparison arm ( Sitrin et al., 2020).

After one year of intervention, the study team interviewed health officials, HEWs, and WDAs for their experiences and perspectives on the acceptability and feasibility of PPFP tracking tools and prompts. This qualitative assessment was done in both intervention and comparison clusters.

The study districts were selected because of the absence of other large-scale FP programs. We aimed to study the perspectives of all stakeholders who had experiences with the PPFP tracking tools and prompts.

The purposive sample for qualitative in-depth interviews (IDIs) included two zonal and four district health officials; four heads of primary health care units, and 23 HEWs from 18 health posts (10 intervention and 8 comparison; 18 interviews total as HEWs were interviewed jointly in health posts where two were available). In addition, we held six focus group discussions (FGDs) with 70 WDAs. No participant selected for IDIs or FGDs declined to participate. Sample size aimed to achieve saturation of themes. Additionally, up to five IMCH cards were pulled at each health post during supervision visits, conducted during implementation, to check for correct completion of the tool. A total of 180 IMCH cards were reviewed.

A conceptual framework describing women’s decision-making and contraceptive use over the continuum of care guided the study design. This framework describes opportunities for two cadres of community health workers (CHWs), HEWs and WDAs, to influence women’s postpartum contraceptive use during service contacts at multiple levels of the health system ( Figure 1). This manuscript focuses on the analysis of HEW and WDA contacts with women and the tools used at the community health system level.

Interventions to strengthen PPFP counseling, provision, and documentation were implemented at all health centers in the study districts, including comparison PHCUs. Because the study was designed to assess the additional benefit of strengthening PPFP through the health extension program, only the intervention arm received the community intervention. The community intervention had three components: 1) training and supervision of HEWs on PPFP with a refresher on implant insertion, 2) giving HEWs and WDAs tools to track women’s PPFP preferences and pregnancy risk, and 3) having HEWs orient WDAs on PPFP messages and the WDA tool. The detail imputes for each study arm are outlined elsewhere ( Sitrin et al., 2020).

The project made small modifications to the existing IMCH card to allow documentation of counseling, method preferences, and PPFP uptake at every contact during pregnancy and within one year after birth. The content of the IMCH card included PPFP counseling, method choice, and method adoption at each contact. The modified version of the IMCH card is available as Extended data; Mossie et al., 2019). In addition, prompts were added to ask women about return of menses, breastfeeding status, and complementary feeding at each growth monitoring visit to assess their return to fecundity and pregnancy risk. The FMOH provided approval to test the modified tools. We conducted regular visits to intervention health posts using a supervision tool (available as Extended data; Mossie et al., 2019) to gather observations on use of tools, capture responses to questions posed to HEWs, and extract PPFP data.

The WDA tool contained pictures of various contraceptive methods to help WDAs educate women on available options. It included boxes to check off a woman’s choice of PPFP method prior to delivery, along with boxes after birth to check breastfeeding status, use of FP, sexual activity, and if menses returned. The tool aimed to prompt WDAs to refer postpartum women to health posts if at risk of unintended pregnancy. The WDA tool is available as Extended data; Mossie et al., 2019).

We created semi-structured interview and FGD guides with questions on experiences of health care workers, HEWs, and WDA on their roles and responsibilities around PPFP and their use of PPFP tracking tools. The tools were prepared in English, translated into local language (Amharic and Afan Oromo), then back-translated. The interview guides were pilot tested in an area outside the study area with similar policy and cultural context.

Qualitative data were collected in May 2018. All FGD and in-depth interviews were audio-recorded and hand-written notes were kept after written informed consent was obtained from all participants. The FGDs lasted about 120 minutes and the IDIs were a maximum of 60 minutes long. The data collection was conducted by four research assistants. All research assistants had a postgraduate degree in public health and had received training for five days prior to the start of data collection. Supervisors oversaw field data collection.

Research assistants who had participated in the data collection transcribed audio data in the original languages, Amharic and Afan Oromo. Two study members reviewed a sample of transcripts while listening to the audio data to confirm that there was congruence between the transcripts and audio data. The authors were public health and family planning program specialists and evaluators with experience in qualitative data analysis methods in health. Several authors, including the lead, are Ethiopian and two authors are from Oromia region and included men and women. After transcription, transcripts were translated to English for coding.

The study team developed initial codes based on the content of the transcripts and the study tools. The team then coded selected transcripts during a qualitative data coding workshop and adjusted the codes. The final codebook with definitions was uploaded to a qualitative data analysis software (MAXQDA 2018; Taguette is an open-access alternative) (Supplement 1). Each transcript was coded by two separate coders sequentially. After the second coding, coders discussed any discrepancies or disagreements to reach consensus on what codes should apply to various segments.

After transcripts were coded, the team analyzed the data using the ‘One Sheet of Paper’ technique ( Ziebland & McPherson, 2006). We reviewed segments for selected codes (including all related to tracking tools), sometimes grouped by type of respondent or study arm. We then summarized the various issues expressed on a single sheet of paper, including participant identifiers and links to specific coded segments (quotes), and developed written summaries for each code. Finally, we convened an analysis workshop, during which the investigators and analysis team reviewed each summary for similarity and outliers in views and opinions of study participants and identified emerging themes. Box 1 depicts the list of themes identified using the ‘One Sheet of Paper’ technique.

To support validity, preliminary findings were presented to select study participants – HEWs, health care workers, and health officials – to check that the results and interpretations agreed with their views and opinions

The study was approved by the JHSPH-IRB (IRB No: 7143). A support letter was obtained from the Ethiopian Ministry of Health to conduct the study. Oral consent was sought and received from all study participants. We did not seek written consent because of some participant’s low literacy level, concern about collecting paperwork with identifying information and the minimal risk associated with the study. Every participant received a copy of the consent form in their preferred language.

The study enrolled 103 participants in both intervention and comparison sites, predominantly WDAs (70 participants in 6 FGDs) and HEWs (23 participants). The characteristics of study participants are presented in Table 1. The following sections explore how HEWs and WDAs used the tools introduced by the study, and the acceptability and feasibility of the modified IMCH card and WDA tool for tracking women’s decision-making and PPFP use.

How do HEWs use the modified IMCH card?

1. As a HEW behavior ‘nudge’

The modified IMCH card serves as a “nudge” reminding HEWs to conduct PPFP counselling during each visit through the continuum of care. A HEW described how recording choice of PPFP method for pregnant women helps providers follow up with women after birth to offer the FP method she chose.

2. As a tool to assess ‘risk of pregnancy’

HEWs thought that the modified IMCH card also helped to assess whether a woman is at risk of pregnancy, and thus may need contraceptives, based on months postpartum, breastfeeding practices, or return of menses. One HEW (HEW33, intervention) stated, it “ helps me, opening a new way for counselling” around risk of pregnancy, noting that in the previous card, “ there was no such opportunity in the past to identify mothers [at risk of pregnancy]”. Another HEW (HEW34, intervention) said: “…for the mother, on her return if she sees her menses, we counsel her to start family planning”. All HEWs appreciated the section on assessing pregnancy risk, though one (HEW38, intervention) talked about being confused on how to complete the pregnancy risk section, even though she strongly supported the idea.

3. To track women’s decision-making through the continuum of care

HEWs noted that they counsel their clients over several visits, and it can take repeated counseling before a woman is ready to adopt PPFP. All HEWs who participated in the study stated that the card helped them track and continuously follow up with their clients, until the woman makes a decision. One modification to the IMCH card was to record method choice during pregnancy, which allowed HEWs to refer to it during later visits, for example, during childbirth. However, women are not expected to stick to their original choices and may choose other methods if they change their mind during subsequent visits. Once they agreed to use contraceptives, HEWs provide the service and record the uptake on the IMCH card. Next, HEWs will regularly check the client if there are reports of complications with the method used. Two HEWs explained:

4. To strengthen client-provider interaction and build trust

Many HEWs commented that the IMCH card strengthened interactions and built trust between themselves and their clients as they received more than one service at a time. Clients received FP counseling while primarily visiting the health care facilities for other services, such as child health care and other services during pregnancy and in the postpartum period. This was clearly narrated by one of the HEWs:

One HEW (33, intervention) commented that the modified IMCH card prompts reminders about breastfeeding and how exclusive breastfeeding or the lactational amenorrhea method (LAM) can be used to delay pregnancy: “When we tell mothers the fact that breastfeeding the child properly will prevent pregnancy, they are very happy. That helped [HEWs] to establish better relationship with them [clients]” (HEW33, intervention).

Did HEWs properly complete the modified IMNCH card?

As shown in Table 2, during each supervision visit, up to five IMCH cards were selected at each health post (180 total) to review data quality and assess PPFP services provided. HEWs recorded almost all women who received FP services. Very few women came to health posts for postnatal care; most of those that did were recorded as having chosen an FP method. Less than half of women had brought their babies for immunization at health posts; those that did were recorded as receiving counseling on FP.

How acceptable and feasible is the modified IMCH card?

HEWs thought the modified IMCH card is feasible to use. The changes to the card helped HEWs to document all services delivered to their clients over time. It also allowed them to document which clients received PPFP counselling and to record method choice. Indeed, HEWs thought that the card has all the information they needed to provide to women during counselling sessions:

Moreover, the contents of the IMCH card were organized in a way that facilitated easy completion. Unlike tools used in the past, HEWs documented all interventions in the modified IMCH card using a code, so it did not take much time to write each of the services provided to clients. As one HEW described:

One of the HEWs noted that the program will be sustained even without external assistance as the card is very helpful. She further noted that, as the copy of the card is with them, they will continue to use it after the project is phased out.

Health officials in the study sites also appreciated the modified card’s simplicity to use. Two officials (LHET10, ZON11) specifically highlighted that the card is integrated, and hence increased its acceptance and sustainability in the healthcare system. Another official characterized the card as ‘clear and productive’ and expressed interest to use it widely:

How do WDAs use the WDA tool?

1. To encourage referrals for PPFP and other health services

FGD participants noted that the WDA tool is helpful to encourage use of services, with pictorials for ANC, childbirth, and immunization. FGD participants from intervention sites noted that after educating women on FP, they often refer them to health care facilities to receive their contraceptive choice. They also said that WDAs accompany their clients while traveling to health care facilities to receive contraceptives. An FGD participant explained:

However, WDAs in one of the comparison areas also said that they advise FP for their clients after birth and refer to health posts to receive their choice, even without the WDA tool. One of the participants, for example stated: “We advise husbands to take his wife to health post and use family planning right before sexual intercourse.” --FGD 06, P#12, comparison.

2. Support WDAs to provide client education and document PPFP services

Some FGD discussions briefly touched on the helpfulness of the WDA tool to provide FP through the continuum of care. WDAs recounted how they documented their work and educated pregnant women according to the guidelines on the WDA tool. As one stated:

FGD participants liked having something to show women; as one declared (FGD 04, P#6, Intervention): “ This card is better, it helps us to show and educate rather than oral talk”.

Despite the benefits mentioned above, some FGD participants showed a lack of knowledge and reported giving inaccurate advice to women or expressing uncertainty over how to speak with them about particular topics. The tool may thus confer them status that is not warranted by their level of expertise. An FGD participant narrated her guidance:

3. To Help WDA and women understand connections between exclusive breastfeeding, complimentary feeding, and pregnancy risk

While the WDA tool was not expected to provide comprehensive nutrition guidance, the inclusion and emphasis on LAM touches on feeding practices and behaviors. For example, one FGD participant stated:

A WDA in the comparison group also frequently mentioned giving nutrition advice. The discussion reflects exposure of WDAs in that ward ( kebele) to nutrition activities. However, in those focus groups, nutrition and FP advice is mentioned more in a list of activities, without linkages between them, along with the importance of institutional delivery and postnatal care.

How acceptable and feasible is the WDA tool?

HEWs have mixed views about WDAs’ understanding of PPFP concepts and their ability to effectively use the WDA tool. Two HEWs (HEW 40 and 41) talked about the need to help the WDAs fill out their tools, as most have no education. In contrast, some participants stated that WDAs can properly use the tool. One (HEW 33) noted that having pictorial cards was helpful to the WDA, though she observed that some WDAs initially removed the cards from the binder ring and gave them away instead of keeping the cards within the ring. However, she reported that card utilization improved over time. Another HEW (HEW 39) noted that WDAs’ use of the card was inconsistent, with some properly using them and others not. One HEW (HEW 36) felt that WDAs were using the card only because HEWs forced them, or because of the study team’s influence. This same HEW felt that WDAs are limited in the FP information they can share with women. Another HEW (HEW 34) reported that the acceptance of WDAs by their peers is low and thus limited their ability to use the tool. She stated the mothers would tell the WDAs, “you are simply boosting yourselves, you don’t know anything.”

This study included the viewpoints of various health system stakeholders with different levels of expertise. Information was sought from health officials, health service providers, and community health workers. This study used rigorous analysis techniques to capture different views and opinions of study participants. There may have been courtesy bias where study participants avoid mentioning limitations of tracking tools. Instead, people may magnify positive comments and remain reserved about negative feelings. The weak functionality of the WDA program is another limitation of this study. In some areas, the WDAs are not actively working and may not use the tool. In such a situation, it would be difficult to judge whether the tool is acceptable and feasible. Lastly, the study did not assess whether the effects of modifying the IMCH card are long-lasting. There are concerns that behavioral nudges may not continue to generate the desired behaviors over time ( Marteau et al., 2011; Matjasko et al., 2016).

This article reports on the qualitative study interviews, conducted in a small geographic area, with a limited set of facilities. Investigators are concerned that de-identification of interview and focus group discussion transcripts to completely exclude deductive disclosure of respondent identities by individuals from the study area is not possible. In consultation with the JHU School of Public Health IRB, the authors have decided not to make the data generated accessible for sharing beyond the data contained within the article, except by request of researchers to Jhpiego’s Open Data help team at opendatahelp@jhpiego.org and upon negotiation of a Data Use Agreement that ensures reasonable protection of research subjects.

Open Science Framework: Counseling at All Contacts for Postpartum Contraceptive Use: Can Paper-based Tools Help Community Health Workers Improve Continuity of Care? A Qualitative Study from Ethiopia. https://doi.org/10.17605/OSF.IO/9FZYJ ( Mossie et al., 2019).

This project contains the following extended data:

Open Science Framework: COREQ checklist for ‘Counseling at all contacts for postpartum contraceptive use: can paper-based tools help community health workers improve continuity of care? a qualitative study from Ethiopia’. https://doi.org/10.17605/OSF.IO/9FZYJ ( Mossie et al., 2019).

Data are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication).