“if you are on duty, you may be afraid to come out to attend to a person”: Fear of crime and security challenges in maternal acute care in Nigeria from a realist perspective

Background: Maternal and Child Health is a global priority. Access and utilization of facility-based health services remain a challenge in low and middle-income countries. Evidence on barriers to providing and accessing services omits information on the role of security within facilities. This paper explores the role of security in the provision and use of maternal health services in primary healthcare facilities in Nigeria. Methods: Study was carried out in Anambra state, Nigeria. Qualitative data were initially collected from 35 in-depth interviews and 24 focus groups with purposively identified key informants. Information gathered was used to build a programme theory that was tested with another round of interviews (17) and focus group (4) discussions. Data analysis and reporting were based on the Context-Mechanism-Outcome heuristic of Realist Evaluation methodology. Results: The presence of a male security guard in the facility was the most important security factor that facilitated provision and uptake of services. Others include perimeter fencing, lighting and staff accommodation. Lack of these components constrained provision and use of services, by impacting on behaviour of staff and patients. Security concerns of facility staff who did not feel safe to let in people into unguarded facilities, mirrored those of pregnant women who did not utilize health facilities because of fear of not being let in and attended to by facility staff. Conclusion: Health facility security should be key consideration in programme planning, to avert staff and women’s fear of crime which currently constrains provision and use of maternal healthcare at health facilities.

Nigeria is a coastal West African country and is comprised of 36 states and a federal capital territory. This evaluation was carried out in Anambra state, located in the south-eastern region. Anambra state was purposively chosen as a case study for in-depth understanding of the inquiry into SURE-P/MCH, because of the researchers’ longstanding engagements in the state. It had a total population of 4,453,964, and a female population of 2,059,844 in the 2006 population census [29]. The potential economic drivers in the state are agriculture (farming, fishery, pasturing and animal husbandry), markets (trade and commerce), transportation (good road networks), natural resources and numerous industries. Anambra state has the largest number of women of child- bearing age and also has the largest number of literate women in the south east zone. There is high uptake of MCH services in the state,although this is predominantly from private facilities [30]. Although there is no insurgency in Anambra state, unlike in some other states in the country, there is however, reported high rates and a general perception that security is a major challenge affecting delivery of good governance in the state, [16, 31]. The SURE-P/MCH programme was initially carried out in 12 primary health care (PHC) facilities, beginning from October 2012. A year later, another 12 PHCs were selected and included in the programme, which ended in November 2015. This study focused on the initial 12 facilities because they had a longer experience of the intervention. The following MCH interventions-antenatal care, facility- based delivery, post-natal care, immunisation and family planning were implemented in all 12 PHCs. Facility managers and health workers (nurses, midwives and community health extension workers) were all females. In addition, each facility was allocated six village health workers, all females, who identified pregnant women in the community and encouraged them to access and use facility based MCH services. Each facility also had a ward development committee (WDC) made up of a gender mix of 8–10 community members. This was a qualitative, exploratory case study using a Realist Evaluation (RE) approach [32, 33]. RE moves beyond cause and effect, to focus on ‘what works, how it works, under what conditions and for whom it works’ using the context, mechanism and outcome (C-M-O) configurations as a heuristic [27, 32, 34]. Context refers to the conditions in which programmes are introduced, and can include political and economic conditions, cultural norms and beliefs. Mechanism includes two aspects, first the process of reasoning of how subjects interpret and act upon programme intervention, at a given time, in a given context, and secondly how they interact with available programme resources. Outcomes are described as the patterns of intended and unintended consequences that result from mechanisms triggered in different contexts, and may be proximal, intermediate or distal [32, 35]. Theories about how programmes are expected to work (“programme theories”) are developed based on this configuration and are then iteratively tested and refined with empirical data gathered through appropriate methods and triangulated with available literature [32, 36]. The overall evaluation approach sought to answer the question ‘what works for whom under what circumstances, how and why’ using qualitative methods. Study was carried out in two phases. Phase 1 (P1) was exploratory and was based on two initial working theories (one Supply and one Demand), largely built from relevant contextual literature and logic map development [28], about how programme interventions introduced into a given context will trigger mechanisms which are acted upon to result in observed or implied outcomes. The Supply side theory which incorporates the state of the health facilities is as follows; “In the context of irregular payment of salaries and poorly functioning facilities in Anambra state (C), if different incentives (e.g. regular payments, training and improved working environment) are provided in a timely manner, then these interventions will make health workers feel motivated (M), and lead to sustained performance, job satisfaction and improved retention of staff (O).” We conducted in-depth interviews to glean initial knowledge to build our programme theories. Security emerged as a distinct theme from these exploratory interviews and formed a distinct programme theory, which was then iteratively tested, validated and refined in phase 2 (P2), based on views of patients and health staff. We report our findings from both phases (P1 and P2) using C-M-O linkages. In Phase 1 (March–October 2016), eight health facilities were sampled to include the four facilities that had the full complement of programme components (additional demand side intervention, although not a focus of this paper) and another four chosen randomly. Interview respondents were purposively selected from these facilities to include the facility managers, a programme midwife and a pre-existing (before programme) health worker and the VHWs. On the demand side, we sampled service users (women who had received maternal care services during the programme intervention (October 2012–November 2015, and were also receiving maternal and child care services during the study period), but who did not necessarily need to be pregnant at the time of the interview. As a result, most of the participants were multiparous (more than one facility delivery). We also sampled WDC members, who are community representatives that oversee the functioning of the facilities. Programme managers and relevant state and local government level policymakers were also interviewed. We also visited the 12 health facilities during both phases of data collection, for direct observation of the structural security components (perimeter fencing, secure gates, security guards and staff accommodation). Data collection included document reviews to ascertain the programme’s approach, in-depth interviews (IDIs) and focus group discussions (FGDs) to explore the views and experiences of a diverse group of stakeholders.35 in-depth interviews (IDIs) were conducted with policymakers (n = 9), programme managers (n = 10), facility managers (n = 8) and facility health workers (n = 8). FGDs were conducted with eight groups of service users (8–10 respondents per group), eight groups of VHWs (6 respondents per group) and eight groups of WDCs (6–8 respondents per group). Health workers comprised nurses, midwives and community health extension workers (CHEWs). Researchers, who were trained in realist qualitative interviewing [33], conducted all interviews. Information gleaned from these interviews were synthesised and informed our programme theory on Security and Safety. In Phase 2(July–December 2018), to test our middle range programme theory which we had built from information gathered from Phase 1, further interviews were conducted. These included 17 IDIs with facility health workers (n = 8), VHWs (n = 9) and FGDs with four groups of service users (5–10 per group) because these were the respondent groups (providers and users) directly involved with providing and utilizing round the clock facility-based MCH services. Although the VHWs were not officially scheduled for night duties, their perception on security, in their interphase role, was explored through IDIs in Phase 2 to further explore their experiences in depth. One of the four FGD groups with service users had 5 participants although 10 women had accepted to participate. A heavy rain on the day of scheduled interview constrained their attendance due to poor road access. The other three FGD groups had 8, 8 and 10 participants respectively. In spite of this constraint, researchers noted that saturation was achieved when compared with the other FGD interviews In addition, there is also evidence in the literature that favours small number (3–5) of participants for FGDs, as this has greater potential to explore complex topics in-depth, while there are also arguments for medium (6–8) and large number (6–12) participants to capture a wider range of views. It is customary to present focus group size in ranges in protocols because of the uncertainty of how many participants will be able to attend on the day. This uncertainty increases when participants are from poorer and undeserved backgrounds which can be particularly affected by unexpected contingencies [37–39]. All Phase 2 interviews were conducted with different respondents (from Phase 1 respondents) but within the study population (providers and users of services in the facilities that had received the SURE-P/MCH intervention). This was in order to validate and further refine our programme theory built from Phase 1 interviews. In each phase, respondents were asked to retrospectively reflect about the programme intervention from their perspective. All interviews lasted between 40 and 60 min and researchers reached saturation. Interview question guides for various groups of respondents were developed for this project and they are included as Supplementary files. Interviews were transcribed verbatim and analysed. Manual thematic analysis was undertaken, systematically identifying emerging themes, which were then organised according to whether they were perceived to be Context, Mechanism or Outcomes and initial linkages between all these were recorded. Each transcript was coded by two researchers, initially individually and then came together to agree on any disparities. Further quality check was conducted on randomly selected transcripts by two other researchers. Data analysis was guided by our hypothesis that the presence or absence of security components given (by the programme) and/or existing resources and how various stakeholders (service users and providers) interacted with these resources to produce behaviours which manifest in their actions. It is these combinations that give rise to the outcome patterns observed and reported here. With information synthesised from Phase 1, we proposed a programme theory for Security and Safety as follows; “In the context where programmes or communities ensure employment of security guards, erect perimeter fences and there is availability of accommodation and adequate lighting in the health facility premises, health workers and service users are likely to feel safer and therefore willing to provide and use round the clock MCH services, leading to improved access and utilization of MCH services.” We tested this theory, by retroductively analysing information gathered from the Phase 2 interviews. We then reported varying explanatory CMO configurations which emerged from our data, in line with the RAMESES Realist Evaluation reporting standards [40]. In the analysis, we acknowledge that realist evaluation is not primarily concerned with whether secure facilities (or insecure facilities) directly lead to increased (or decreased) facility access and utilization (i.e. causation), rather we used RE to explore how participants interacted with the resources (emotional, social, material, economic and sometimes political) offered by the SURE-P/MCH programme to produce actions which led to observed programme outcomes [34]. Ethical approvals were granted by the School of Medicine Research Ethics Committee at the Faculty of Medicine and Health at the University of Leeds (ref: SoMREC/14/097) and the Health Research Ethics Committee at the University of Nigeria Teaching Hospital (ref: NHREC/05/02/2008B-FWA00002458-1RB00002323). Written informed consent was obtained from all study participants and they were assured of confidentiality during reporting of findings.

N/A

• Dima Health