Understanding mistreatment during institutional delivery in Northeast Nigeria: A mixed-method study

Background: Improving quality of care including the clinical aspects and the experience of care has been advocated for improved coverage and better childbirth outcomes. Objective: This study aimed to explore the quality of care relating to the prevalence and manifestations of mistreatment during institutional birth in Gombe State, northeast Nigeria, an area of low institutional delivery coverage. Methods: The frequency of dimensions of mistreatment experienced by women delivering in 10 health facilities of Gombe State were quantitatively captured during exit interviews with 342 women in July-August 2017. Manifestations of mistreatment were qualitatively explored through in-depth interviews and focus groups with 63 women living in communities with high and low coverage of institutional deliveries. Results: The quantitative data showed that at least one dimension of mistreatment was reported by 66% (95% confidence interval (CI) 45-82%) of women exiting a health facility after delivery. Mistreatment related to health system conditions and constraints were reported in 50% (95% CI 31-70%) of deliveries. In the qualitative data women expressed frustration at being urged to deliver at the health facility only to be physically or verbally mistreated, blamed for poor birth outcomes, discriminated against because of their background, left to deliver without assistance or with inadequate support, travelling long distances to the facility only to find staff unavailable, or being charged unjustified amount of money for delivery. Conclusions: Mistreatment during institutional delivery in Gombe State is highly prevalent and predominantly relates to mistreatment arising from both health system constraints as well as health worker behaviours, limiting efforts to increase coverage of institutional delivery. To address mistreatment during institutional births, strategies that emphasise a broader health systems approach, tackle multiple causes, integrate a detailed understanding of the local context and have buy-in from grassroots-level stakeholders are recommended. Gombe State has an estimated population of 2.6 million based on the last census in 2006. Gombe State has 11 Local Government Areas, is multi-ethnic and 80% rural [32, 33]. The state has a high maternal mortality ratio, estimated at 1549 maternal deaths per 100,000 live births, neonatal mortality is estimated at 35 neonatal deaths per 1000 live births and just 29% of women had delivered their most recent newborn in a health facility [26, 32–34]. Public health services account for 98% of institutional deliveries in the state [26, 32–34]. In Gombe State, approximately 486 public health facilities provide labour and delivery services 460 of which are primary health facilities and 28 are referral facilities offering both labour and delivery services and specialised care [35]. Lower cadre health care workers, for example, community health extension workers (CHEW), junior community health extension workers (JCHEW) and community health officers (CHO) comprise the majority of the health care workforce [36]. In primary health care facilities, nurses or midwives are responsible for the organisation and provision of MNH services with the assistance of CHEWs, JCHEWs or any lower cadre health worker available. In the absence of nurses of midwives these lower cadre health workers must take full responsibility. In most of the PHCs, health care providers are not available 24 h a day, limiting access to facility-based care in case of an emergency or a delivery outside daylight hours. This problem is less acute in referral health facilities where nurses or midwives contribute to the organisation and delivering of MNH services under the supervision of a medical doctor. Quantitative and qualitative data were collected in 2017, as part of a programme of work to understand the quality of maternal and newborn care in Gombe State. The quantitative study involved conducting structured exit interviews with recently delivered women upon discharge after institutional delivery, in 10 primary health care facilities. The qualitative study included in-depth interviews (IDIs) and focus groups with 63 women who had recently delivered in a health facility in two local government areas (LGAs) (districts): Kaltungo, characterised by higher levels of facility births, and Kwami, where there are low levels of facility deliveries, and are reported below in line with the Consolidated Criteria for Reporting Qualitative Research (COREQ) [37]. This research was conducted with approval from the ethics review boards of the Federal Ministry of Health Abuja, Nigeria, the State Ministry of Health Gombe State, Nigeria and the London School of Hygiene & Tropical Medicine (reference 12,181). To obtain participant’s informed consent, all potential participants were provided with study information sheet and a consent form. The information sheet provided information to the participants about their right to participate or refuse to participate in the study, the right to change their mind about participating during the course of the study, and the right to withdraw from the study at any time. The information sheet was read and explained to those participants that cannot read. The free and written informed consent of all interviewees was obtained. Participants unable to sign the consent form were allowed to thumb print, to affirm their consent. A random sample of 107 health facilities was drawn from approximately 500 public primary health facilities. Volume of births occurring in the previous six months in the 107 sampled primary health facilities was determined by reviewing their maternity registers, and the 10 primary health facilities with the highest volume of births in the state were selected for the study. The 10 selected facilities had an average of 15.7 births (SD 12.0) per month, which is higher than the state-level average of 4.3 births (SD 6.3) per month in primary health facilities [35]. The facilities were primary health facilities, providing all services (e.g. primary care, pregnancy care, labour and delivery services). Emergency care and complicated cases from these health facilities are referred to referral health facilities. All women giving birth in these facilities in July–August 2017, and who had a live newborn, were invited to complete an exit interview about events that occurred during their labour and delivery, and their perception of the care that health workers provided, including respectful care. To ensure confidentiality, the exit interviews were conducted in a separate room or area reserved for the interviews within the health facilities. In each of these 10 facilities, two trained data collectors and a supervisor were posted in shifts covering day and night deliveries, seven days a week for approximately four weeks. This was determined to be the amount of time needed in these high-volume facilities to recruit a sample of 320 births. The sample size was calculated based on the assumed prevalence of respectful maternity practices of 10–20%, a power of 80%, a 95% confidence interval. The study instrument was a structured questionnaire covering the demographic information of study participants, the content of care provided to the mother and the newborn, and respectful care during institutional birth. The study tool was operationalised based on revised typology by Bohren et al. [18] The tool had 31 items, structured around the seven domains of mistreatment proposed by Bohren and others: physical abuse (2-items), sexual abuse (1-item), verbal abuse (2-items), stigma and discrimination (1-item), failure to meet professional standards of care (6-items), poor rapport between women and health providers (11-items), and health system conditions and constraints (8-items) [18]. Additional file 1: Table S1 present questions used to assess mistreatment during institutional delivery in Gombe State. Women were asked to provide Yes or No response to the questions listed in Additional file 1: Table S1. This method is widely used in attitude measurement, for example, in Abuya et al. [23] and easily understood by respondents [38]. The study instrument was first reviewed and validated for content in collaboration with a group of health workers (doctors, nurses and midwives) working in Gombe. The tool was pilot tested within the same study health facilities with post-partum women. The feedback from the health workers and the pilot informed further refinement and finalisation of the study instrument. Data were collected using personal digital assistants, programmed in Census and Survey Processing System (CSPro), and took about an hour to complete. Descriptive statistics about the study sample of postpartum mothers and their reported experience of respectful maternity care were tabulated. An aggregate outcome variable on the report of any abuse was computed from each of the seven domains. We used the svyset command in Stata 15 to account for clustering at the facility level. In December 2017, IDIs and focus groups were conducted with consenting women who had recently delivered in a health facility in Kaltungo and Kwami LGAs. Similar to rest of Gombe State, Kaltungo and Kwami LGAs are multi-ethnic and mostly rural. Public health services account for almost all institutional deliveries in these LGAs, provided through 38 public and six private health facilities in Kaltungo LGA, and 42 public health facilities in Kwami LGA, where there are no private health facilities. Recruitment was purposeful and involved the lead researcher with the assistance of health facility staff identifying women who had delivered in a health facility, from the records of two health facilities in Kaltungo LGA and two health facilities in Kwami LGA, and whose infants were under 6-months of age. The homes of 64 eligible women were then identified with the assistance of community leaders specifically ward focal persons. The IDIs were conducted in the two most commonly spoken local languages in Gombe, Hausa and Fulfulde, and in English, using a pretested semi-structured interview guide. Out of the 64 women identified and recruited, 31 women – 15 in Kaltungo and 16 in Kwami – were interviewed in-depth before sufficient saturation level was achieved, determined in reflective meetings in-between interviews [39]. One of the women recruited for the IDIs in Kaltungo could not participate because she travelled out of town before the interview. The interviews were conducted in the women’s homes or in a private place of their choosing, by a trained female interviewer. Interview sessions lasted between 60 and 90 min, and were audio recorded, the research lead ensured the quality of the data collected from the IDIs through reflective meetings with the interviewer between the interviews. The interviews focused on problems faced by pregnant women, their motivations for giving birth in a health facility and for selecting a particular health facility, their attitudes towards health facility deliveries, their perceptions of quality of care and experience of mistreatment during institutional delivery. The IDIs were followed by focus group discussions (FGDs) to gain further insights into the women’s shared understandings of respectful care during institutional delivery and to gain group consensus around themes identified in the IDIs as reasons for giving birth in a health facility and perceptions of quality of care, including respectful care received. Four FGDs were conducted, two in Kwami LGA and two in Kaltungo LGA, with the remaining 32 women from the 64 originally identified. Eight women participated in each FGD with no stratification. For optimal results, 6–12 participants per focus group were recommended [40, 41]. The FGDs were also conducted by a trained female interviewer, assisted by the research lead. To encourage participants to speak freely the FGDs were conducted in neutral settings: empty primary school classrooms and community centre conference rooms. The FGD sessions lasted between 90 and 120 min [42]. The same trained female interviewer carried out both the IDIs and FGDs. IDI and FGD guides were used to ensure that all relevant issues were covered. Data collection took the form of field notes, supported by recordings. At the end of each data collection session, the sound recordings, field notes and consent forms, were stored securely. The recorded qualitative interviews were transcribed verbatim and translated from Hausa or Fulfulde into English. To ensure that the original meanings conveyed by the participants were fully captured the data collectors also carried out the translations. A thematic content analysis a form of qualitative analysis that allows the use of quantitative results as the basis for a priori themes was used to analyse the data, with a manifest approach [40], in which the data analysis focused on what women said of their experience during labour and delivery. The data analysis was carried out in three stages. First, familiarisation involving reading and re-reading the transcripts to aid understanding of the data. Second, organising and coding the data. The coding was determined a priori to align the qualitative findings to the quantitative results, to aid understanding how the quantitative findings were manifest. These were physical abuse, verbal abuse, sexual abuse, stigma and discrimination, failure to meet professional standards of care, poor rapport between women and providers and health system conditions and constraints. The coding was done using NVivo software version 12. Third, data from each code point were reviewed and summarised to reduce the number of words without losing the content or context of the text and to ensure themes were internally consistent. The a priori themes helped in identifying broad initial themes, further themes that emerged from the analysis were considered sub-divisions of those broad initial themes, but may also be standalone. The qualitative study findings were drawn from individual themes and sub-themes, and from exploring the relationship between themes. Some representative anonymised quotes of women’s own words were used to describe the manifestations of their experiences. The credibility of the data was determined by triangulating data between data collection methods.