Using formative research to promote antenatal care attendance and iron folic acid supplementation in Zinder, Niger

Abstract

In Niger, use of antenatal care (ANC) and iron folic acid (IFA) supplements is suboptimal. The objectives of this paper are as follows: (a) to conduct formative research to understand barriers and beliefs among pregnant women related to ANC, IFA supplementation, and pregnancy outcomes; (b) assess the quality of currently provided ANC services; (c) use the findings to guide the development of programmatic interventions to improve coverage of ANC services and IFA supplementation of pregnant women. Structured in‐home interviews (n = 72) and focus groups (n = 4) were conducted with pregnant women in 4 randomly selected villages in rural Zinder. ANC consultations (n = 33) were observed in 5 randomly selected health centres, and exit interviews were conducted with all pregnant women and seven health agents following these observations. During workshops with stakeholders, results of the formative research were interpreted, and programmatic interventions were developed. In home interviews, 72% of women reported having attended at least one ANC visit. They also reported husbands (71%), mothers (40%), and friends (33%) supporting ANC attendance. Among those having attended ANC, only 65% reported taking IFA the day prior to the interview. Three of five health centres visited had IFA in stock. Health staff did not provide IFA supplements during 18 of 33 observed ANC consultations of which only 7 cases could be explained by the lack of IFA supplements in stock. Findings were used to design a 3‐pronged intervention: (a) behaviour change communication activities in communities; (b) quality improvement activities in health centres to strengthen ANC; and (c) provision of key supplies required for ANC.

1. INTRODUCTION

The total fertility rate of 7.6 in Niger is one of the highest in the world (United Nations et al., 2014). Similarly, the adolescent birth rate in Niger is the highest in the world with 205 births per 1,000 women (United Nations Development Programme, 2016). Considering these high fertility rates and the high rates of maternal mortality (520 deaths per 100,000 live births) and infant mortality (57 deaths per 1,000 live births between birth and 1 year of age; United Nations Children's Fund, 2016), improving health services for pregnant women in Niger is a high priority.

Until 2016, the World Health Organization (WHO) recommended at least four antenatal care (ANC) visits for normal pregnancies with the first visit occurring during the first trimester of pregnancy (World Health Organization, 2002). In the most recent WHO recommendations on ANC, the recommended number of contacts was increased to 8 with the initial contact in the first trimester, 2 contacts in the second trimester, and 5 in the third trimester (World Health Organization, 2016). According to available scientific evidence, the increased number of contact may facilitate assessment of the pregnant woman's health status and provide an opportunity to intervene if problems are identified (World Health Organization, 2016). However, in Niger, ANC attendance is far below either one of these recommendations as reported in the 2012 Demographic and Health Survey in Niger (Institut National de la Statistique & ICF International, 2013). Although the majority of women (83%) reported seeking ANC at least once during a pregnancy, only 33% of women reportedly attended at least four visits.

Adequate maternal nutrition is also important for healthy pregnancy outcomes. Maternal undernutrition contributes to foetal growth restriction, which increases the risk of neonatal deaths and stunting by 2 years of age (Black et al., 2013). To improve maternal and newborn nutrition and health, the WHO globally recommends daily iron and folic acid (IFA; 30–60 mg elemental iron with 400 μg folic acid) supplementation throughout the pregnancy and nutrition counselling about healthy eating (World Health Organization, 2013; World Health Organization, 2016). Although Niger has adopted the policy of providing IFA supplementation during pregnancy through routine ANC (Wuehler & Biga Hassoumi, 2011), less than a third of pregnant women reported consumption of >90 IFA supplements, and the prevalence of anaemia in pregnant women is high at 58.6% implying a severe public health problem (Institut National de la Statistique & ICF International, 2013). Increasing the coverage of and adherence to prenatal IFA supplementation is important as maternal anaemia during pregnancy is associated with increased risk of maternal mortality (Black et al., 2013). Prenatal iron or IFA supplementation may also be associated with a reduced risk of low birthweight (Imdad & Bhutta, 2012; Pena‐Rosas, De‐Regil, Garcia‐Casal, & Dowswell, 2015).

The present project was implemented in rural communities of the Zinder region, which is a mainly agricultural zone in East Central Niger. Little information is available about the health and nutritional status of pregnant women in Zinder, but it is expected that inadequate nutritional status is a serious concern. Results from the 2012 Demographic and Health Survey in Niger indicate that rates of underweight or body mass index <18.5 (23.5%) and anaemia (50.3%) among women of reproductive age in the Zinder region are among the highest in the country (Institut National de la Statistique & ICF International, 2013).

The importance of using formative research to identify cultural barriers and enablers to guide the implementation of locally adapted, context‐specific programmatic interventions is well recognized (Craig et al., 2013; Fabrizio, van Liere, & Pelto, 2014; Neonatal Mortality Formative Research Working, 2008). Extensive research in several countries found that prenatal IFA supplementation programs encounter barriers at the health system, health facility, community, family, and individual levels (Galloway et al., 2002; Nagata, Gatti, & Barg, 2012; Yip, 2002). Similarly, formative research conducted to develop community‐based intervention strategies to save newborn lives found that interventions need to start during pregnancy and target not only women but also family members, traditional birth attendants, and professional health staff (Neonatal Mortality Formative Research Working, 2008). In the present project, we aimed to (a) conduct formative research to understand knowledge, attitudes, and practices (KAP) among pregnant women related to ANC, IFA supplementation, nutrition, and health during pregnancy; (b) assess the quality of currently provided ANC services; and (c) use the findings to guide the development and implementation of programmatic interventions to improve coverage of ANC services and IFA supplementation of pregnant women in rural Zinder.

Key messages.

• Despite a recognition of the importance of attending ANC for their own and their baby's health, ANC attendance is less than recommended and frequently starts in the second trimester.

• IFA supplementation is also recognized as important, but inadequate supplies at health centres and counselling result in low reported IFA consumption.

• To improve health and nutrition among pregnant women and encourage early and frequent ANC attendance, a multilevel approach informed by research is required including behaviour‐change communication activities at the community level and quality improvement strategies at the health centre level.

2. METHODS

2.1. Brainstorming session to initiate the project

The objectives for the Niger Maternal Nutrition (NiMaNu) project were defined during a 1‐day workshop in Niamey, Niger, in April 2012. The primary purpose of the workshop was to inform key stakeholders about the planned project, jointly identify gaps in available information on maternal nutritional status in Niger, and decide on the primary project objectives. Representatives of different departments of Niger's Ministry of Health (MOH), and of Helen Keller International (HKI), International Nutrition (former Micronutrient Initiative), the United Nations Children's Fund (UNICEF), the World Food Programme, and local researchers were invited. Consensus was reached that the intervention should optimize programme coverage and quality of the ANC package currently recommended by the MOH of Niger. In addition, all participants agreed that additional information on maternal health and nutrition status and related KAP was needed to guide the MOH and its partners in improving pregnant women's health status.

The location of the project site was also discussed, and consensus was reached that Zinder would be suitable because of a strong regional presence of HKI, who would lead the programmatic intervention, and available information indicating that ANC coverage was low in that region. The intervention zone would be comprised of the catchment area of 12 integrated health centres (IHCs) Type 1. These primary healthcare clinics each serve a population of approximately 10,000 and provide the minimum packet of preventive and treatment services defined by the MOH, including a newly revised protocol for “refocused” ANC aiming to strengthen implementation of the essential services (Ministère de la Santé Publique, 2006). It was further agreed that an advisory committee would be established for the duration of the project and representatives of the MOH departments (Directorate for Maternal and Child Health, Directorate of Research and Programming, and Directorate of Health Care Organization), HKI, and International Nutrition would be invited.

A detailed proposal was developed, and ethical approval was provided by the National Consultative Ethical Committee (Niger) and the Institutional Review Board of the University of California, Davis (USA). The NiMaNu Project was registered with the U.S. National Institutes of Health (http://www.ClinicalTrials.gov; NCT01832688). Considering that the majority of pregnant women in this region are illiterate and that data collection involved only oral questionnaires and very limited personal data were collected, oral consent was considered appropriate by the ethical committees. The consenting procedure was conducted in two steps. First, the information was presented following a prepared script and describing all the study procedures, including the right to stop participation at any time. Second, the oral consent was documented in standard log forms.

2.2. Formative research

Prior to the implementation of the project activities at the community level, formative research was conducted in four randomly selected villages and in five randomly selected IHCs within the project intervention zone. The four villages were randomly selected from 168 villages of the catchment area of 18 IHC, and the selection was independent from the five IHCs.

The objectives of the formative research were to understand KAP among pregnant women related to ANC, IFA supplementation, nutrition, and health during pregnancy and to evaluate the quality of the ANC services offered by the IHCs. The research activities included (a) structured in‐home interviews with pregnant women, (b) focus group discussions with pregnant women conducted according to an interview guide, (c) direct observations of ANC consultations, (d) structured exit interviews with pregnant women following the observed ANC consultations, (e) structured exit interviews with health staff who conducted the observed ANC consultations, and (f) IHC infrastructure assessments. Data collection tools used for interviews of pregnant women and health staff employed a combination of structured and open‐ended questions. In most cases, the interviewer asked a question and the respondent was given the opportunity to freely respond. This information was then documented by the field worker using the prepared list of codes on the questionnaire. Specific aspects of the data collection tools are described briefly below, and the tools are accessible on Open Science Framework (Hess & Ouédraogo, 2017).

2.3. Home interviews

In each selected community, pregnant women were identified for the home interviews with the help of the traditional birth attendants and/or the local community health workers through the random walk method (United Nations, 2008). The starting point to initiate the random walk method in each village was randomly selected in each village. A structured questionnaire‐based interview was conducted with pregnant women (n = 72) on KAP related to ANC, IFA supplementation and nutrition and health during pregnancy, and birth plans. The questionnaire was informed by the content of the Essential Nutrition Actions (ENA) framework (Guyon & Quinn, 2011; USAID et al., 2014) to compare women's actual practices to recommended practices during pregnancy.

2.4. Focus groups

In the same four villages, 8–10 pregnant women, who were not interviewed in the home, participated in focus group discussions. The purpose of the focus groups was to complement the information collected through structure questionnaires mentioned above with qualitative information obtained through focus group discussions. A focus group guide was developed prior to the focus group sessions, and Hausa‐speaking facilitators were trained on guiding the discussions. The focus group discussions covered topics including KAP related to IFA supplementation, diet, ANC and other preventive health services, services received during ANC, plans for delivery and breastfeeding, and wishes for the baby.

All focus group discussions were audio‐recorded for complete transcripts, and two observers took notes. The focus groups occurred in Hausa, and notes were taken in French. Immediately following each focus group, the study staff conducted a team debriefing to discuss and complete the records. The transcripts were then analysed with inductive methods to identify themes around beliefs about ANC, key influencers of care seeking and nutrition health practices, and aspirations of women for themselves and their children.

2.5. ANC observations and follow up interviews

On randomly selected days, we visited the five selected IHCs, observed 5–10 ANC consultations in each IHC conducted on these days (n = 33), and interviewed all health staff involved in the ANCs (n = 7). All 33 pregnant women who had visited the IHCs for ANC on the days of observation were interviewed individually following the observed ANC sessions. Structured checklists were developed based on the MOH's worksheet for ANC visits and pretested (Ministere de la Sante Publique, 2008). Field workers were instructed to not interfere with the flow of the ANC and to avoid any distractions to the extent possible.

2.6. Infrastructure assessments and staff

The availability of supplies and equipment required for ANC was recorded during an infrastructure assessment at the same five IHCs. The infrastructure data collection tool was developed following the guidelines by the MOH for supplies relevant for ANC (Ministere de la Sante Publique, 2008). The number of all staff at the IHC, and specifically those involved in any aspect of ANC, and their training and level of experience was also documented.

2.7. Data interpretation workshops with key stakeholders

After completion of the formative research assessments, two workshops were organized, one in Zinder and one in Niamey, to share and interpret results and develop programmatic interventions together with key stakeholders. Representatives of IHCs, health districts, the regional directorate of public health in Zinder, NiMaNu project staff, and team members of the University of California Davis, HKI and Micronutrient Initiative participated at the regional workshop in Zinder. Many of these same individuals also participated at the national workshop in Niamey, in addition to an independent consultant and national representatives of the MOH. Results of the formative research were presented, the essential antenatal interventions included in the national policy were reviewed, and an optimization strategy to improve ANC services and women's attendance was discussed.

3. RESULTS

3.1. Results of formative research

3.1.1. Knowledge, attitudes, and practices related to ANC

The 72 pregnant women, who participated in the home interviews, had a mean age of 26.3 ± 6.6 years, and the majority had been pregnant more than 3 times (Table 1). All women reported believing that attending ANC was important for their own health and the health of their foetus. Less than half of the women (44%) knew that the recommended number of ANC visits was four. Others either did not know (7%) or thought the recommended number of visits was 3 (18%) or >4 (30%). About three quarters of women reported having attended at least one ANC consultation during the present pregnancy and among them 11 out of 50 reported having begun ANC in the first trimester. The majority of the women reported that they received support to attend ANC from their husbands (71%), their mother (40%), friends and neighbours (33%), and their mother‐in‐law (14%), and most (96%) reported that no one advised them against attending ANC. The majority of interviewed pregnant women (92%) did not report anything negative about their ANC experiences. But among the rest, the following statements were made by individual women:

Table 1.

Baseline characteristics of pregnant women interviewed at their homes, and following ANC observations

Characteristics	Home interviews	ANC observations and follow up interviews
N	72	33
Age (yrs), mean ± SD	26.3 ± 6.6	26.2 ± 7.3
Marital status, n (%)
Married	70 (97.2)	33 (100)
Single, divorced, widowed	2 (2.8)	0
Education, n (%)
Formal	20 (27.8)	6 (18.2)
Koranic	29 (40.3)	5 (15.2)
None/literacy	23 (31.9)	22 (66.7)
Ethnicity, n (%)
Hausa	67 (93.0)	27 (81.8)
Touareg	2 (2.8)	3 (9.1)
Foulani	2 (2.8)	2 (6.1)
Others	1 (1.4)	1 (3.0)
Gravidity, n (%)
1	10 (13.9)	7 (21.2)
2–3	16 (22.2)	8 (24.2)
>3	46 (63.9)	18 (54.5)
Estimated number of months since last menstrual period
1–3 mo	8 (11.1)	1 (3.0)
4–6 mo	32 (44.4)	13 (39.4)
7–9 mo	31 (43.1)	18 (54.5)
Irregular or no menstrual period since last pregnancy	1 (1.4)	1 (3.0)

Note. ANC = antenatal care.

Poor hospitality at health center.

Lack of medications at health center.

Absence of health care provider.

Long distance to attend ANC visits.

Fifty women participated in focus groups held in four villages. Discussions during the focus groups corroborated the reported attitudes and practices regarding ANC stated during home interviews; namely, women in the focus groups agreed that ANC was important for their health and that of their babies. However, they disagreed in regard to visiting the ANC in the first trimester. In three of the four villages, women said that they generally start seeking ANC between the fourth and fifth month of gestation. The women attending the focus group in the remaining village reported starting ANC visits between the third and fourth month of gestation. An important statement they agreed on was as follows:

“In our culture a woman should be showing before attending any ANC.”

3.1.2. Knowledge, attitudes, and practices related to IFA supplementation

When shown a sample, almost all women (96%) recognized the IFA supplements as the “medicine to increase blood,” which is the local name used for IFA supplements. Fewer knew that one supplement should be taken per day (52%) and that it should be taken for the duration of pregnancy (59%). Among those having attended ANC (n = 52), only 65% reported having taken IFA the day prior to the interview, and the same proportion reportedly consumed it sometime during the previous month.

Several positive statements about IFA supplements were made by individual women during open‐ended questions:

Increases blood and prevents poor blood.

Gives health.

Reduces the complications during pregnancy and delivery.

However, a small minority (7%) of women voiced concerns and risks associated with IFA supplementation:

Causes nausea and dizziness.

Excessively increases blood.

Increases bleeding during delivery.

3.1.3. Knowledge, attitudes, and practices related to nutrition, physical work, and malaria prevention

Most pregnant women reportedly changed their dietary habits because they became pregnant and almost all wished that they could eat additional foods (Table 2). In particular, many of the respondents at the home interviews stated that they wished to eat more meat (50%), fish (31%), vegetables (47%), fruits (35%), rice (42%), spaghetti (42%), milk (24%), and eggs (14%). Half of all women reported that their husbands recommend that they work less during pregnancy. However, 32% reported that nobody helped them with their work load, and very few women (8%) said that they received help from their husbands. Women were well aware of the risks related to malaria, and many practiced recommended malaria prevention strategies (Table 2).

Table 2.

Knowledge, attitudes, and practices of pregnant women related to nutrition, physical work, and malaria prevention

	N pregnant who received advice / total N
All	72
Diet
Changed dietary practices due to pregnancy, n (%)	52 (72.2)
Wish to eat other/additional foods, n (%)	62 (86.1)
Would like to eat more than usual, n (%)	57 (80.3)
Physical work and rest
Changed work due to pregnancy, n (%)	51 (70.8)
Receive advice from family members to work less, n (%)	52 (72.2)
Malaria prevention
Believe that consequences of malaria are worse for pregnant women, n (%)	71 (98.6)
Sleep under bed net, n (%)	49 (68.1)
Obtained intermittent preventive antimalarial treatment, n (%)	54 (75.0)

3.2. Results of ANC observation, interviews with pregnant women and health staff, and infrastructure assessments

3.2.1. Direct observations of ANC visits

All of the 33 observed ANC consultations were conducted by female health staff who were certified as midwives. Pregnant women were on average 26.2 ± 7.3 years old (Table 1). For half, this was their first ANC visit in the present pregnancy. In the majority of ANC consultations, the midwife greeted the patient (30/33) and/or offered her a seat (31/33). Danger signs during pregnancy were reviewed in 23 of the 33 visits, and the midwife listened to the foetus' heartbeat in 25 of 33 cases. The assessment of pre‐eclampsia risk factors was not done frequently: Only half of women (16/33) had their blood pressure measured, very few were asked for potentially related danger signs (6/33), and proteinuria was assessed among even fewer (3/33). Haemoglobin concentration was not assessed in any of the 33 patients; however, the majority were examined for conjunctival and/or palmar pallor (29/33), and about half were asked about fatigue (15/33). Eight were diagnosed as anaemic based on pallor. Additional physical exams included vaginal exam (10/33); exam for oedema (33/33); body temperature (3/33); respiratory rate count (1/33); and measurement of height (6/33), weight (21/33), and mid‐upper arm circumference (16/33).

IFA supplements were discussed with less than half of the women. In particular, the midwives asked only a few (9/33) whether they were currently taking IFA and confirmed with only four whether they still had supplements available at home. IFA supplementation was provided to less than half (15/33), and very few were informed about why and how to take the supplement (2/15) or of potential adverse effects (0/15). By contrast, many midwives mentioned the importance of sleeping under a bed net (24/33) and of returning to the health clinic in the event of any danger signs (25/33). Only one woman received advice on the recommendation to attend at least four ANC visits during the pregnancy. Nutritional and other advice according to the ENA framework (Guyon & Quinn, 2011; USAID et al., 2014) was provided infrequently (Table 3). None of the midwives invited the women to ask questions during or at the end of the visit, but almost all (32/33) made an appointment for the next visit.

Table 3.

Nutritional and other advices provided during directly observed ANC visits

Advice	N pregnant who received advice / total N ANC observed	N midwives who believed that the following advicea should be given to pregnant women / total N interviewed
N total	33	7
Eat more than usual	5/33	2/7
Eat a large variety of different foods	1/33	2/7
Eat orange‐flesh fruits and vegetables	7/33	3/7
Eat meat, eggs and dairy products	11/33	5/7
Eat green‐leafy vegetables	20/33	6/7
Eat iodized salt	0/33	1/7
Work less and avoid lifting heavy subjects	4/33	n/a
Try to rest more	2/33	n/a
Breastfeed your baby immediately after birth	10/33	n/a
Breastfeed your baby exclusively for 6 months	17/33	n/a

Note. ANC = antenatal care; n/a = not available.

^aFree response, that is, individual guidelines were not provided by the interviewer.

3.2.2. Interviews with pregnant women directly following the observed ANC visits

For over half of the women interviewed following the observed ANC visit (n = 33), the present pregnancy was at least their fourth pregnancy (Table 1). The majority (22/33) said that they expected to be able to follow at least some of the advice they received from the midwife. Fourteen reported to have consumed IFA in the previous month, of whom nine reported daily consumption and only one mentioned experiencing adverse side effects. The majority (31/33) said they believed it was beneficial to consume IFA supplements, although four expressed concern about adverse side effects such as vomiting, dizziness, and discomfort.

When asked about their experience in the consultation, the majority felt they had been treated respectfully (32/33), were satisfied with the service they received (32/33), and all found that it was important for their own and their baby's health. Only three women had expected additional services, such as urinary analysis, receipt of a bed net, medicine, or vaccine or additional advice. As with the pregnant women interviewed by home visit, the majority (31/33) reported their husbands were supportive of their attending ANC. One said that one of her co‐wives was opposed to ANC, and two women said their friends/neighbours advised against ANC. All but one intended to attend the next scheduled ANC appointment.

3.2.3. Interviews with health staff directly following the ANC

Seven midwives were interviewed following the direct ANC observations. Their work experience ranged from less than 1 year to over 15 years. Three of the seven midwives had received a supervisory visit by the health district within the past year, whereas four had not been supervised in ≥2 years. Almost all (6/7) said they did not assess the haemoglobin concentration of pregnant women because of lack of supplies. Similarly, proteinuria was not diagnosed because midwives reported that urine rapid diagnostic test kits were generally not available (4/7). Diagnostic tests for malaria were only performed in patients if there was some concern (5/7).

The interviews suggested that not all the relevant ENA messages (Guyon & Quinn, 2011) were known to the midwives (Table 3). Most (6/7) said that they always give IFA supplements to their patients, the remaining one said that she did so when they were available. All reported to always ask the patient if she is taking IFA, and the majority (5/7) claimed they always asked about side effects.

The seven midwives were also asked about their work with open‐ended questions. They all expressed the opinion that the new national ANC guidelines required too many activities for one midwife working alone to provide: a physical examination, tetanus vaccines, IFA supplements, presumptive treatment of malaria, nutrition counselling, HIV counselling, among other tasks. They also expressed frustrations with challenges such as shortages of supplies and medication, and pregnant women who refuse to be physically examined or who do not follow advice given, in particular related to family planning. On the other hand, when supplies and medication were available, the job was perceived as enjoyable. Another positive aspect mentioned was the team work among the health staff. When asked what would make ANC services more effective, the most common response was ensuring the availability of supplies and medication needed for ANC.

3.2.4. Health centre infrastructure assessment

Among the five IHC inspected for adequacy of staffing, infrastructure and current inventory, just two had electricity and two had a source of running water. All IHC had at least two qualified nurses (infirmiers diplomé d'état) on staff, and all of the IHC had at least one qualified midwife with a state diploma (sage‐femme d'état). All midwives were involved in ANC. No IHC had a lab technician, and all utilized a network of at least seven community health workers. The infrastructure assessment revealed a lack of many supplies and equipment needed to conduct ANC services (Table 4).

Table 4.

Supplies and equipment available at the time of the formative research in five health centres

Well‐functioning supplies / equipment	N available at the time of the interview / total health centre
Total health centres (N)	5
Thermometer	3/5
Adult scale	4/5
Adult stadiometer	4/5
Fundal height measuring tape	5/5
Obstetrical stethoscope	4/5
Blood pressure device	2/5
Stethoscope	3/5
Supplies for haemoglobin assessment	3/5
Supplies for malaria assessment	3/5
Supplies to assess proteinuria	3/5
Exam table for antenatal care	4/5
Mother–child health cards	3/5
Iron folic acid supplements	3/5
Intermittent preventive antimalarial treatment	3/5
Impregnated bed nets	1/5
Preventive antihelminithic medication	4/5
Antibiotics	1/5
Tetanus toxoid vaccines	5/5

3.3. Design and implementation of the programmatic intervention strategy

During the above‐described workshops held to interpret results with local, regional, and national representatives of the MOH and other stakeholders, agreement was reached that a multilevel approach was required to improve ANC attendance and experiences, and nutrition and health practices during pregnancy. Thus, a programmatic strategy was developed that encompassed three strategies: (a) community‐based behaviour‐change communication (BCC) activities, (b) quality improvement at the IHCs, and (c) donation of supplies essential for ANC.

3.3.1. Community‐based BCC activities

On the basis of the analysis of the formative research findings, the team identified eight specific priority practices to promote through individual counselling and group discussions (Table 5). The strategy at the community level was guided by the ENA framework (Guyon & Quinn, 2011; USAID et al., 2014), a training curriculum, and BCC approach designed to help health workers and community volunteers encourage women and their families to adopt optimum nutrition practices. Recognizing the importance of involving husbands and other members of the pregnant woman's family (World Health Organization, 2015), the community‐based activities aimed to reach multiple target populations (Table 5). The theoretical framework of the BCC strategy recognizes that individual behaviours are shaped by personal choices, social norms, environmental and economic constraints, and public policies (Kok et al., 2016), and therefore, effective strategies must work on multiple levels within the resources available.

Table 5.

Practices, barriers, facilitators, and strategies identified for behaviour‐change communication activities of the NiMaNu project

Practicea	Target audiences	Barriers	Facilitators	Behaviour‐change communication strategies
Attend ≥4 ANC visits beginning in first trimester	Pregnant women Husbands Mothers‐in‐law Neighbours/friends (include village chiefs on importance of ANC)	1. Lack of knowledge of recommendations 2. Distance/lack of transport 3. Poor treatment at health centre 4. Taboos against admitting pregnancy before physical signs	1. Recognition of importance of ANC for mother and foetus 2. Support of husbands and often other family members 3. Desire to know about health of foetus	1. Provide information and explain benefits via mothers & husband groups 2. Problem solving in mother/husband group discussions 3. Community groups to develop transport solutions 4. Community theatre
Take daily IFA during pregnancy and for 6 months post‐partum	Pregnant women Husbands Mothers‐in‐law Neighbours/friends	1. Low rate of ANC visits 2. Lack of knowledge of recommendations 3. Stock outs at IHC	1. Understand benefits 2. Side‐effect issue rare	1. Inform about frequency and duration of IFA (and managing side‐effects) 2. Use mothers' groups to motivate and form habit 3. Community theatre
Pregnant women sleep under insecticide‐treated bed net	Pregnant women Husbands Mothers‐in‐law Neighbours/friends	1. Lack of treated ITN 2. Cost of ITN	1. Understand benefits and importance	1. Encourage women's savings groups to facilitate purchase 2. Motivate husband groups to purchase 3. Problem‐solving in discussion groups 4. Community theatre
Pregnant women seek IPT‐p ≥ 2 times during pregnancy	Pregnant women Husbands Mothers‐in‐law Neighbours/friends	1. Low rate of ANC visits 2. Lack of awareness among mothers/fathers 3. Stock‐outs at IHC or non‐compliance by nurses	1. Desire for health baby 2. Understanding of dangers of malaria	1. Mother/husband group discussions to motivate and problem solve 2. Community theatre
Pregnant women eat iron‐rich foods, especially animal source	Pregnant women Husbands Mothers‐in‐law Neighbours/friends	1. Cost and availability 2. Lack of knowledge of good plant sources and combinations and/or importance 3. Increasing food insecurity	1. Eating meat is valued; men more likely to consume 2. Women wish to control symptoms of anaemia 3. Fathers want healthy babies 4. Crickets are consumed when available	1. Work with husband groups to motivate to buy animal foods for wives 2. Women's Savings groups for income generation 3. Promote production & consumption of beans and without coffee/tea 4. Community theatre
Pregnant women increase dietary diversity and eat additional meal	Pregnant women Husbands Mothers‐in‐law Neighbours/friends	1. Poverty 2. Seasonal availability 3. Lack of understanding of “diversity” and “food groups” 4. Perception of fruits & vegetables as “light” (less nutritious than grains) 5. Lack of understanding by husbands	1. Rich diet valued 2. Mothers and fathers want healthy baby	1. Discussion groups teach what is diversity and problem solve about overcoming barriers 2. Promote homestead food production 3. Community theatre
Families respond immediately to first signs of danger	Pregnant women Husbands Mothers‐in‐law Neighbours/friends	1. Lack of transportation 2. Preference for home delivery	1. Danger signs well known 2. Desire for safe and healthy delivery	1. Problem solving via women and husband groups 2. Community theatre
Families prepare a birth planb	Pregnant women Husbands Mothers‐in‐law Neighbours/friends Village chiefs	1. Lack of understanding of purpose/value 2. Lack of transportation	1. Risks known 2. Desire for safe and healthy delivery	1. Discussion groups teach value and motivate planning 2. Husband groups work to find solutions 3. Community theatre
Importance of reducing heavy physical labour during pregnancy	Husbands Mothers‐in‐law Neighbours/friends	1. Heavy labour burden and lack of family members to take over	1. Desire of both mother and father for healthy baby 2. Recognition that heavy labour may cause risks 3. Most women have reduced labour to some extent	1. Work with husband groups to find solutions 2. Community theatre about risks/solution

Note. ANC = antenatal care; IFA = iron folic acid supplement; IHC = integrated health centre; IPT = intermittent preventive antimalarial treatment; ITN = insecticide‐treated bed nets.

^aBehaviour‐change communication activities at the community level were guided by the Essential Nutrition Actions framework (Guyon & Quinn, 2011; USAID et al., 2014).

^bAlthough not probed in formative research, its importance emerged and thus was added during development of communications strategy.

In each target community, four community health volunteers were trained by a NiMaNu project BCC coordinator and a health agent from the supervising IHC; in all, 233 volunteers participated in a 3‐day training and received illustrated job aids with key messages and background rationale. Training reviewed the key practices together with techniques for counselling for behaviour change, essentially identifying a clear behaviour, and forging an agreement with an individual to try the new practice for a few days, then returning to review the experience.

After the training, the volunteers worked in pairs, one male and one female, and were expected to organize a minimum of two‐group discussions per month to review the advantages of the various priority practices and discuss barriers and solutions. A plan was established to ensure all priority practices were covered over a period of 6 months. The BCC sessions could involve homogeneous or mixed participants from the primary and influencing audiences (pregnant women, their husbands, senior women, village chiefs, and other community members). The sites for group discussions were opportunistic, ranging from water collection points and savings groups for women to mosques and boutiques for men. In addition, many communities organized theatrical performances to illustrate the benefits of ANC using characters confronting real‐life situations.

Two field workers and a BCC coordinator assured regular supervision, including monthly meetings with groups of volunteers to review their log books, discuss and resolve difficulties, and motivate the volunteers.

3.3.2. Quality improvement approach

In addition to the community‐level intervention that targeted service demand, the project team and government partners engaged a quality improvement expert to conceive the strategy for strengthening ANC services by IHCs and to address supply‐side constraints. The approach built on successful methodologies applied elsewhere in Niger and neighbouring countries (Legros et al., 2002). The ANC staff of the 12 targeted IHC were grouped into three quality improvement teams. After an initial introduction to principles of quality improvement, each team worked together to develop a plan, based on their own solutions but with input from regional and local MOH coaches (supervisors), for improving performance in four key identified weaknesses: supply management to prevent stock‐outs; work organization to ensure service providers for all clients; capacity building in clinical procedures, counselling techniques, and record‐keeping; and strengthening partnerships with and feedback from communities served. The plans included clear indicators for assessing progress. The teams then implemented the plan for a period of 4–6 weeks with weekly visits from MOH coaches and then met together again to review achievements, challenges, and revisions to the plan. The process followed the plan‐do‐study‐act cycle and was based on the principle that engaging staff to find their own solutions and cooperating to implement them is motivating and sustainable (Berwick, 1996). Each team was expected to pursue an improvement plan intensively for 6 months, then share their best practices with other teams.

3.3.3. Supply donation approach

To address the shortages of key medical inputs and weaknesses in supply management identified by the formative research, an agreement was established with UNICEF to donate IFA supplements, insecticide‐treated bednets, albendazole, scales, and stadiometers. The NiMaNu project financed other essential supplies that were identified as lacking, including health cards for mothers and infants, albendazole, cotrimoxazole, malaria rapid diagnostic kits, urinalysis rapid diagnostic test kits, intermittent preventive sulfadoxine/pyrimethamine treatment, amoxicilline, magnesium sulfate, thermometers, stethoscopes, and blood pressure monitors. The rationale was to ensure the project's ability to assess the impact of a well‐functioning ANC service.

4. DISCUSSION

In the present, paper we describe how formative research was used to inform a programmatic intervention aiming to improve nutrition and health of pregnant women in rural Zinder. In particular, findings revealed a need to improve ANC attendance (early seeking and increased frequency) as well as the quality of ANC services (reducing supply stock‐outs and improving health agent knowledge and counselling skills). Consequently, the programmatic intervention targeted demand through BCC activities at the communiy level and supply through a quality improvement strategy at the health centre level. New ANC guidelines issued by WHO highlight the importance for pregnant women to have a positive experience during ANC visits and the need for such an intervention strategy at multiple levels (World Health Organization, 2016). Similar conclusions were drawn from research to understand how to improve prenatal IFA supplementation and ANC experiences in other settings (Galloway et al., 2002; Neonatal Mortality Formative Research Working, 2008).

The goals of the community‐based BCC activities were to convey the importance of attending ANC early and regularly throughout the pregnancy and to share information about important nutrition and health practices during pregnancy. Additional programmatic strategies may also be useful to increase ANC attendance. For example, community outreach to bring health services to communities far from health centres may overcome the barrier of distance previously identified as a risk factor for low use of health services in Niger (Blanford, Kumar, Luo, & MacEachren, 2012).

About three quarters of women interviewed in our study reported having attended ANC in the present pregnancy. Most stated that they believed attending ANC was important for their own and their babies' health and that their husbands and other family members were also supportive of ANC. Although the facility assessment found a shortage of essential supplies and ANC observations revealed that not all recommended services were provided, very few women expressed dissatisfaction with the experience. However, we were unable to identify reasons for this in the present study. It may in part be due to cultural or social norms for women to not speak up or to the fact that women in this region have not experienced high‐quality ANC services and thus are not aware what they could expect more. Poor quality of health services has been found in other low countries. In particular, formative research using a variety of methods to collect information from a range of respondents in Ghana, Mali, Bangladesh, India, and Nepal found that the perceived quality and usefulness of visits were low, that counselling skills of ANC providers were inadequate, and that delivery of health education messages was ineffective (Neonatal Mortality Formative Research Working, 2008). In addition, a study in Cameroon found poor awareness of essential reproductive health interventions among health staff (Tita, Selwyn, Waller, Kapadia, & Dongmo, 2005).

In our study, 22% of the women interviewed reported to have attended ANC in the first trimester. However, it is likely that women's perception of their first trimester is not in line with the medical definition, because when asked during focus group discussions, the majority of women agreed that the first ANC visit should be between the fourth and the fifth months of gestation. It was also mentioned that the pregnancy should show before attending ANC. These findings suggest strong cultural barriers to beginning ANC in the first trimester. The practice to keep a pregnancy a secret for as long as possible is common in many other parts of the world, as it is often seen as a time of increased vulnerability (Neonatal Mortality Formative Research Working, 2008). BCC strategies therefore need to address traditional social norms that may be rooted in well‐founded fears of the adverse pregnancy outcomes faced in low resource settings, whereas quality improvement strategies must effectively demonstrate that early care seeking reduces these risks.

Most women in the present study recognized IFA tablets and knew their benefits; indeed, locally they were known as “medicine to increase blood.” During the home interviews and ANC follow up interviews, the majority reported that they believed that IFA supplements were important, and very few raised concerns about side‐effects or excessive bleeding during delivery. Nevertheless, IFA coverage was low. A study in Senegal found that adherence to IFA supplements can be increased by providing women with clear instructions about the recommended intake and about health benefits associated with optimal adherence (Seck & Jackson, 2008). In our observations, this advice was provided to only two pregnant women during ANC visits. An additional concern is an inadequate duration of IFA supplementation, potentially exacerbated by initiating ANC visits late in pregnancy. When ANC is delayed until late in pregnancy, women do not receive the full dose of IFA recommended in the national policy (Ministere de la Sante Publique, 2008), an important concern in a setting where anaemia is high. Inadequate duration of IFA supplementation was also found in a multi‐country assessment of barriers related to IFA supplementation (Galloway et al., 2002). Another important barrier in our setting was stock‐out in the supplies. Two out of five IHC had no IFA supplements available at the time of our infrastructure assessment. Inadequate supplies of IFA supplements have also been identified as the most significant barrier of iron supplementation in many parts of the world in an extensive review of qualitative research by Nagata et al. (2012). Another barrier identified in our study was the failure of health staff to provide IFA supplements, even when the IFA supplements were available. Indeed, knowledge gaps in several pregnancy‐related health aspects and recommended services by the health staff were identified during interviews with midwives in the present study. Our intervention therefore attempted to guarantee adequate supplies as well as training and coaching for health centre staff to provide IFA distribution during each ANC visit and counsel pregnant women on the importance of daily consumption.

The key messages for nutrition counselling in the national MOH manual for ANC training focus on counselling women to (a) consume an increased variety of healthy foods, (b) reject dietary taboos that advise restricting the consumption of healthy foods, (c) eat sufficient amounts, and (d) avoid strenuous physical activities (Ministere de la Sante Publique, 2008). Our ANC observations found that the most common recommendations by midwives were “to consume green leafy vegetables,” “to eat meat, eggs, and dairy products”; and “orange‐flesh fruits and vegetables.” However, the proportion of women who can indeed increase consumption of these foods considering the depth of poverty and limited availability of foods needs further exploration. Indeed, many interviewed pregnant women stated that they wished to eat more diverse foods. In view of the high prevalence of multiple micronutrient deficiencies among pregnant women recently documented in this region of Niger (Hess et al., 2017; Wessells et al., 2017), IFA supplementation and nutrition counselling alone may not be adequate and multiple micronutrient supplements along with other nutritional interventions may be required.

A strength of the present project was the use of formative research and collaboration with government partners to inform the development of a programmatic intervention. In particular, the results of the formative research broadened our programmatic approach to address service quality as well as community‐based BCC. Our research applied a variety of qualitative and quantitative methods, ranging from structured home interviews and focus groups to direct ANC observations. An important limitation of the ANC observations was our inability to determine what the appropriate services and exams should have been in relation to the gestational age and pregnancy history of each pregnant woman due to our commitment not to interfere with the provision of the ANC. Thus, we report the findings without making specific judgements.

5. CONCLUSION

Challenges to achieve optimal health and nutrition practices during pregnancy have been documented in many low‐income settings. Although pregnant women participating in formative research in rural Zinder reportedly appreciated the importance of ANC and IFA for their own health and that of their baby, their use of these essential services was less than recommended. Results suggested a need for strategies to improve ANC attendance (early seeking and increased frequency) as well as the quality of ANC services (reducing supply stock‐outs and improving health agent knowledge and counselling skills) and confirmed findings of other studies that interventions should target the multiple barriers at both the community and health centre levels.

CONFLICT OF INTEREST

Banda Ndiaye is an employee and Maimouna Doudou was an employee of Nutrition International at the time of the project implementation. All other authors declare that they have no conflicts of interest.

CONTRIBUTIONS

The formative research study was designed by SYH, CTO, IFB, KRW, NK, MD, and JN. Data were collected by CTO with guidance from SYH, KRW, and NK. Results were interpreted by all co‐authors. SYH drafted and JN edited the manuscript. All co‐authors critically reviewed and approved the final version of the manuscript.

Hess SY, Ouédraogo CT, Bamba IF, et al. Using formative research to promote antenatal care attendance and iron folic acid supplementation in Zinder, Niger. Matern Child Nutr. 2018;14:e12525 10.1111/mcn.12525