Background: Interventions to reduce maternal mortality have focused on delivery in facilities, yet in many low-resource settings rates of facility-based birth have remained persistently low. In Tanzania, rates of facility delivery have remained static for more than 20 years. With an aim to advance research and inform policy changes, this paper builds on a growing body of work that explores dimensions of and responses to disrespectful maternity care and abuse during childbirth in facilities across Morogoro Region, Tanzania. Methods: This research drew on in-depth interviews with 112 respondents including women who delivered in the preceding 14 months, their male partners, public opinion leaders and community health workers to understand experiences with and responses to abuse during childbirth. All interviews were recorded, transcribed, translated and coded using Atlas.ti. Analysis drew on the principles of Grounded Theory. Results: When initially describing birth experiences, women portrayed encounters with providers in a neutral or satisfactory light. Upon probing, women recounted events or circumstances that are described as abusive in maternal health literature: feeling ignored or neglected; monetary demands or discriminatory treatment; verbal abuse; and in rare instances physical abuse. Findings were consistent across respondent groups and districts. As a response to abuse, women described acquiescence or non-confrontational strategies: resigning oneself to abuse, returning home, or bypassing certain facilities or providers. Male respondents described more assertive approaches: requesting better care, paying a bribe, lodging a complaint and in one case assaulting a provider. Conclusions: Many Tanzanian women included in this study experienced unfavorable conditions when delivering in facilities. Providers, women and their families must be made aware of women’s rights to respectful care. Recommendations for further research include investigations of the prevalence and dimensions of disrespectful care and abuse, on mechanisms for women and their families to effectively report and redress such events and on interventions that could mitigate neglect or isolation among delivering women. Respectful care is a critical component to improve maternal health.
In Tanzania, the maternal mortality ratio is 454 deaths for 100,000 live births. One in 38 women have a lifetime risk of death due to maternal causes [27] and for every 1,000 births, 4–5 women die from pregnancy-related causes [28]. Nationwide, 50.2% of births are facility-based and 50.6% of all births are in the presence of a skilled attendant [28]. Since the early 1990s, the national rate of facility-based birth has remained below 52.6% [28, 29]. In rural areas, less than half of births are facility-based (41.9%) and 42.3% of all rural births are in the presence of a skilled attendant [28]. This study was based in 16 villages across 4 districts of Morogoro Region, in eastern Tanzania. Compared to national averages, slightly more women in the region deliver in a facility (58%) and more births are attended by a skilled provider (60.6%) [28]. Throughout the country’s Eastern Zone, which encompasses the region, hospitals and health centers are ill equipped to provide basic or comprehensive emergency obstetric care (EmOC). Basic EmOC is available in 11% of facilities and comprehensive EmOC is available in 10% of facilities [30]. In terms of personnel, facilities in Morogoro Region are understaffed, which reflects national trends. The Region’s density of doctors (0.2), assistant medical officers (0.3) and clinical officers (2.1) per 10,000 people attests to severe human resource limitations [30]. Less than half of all facilities in the Zone (47%) have at least 2 qualified providers assigned to a facility to support basic emergency services 24-hours [30]. Supportive management practices, which are critical for supporting quality care, are also limited. While many facilities in the Eastern Zone receive an external supervisory visit (79%), 34% of facilities provide routine staff training and only 25% of facilities provide “supportive management practices” (an external supervisory visit, routine training and personal supervision) [30]. This qualitative, cross-sectional study employed in-depth interviews (IDIs) with women, their male partners, community health workers (CHWs) and community leaders. At eight health centers across four districts, health center staff were asked to identify one village with difficult access to the health center, yet within the center’s catchment area. The data collection team then presented the study to leaders in both the village encompassing the health center and the village described as having difficult access. In Tanzania, the long-standing policy has been for every village to have a village health committee, which appoints two CHWs. Leaders interviewed included religious leaders, as well as members of an elected village board and/or village health committee who identified CHWs. Leaders and CHWs were interviewed irrespective of gender, age, education level, or length of service. Leaders as well as CHWs helped identify women in the village who had delivered in the preceding 14 months. In addition, data collectors canvassed the village and invited eligible mothers and fathers to participate. For a breakdown of respondent groups by distance to facility and district, see Table 1. Respondent groups by distance to facility and district *Women who delivered a child within the preceding 14 months. **Includes any male partner regardless of legal marriage status. Women and their partners were eligible if they had delivered a baby within the preceding 14 months regardless of reports on quality of care, or experiences of disrespectful care. An emphasis was placed on identifying women who had non-complicated, normal deliveries. Women who reported severe vaginal bleeding, eclampsia, obstructed labor, retention of placenta, severe anemia or whose births required vacuum or forceps extraction, or cesarean section were not included with the rationale that such births alter not only careseeking behaviors (often necessitating referrals) but also entail a vastly different subjective sense of the birth experience. For discussion on how a birth experience alters later assessment of quality of care (described as “fulfillment theory”), see Bramadat [31]. All women providing consent were interviewed, until 2–4 women had been interviewed for that site. Five Tanzanian research assistants fluent in Swahili with graduate-level training in education, public health, and social sciences were trained for five days to collect the data using instruments, which were pre-tested and revised before starting interviews. Training topics included maternal and newborn health, interview techniques, research ethics and qualitative methods. IDIs were recorded and conducted one-on-one, in a private place of the respondent’s choosing following verbal consent. IDIs focused on experiences related to care seeking during a most-recent pregnancy and birth. At the outset of data collection, the research team did not intend to explicitly investigate experiences of abuse, but rather to explore careseeking for birth in facilities. The abuse theme emerged in the earliest interviews, however, and was probed more explicitly as data collection progressed. A supervisor conducted daily debriefing sessions with data collectors to discuss and triangulate key findings, refine lines of inquiry, and identify saturation of themes. A main product of these debriefings were memos, first generated as a version of meeting notes from debriefings and later amplified by the data collection supervisor to incorporate reflexive notes, contextual information and emerging understandings that could be shared and commented upon by the wider research team. Data collection lasted approximately two months during July and August 2011. In-country debriefings with national stakeholders following the close of data collection corroborated and refined the framework for thematic analysis. All interviews were recorded and transcribed into Swahili. An initial phase of open, inductive coding on a selection of rich, diverse and representative transcripts was conducted based in part on Grounded Theory [32]. This resulted in the creation of a codebook that was validated by co-authors. A co-author fluent in Swahili and English applied these broad codes to remaining transcripts using ATLAS.ti [33]. Coded data were then translated from Swahili to English and a second phase of detailed coding was undertaken by a social scientist. During the analysis process, a subset of co-authors discussed codes and themes, and drew comparisons across respondent groups and regions, and by distance to facility. This aided in triangulation of findings and provided texture and nuance to descriptions. Drawing on the principles of Grounded Theory, a literature review followed the completion of coding [32]. The study received ethical approval from the Muhimbili University of Health and Allied Sciences and Johns Hopkins School of Public Health Institutional Review Boards. Names used in this paper are pseudonyms to protect the privacy of interviewees.
N/A