Menu
Menu
Menu
Menu
Background: North Eastern Kenya has persistently had poor maternal, new-born and child health (MNCH) indicators. Barriers to access and utilisation of MNCH services are structural, individual and community-level factors rooted in sociocultural norms. A package of interventions was designed and implemented in Garissa sub-County aimed at creating demand for services. Community Health Volunteers (CHVs) were trained to generate demand for and facilitate access to MNCH care in communities, while health care providers were trained on providing culturally acceptable and sensitive services. Minor structural improvements were made in the control areas of two facilities to absorb the demand created. Community leaders and other social actors were engaged as influencers for demand creation as well as to hold service providers accountable. This qualitative research was part of a larger mixed methods study and only the qualitative results are presented. In this paper, we explore the barriers to health care seeking that were deemed persistent by the end of the intervention period following a similar assessment at baseline. Methods: An exploratory qualitative research design with participatory approach was undertaken as part of an impact evaluation of an innovation project in three sites (two interventions and one control). Semi-structured interviews were conducted with women who had given birth during the intervention period. Focus group discussions were conducted among the wider community members and key informant interviews among healthcare managers and other stakeholders. Participants were purposively selected. Data were analysed using content analysis by reading through transcripts. Interview data from different sources on a single event were triangulated to increase the internal validity and analysis of multiple cases strengthened external validity. Results: Three themes were pre-established: 1) barriers and solutions to MNCH use at the community and health system level; 2) perceptions about women delivering in health facilities and 3) community/social norms on using health facilities. Generally, participants reported satisfaction with services offered in the intervention health facilities and many indicated that they would use the services again. There were notable differences between the intervention and control site in attitudes towards use of services (skilled birth attendance, postnatal care). Despite the apparent improvements, there still exist barriers to MNCH services use. Persistent barriers identified were gender of service provider, insecurity, poverty, lack of transport, distance from health facilities, lack of information, absence of staff especially at night-time and quality of maternity care. Conclusion: Attitudes towards MNCH services are generally positive, however some barriers still hinder utilization. The County health department and community leaders need to sustain the momentum gained by ensuring that service access and quality challenges are continually addressed.
The study was conducted in three locations within Garissa sub-County. Iftin and Township were selected as intervention sites while Madina was the control site. In each of the three locations, there are community units (CUs) overseen and served by Community Health Volunteers (CHVs) who in turn report to the respective designated health facilities. Iftin location is served by Iftin sub-County Hospital, Township area is served by Sisters Maternity Home (SIMAHO) and Madina area is served by Madina Health Centre. The two intervention sites received the intervention package and the control site received the usual standard of care. The study utilized an exploratory qualitative design with participatory approach using focus group discussions (FGDs) and Key informant interviews (KIIs) as part of impact evaluation of the project. Purposive sampling procedures were adopted to identify and conveniently select a sample of opinion leaders (n = 15), women of reproductive age -who had childbirth experiences during intervention period (n = 53) and married men (n = 57) to provide data in different qualitative interviews. Health managers were purposively selected to submit further insights into expectant mother-ANC provider relationships and uptake of medical services. All the participants were approached by either email or trained research assistants to participate in the interviews. In total, 150 participants were approached and 25 refused to participate citing “lack of confidentiality” and “busy schedules”. A summary of all participants can be found in Table 1. Distribution of participants Some of the interviews were conducted by two of four authors (IK, RN, AAN & AKZ) who are researchers with a professional interest in MNCH. All have a health background and had experience with in depth interviews. Prior to the interviews, none of the participants were known to the interviewers, either personally or professionally. Initial interview guides were based on questions the researchers had identified after studying literature and then adjusted throughout the study as new themes arose. An interview guide containing structured and unstructured questions were applied to health professionals and opinion leaders. A similar semi-structured discussion guide was used for FGDs with community members to elicit in-depth community perspectives of the project interventions, the causes of poor MNCH outcomes, sociocultural beliefs and practices impacting the use of maternal and newborn health services and barriers to healthcare utilization. Data were collected between March 2019 and April 2019 and all participants gave verbal consent for their quotes to be used. All the interviews took place in a location chosen by the participants, which was most often either in their home or place of work and rarely in a public place like health facilities. All interviews were done in private. The interviews were allowed to take a spontaneous course and lasted between 45 and 60 min. A total of 14 FGDs and 15 KIIs were conducted. Table Table11 represents the type of interviews conducted and distribution of study participants using the different study methods. In each study site, a tested interview guide was used to obtain information primarily on: (i) views on women giving birth in health facilities; (ii) barriers in accessing and using MNCH services at the community and health facility level, and (iii) possible quality improvement at health facilities. All field interviewers were trained on the study rationale, the objectives, the study approach and the data collection procedures. Interviews were conducted in Somali by trained Somali native research assistants. A note taker worked alongside with the interviewer taking notes. In both KIIs and FGDs interviews, the research assistant let the participant respond and then probed where necessary to obtain more information before proceeding to the next topics. FGD interviews included women of reproductive age (who had ever given birth over the intervention period) and men (whose wives had delivered during the intervention period) at the community level, key informants included community leaders (religious, clan leaders), NGO personnel (NGOP), healthcare providers/managers (HM), traditional birth attendants (TBAs), political leaders MCA) and community health volunteers (CHVs). Two FGDs were conducted in each of the three study sites among married women aged 15–29 and 30 and above separately. Similarly, two FGDs were conducted with married men aged 15–29 years and the other with older men (30 years and above) in each site. We also conducted FGD among male champions. Each focus group discussion had 6–10 participants. Audio files were transcribed verbatim into English by native Somali speakers. The transcription was supplemented by field notes taken by note taker. The transcripts were marked according to the area where the interviews took place, date of discussion, type of group in terms of gender and role in the community. The transcripts were analyzed using content analysis, by reading through the transcripts to code important information. Two members of the research team reviewed and coded the transcripts. After coding the transcripts, the two members of the research team identified patterns from the coded data and made connections to recurrent themes and pre-established themes from the quantitative survey -such as views on health facility delivery, reasons for home delivery. Outcomes were compared for consensus. A discussion was held in case of mismatch and an agreement arrived upon to assign appropriate codes. Due to the fact that the target population share the same culture and views, data saturation occurred at the analysis stage. Themes were re-appearing in most transcripts, relaying the same information. Data were coded using QSR International’s NVivo 12 software to identify primary and meta-codes and major themes. The themes were identified with attention to contradictions and diversity of experiences, perception and attitudes across different interviewees. The coding frame agreed by both researchers was used to systematically assign the data to the thematic categories. Coding was undertaken by a single researcher. The interpretation was undertaken by at least two members of the research team to ensure objectivity and consistency of coded information. Data from different participant groups were analyzed separately and then compared for areas of convergence and divergence. Interview data from different sources (women, their partners, opinion leaders and their health care professionals) on a single event were triangulated to increase the internal validity of this study [23]. Analyzing multiple cases strengthened external validity. The following themes were pre-established: barriers and solutions to MNCH use at the community and health system level; perceptions about women delivering in health facilities and community/social norms on using health facilities.
N/A
