Community perceptions of pre-eclampsia and eclampsia in southern Mozambique

Background: Sub-Saharan Africa has the highest maternal mortality ratio at 500 deaths per 100,000 live births. In Mozambique maternal mortality is estimated at 249-480 per 100,000 live births and eclampsia is the third leading cause of death. The objective of this study was to describe the community understanding of pre-eclampsia and eclampsia, as a crucial step to improve maternal and perinatal health in southern Mozambique. Methods: This qualitative study was conducted in Maputo and Gaza Provinces of southern Mozambique. Twenty focus groups were convened with pregnant women, partners and husbands, matrons and traditional birth attendants, and mothers and mothers-in-law. In addition, ten interviews were conducted with traditional healers, matrons, and a traditional birth attendant. All discussions were audio-recorded, translated from local language (Changana) to Portuguese and transcribed verbatim prior to analysis with QSR NVivo 10. A thematic analysis approach was taken. Results: The conditions of “pre-eclampsia” and “eclampsia” were not known in these communities; however, participants were familiar with hypertension and seizures in pregnancy. Terms linked with the biomedical concept of pre-eclampsia were high blood pressure, fainting disease and illness of the heart, whereas illness of the moon, snake illness, falling disease, childhood illness, illness of scaresand epilepsy were used to characterizeeclampsia. The causes of hypertension in pregnancy were thought to include mistreatment by in-laws, marital problems, and excessive worrying. Seizures in pregnancy were believed to be caused by a snake living inside the woman’s body. Warning signs thought to be common to both conditions were headache, chest pain, weakness, dizziness, fainting, sweating, and swollen feet. Conclusion: Local beliefs in southern Mozambique, regarding the causes, presentation, outcomes and treatment of pre-eclampsia and eclampsia were not aligned with the biomedical perspective. The community was often unaware of the link between hypertension and seizures in pregnancy. The numerous widespread myths and misconceptions concerning pre-eclampsia and eclampsiamay induceinappropriatetreatment-seeking and demonstrate a need for increased community education regarding pregnancy and associated complications. Trial Registration: NCT01911494 This is an ancillary study of a multinational cluster randomized control trial in Nigeria, Mozambique, Pakistan and India (the Community Level Interventions for Pre-eclampsia trial-CLIP) ({“type”:”clinical-trial”,”attrs”:{“text”:”NCT01911494″,”term_id”:”NCT01911494″}}NCT01911494) [14]. For this qualitative study, four study regions in Mozambique were selected, two from Maputo Province and two from Gaza Province (Fig. 1). Each study region was equivalent to an Administrative Post (AP), with the exception of Ilha Josina Machel and Calanga administrative posts, which were combined for the purposes of fulfilling the minimal population size for a study cluster within the context of the CLIP trial, and given that they are neighbouring APs. Each region was purposively selected to reflect a variety of socioeconomic and demographic characteristics, such as level of urbanization, population density, distance to a trading centre, and presence of a referral facility. Map of study areas, southern Mozambique The Ilha Josina Machel-Calanga region is located in north-east Maputo Province, populated mainly by farmers and fishermen. This area is characterised by extremely poor transportation networks, which further deteriorate due to flooding in the rainy season. Três de Fevereiro is located in the north of Maputo Province, it is intersected by the 1st National Road (the major two-way highway in Mozambique, and the only connection between the northern, central and southern regions of the country) and has reasonable infrastructure such as modern communication networks, some secondary roads, and public services. Most residents of this AP are employed by the Xinavane Sugar Company and other private sugar and rice farms. This area is an important informal business centre, with a large sector of the young adult male population employed in the mining sector in South Africa. The two regions in Gaza Province were Messano and Chongoene. Messano,in the southwest, has a weak community infrastructure set-up including poor access to the main road. The primary occupation of residents is small-scale farming. Chongoene is a coastal region in northern Gaza. It is the newly appointed district head office, which has led to improvements in commerce, administrative services, tourism, and the agriculture sector. Most residents of the four regions belong to the Changana ethnic group. The predominant occupation is farming, especially among women. Raising livestock, informal trading, and handicrafts are the other sources of income. Most men migrate to South Africa, Swaziland and other cities in Mozambique for work. Education indicators vary between the two provinces, with a 22 % illiteracy rate in Maputo and 38 % in Gaza, in both cases literacy is lowest among women [14]. For more detailed study site characteristics see Table 1. Study site characteristics Source: Unpublished data from demographic census(2014) and demographic rounds (2015) of the CLIP study This article is a component of a larger formative study prior to the CLIP trial. While the formative research was based on a mixed methods approach, the present article focuses on the qualitative component, comprised of focus group discussions and in-depth interviews with community stakeholder groups (see Tables 2 and ​and33 for participant characteristics). Characteristics of focus group discussion participants aDespite the fact that these two Administrative Posts were combined into one single cluster, the data was collected separately bMissing data Characteristics of interview participants Focus groups were chosen to best capture community members’ views, while enabling open discussion between participants. It was difficult to convene focus groups for traditional healers and matrons due to the limited number available; therefore individual interviews were conducted with these two stakeholder groups. Data collection took place between September 2013 and May 2014. This process was conducted by a team comprising a Mozambican social scientist and four trained interviewers, all employed by the Manhiça Health Research Centre (CISM). All data collectors were fluent in Portuguese and Changana, the predominant local language. As part of the rapport-building stage, the first contact was made with the community chief at the Administrative Post level, to obtain permission for data collection. Following this, a neighbourhood was randomly selected for data collection within each AP. Neighbourhood chiefs (known as secretários dos bairros) supported the study team in the identification of participants who fulfilled the inclusion criteria for interviewsand focus groups. Participants had to belong to one of the following categories:pregnant, partners or husbands of women of reproductive age (WRA), mothers or mothers-in-law of WRA, matrons or traditional birth attendants (TBA), elders and traditional healers. The team made the final selection by verifying the characteristics of the potential participantsand the number needed for interviews and focus groups. The secretários dos bairros were instructed to identify participants from different quarteirões (the set of houses located in the same block within a bairro). Focus groups were conducted either at the círculos (the usual community gathering location), or at the community leaders’ house, as groups could easily be convened in these locations. A total of 20 focus groups were conducted with an average of 7 [6–14] participants in each session. Groups were homogeneous according to the main inclusion criterion. However, there was heterogeneity within each focus group in terms of age, residence (quarteirão), occupation and education, as captured in Table 2. Each discussion lasted for 30 to 80 min. A total of 10 interviews were conducted with community members (traditional healers and matrons). Interviews were conducted one-on-one in the home or workplace of participants, and were 30–60 min in length. Data collection instruments served asguides for the discussions, allowing for probing and follow-up questions whenever necessary. These interview and focus group guides had been usedin Nigeria, India and Pakistan in the context of the CLIP trial, and were subsequently adapted to the local context during the piloting process in Mozambique. Theguides differed slightly according to the stakeholder groups, but in general they touched upon similar themes. Although the guides were written in Portuguese, data collection was conducted primarily in the Changanalocal language. The choice of language was determined by participants’ preference. Ethical approval for this study was granted by the CISM Institutional Review Board (CIBS_CISM/08/2013), as well as by the University of British Columbia in Canada (H12-00132). Focus group discussions (FGD) and in-depthinterviews (IDI) were digitally recorded using Olympus AS-2400 PC®; IDIs and FGDs were transcribed verbatim and translated simultaneously from Changana to Portuguese for analysis at CISM. On site, quality control was ensured by a secondary review of 20 % of the transcripts against the audio recordings to confirm accuracy. Two social science researchers coded all the data, which was originally transcribed in Portuguese, in Mozambique. Twenty-six percent of all transcripts were translated into English and re-analysed by an external collaborator from UBC for quality controlandto contribute to interpretation of the data. Data from Ilha Josina Machel and Calanga were analysed separately and subsequently combined for presentation of qualitative findings. Data saturation was sufficiently met after 20 focus group discussions and 10 individual interviews. Data analysis was performed using NVivo version 10.0 (QSR International Pty.Ltd. 2012). A thematic analysis approach was taken. The coding structure was developed in advance of analysis through collaboration among researchers. Themes were subsequently adjusted and new themes were added as they emerged from the data (Fig. 2). Thematic categories used in analysis