Access barriers to obstetric care at health facilities in sub-Saharan Africa—a systematic review

We searched the online databases PubMed, CINAHL, and Scopus. Search terms used to locate relevant articles included ‘obstetric care’ with ‘access’, ‘barriers’, ‘developing countries’, ‘pregnancy’, ‘morbidity’, ‘mortality’, ‘haemorrhage’, ‘eclampsia’, ‘sepsis/infection’, ‘obstructed labour’, ‘abortion-related complications’ and ‘sub-Saharan Africa’ (Additional file 1). Articles that met the inclusion criteria were primary research studies that examined barriers to obstetric care, targeted women/service users accessing such care as well as maternity care workers. Studies were also published in a peer reviewed scientific journal in English between 2000 and 2015 and conducted in sub-Saharan Africa. Studies which employed quantitative, qualitative or a mixed-methods design were included. Articles were excluded if the reports were based on secondary data analyses or if data regarding obstetric care barriers is not extractable from the text.

The study selection stage, which involved screening of titles and abstracts and retrieval of full texts, was carried out by one author (MKN). Full-text articles were extracted and assessed against the inclusion and exclusion criteria. Selected full-text articles were re-evaluated for data extraction and assessed for quality.

The quality of included studies was assessed using the mixed methods appraisal tool (MMAT) by Pluye and colleagues [12]. The tool was suited for this review as it was specifically developed for quality appraisal in systematic reviews involving qualitative, quantitative and mixed-methods designs. MMAT is reported to have an inter-rater reliability score ranging from moderate to perfect [13] and has been applied by other researchers [14‐16] in mixed-methods systematic reviews. There are five sections in the criteria which include qualitative, randomised controlled, non-randomised, descriptive and mixed-methods studies. Qualitative and quantitative sections have four criteria each, and studies are scored by dividing the number of criteria met by four to arrive at a value ranging from 25 to 100%. For studies with mixed-methods designs, the overall quality score is the lowest score of the study components [17]. All studies were included regardless of their quality ranking since the focus of the review was to examine the context in which barriers to obstetric care occurs (Additional file 2).

Given the range of the outcomes in the studies, identified barriers to obstetric care has been summarised using narrative synthesis. The Popay et al. guidance on the conduct of narrative synthesis in systematic reviews was employed in the synthesis [18]. The elements of the narrative synthesis process are developing a theory, developing a preliminary synthesis, exploring relationships within and between studies, and assessing the robustness of the synthesis. No new theory was developed as part of this review; instead, a pre-existing analytical framework [11] was adopted to facilitate organisation and interpretation of the data. The subsequent step involved the creation of a large table for the extraction of relevant data such as author, year of publication, country, study design, sample characteristics, study objectives, data analysis, major findings and quality assessment (Additional file 3). This important stage was iterative and involved studying the articles, taking notes and making initial comparisons so as to gain familiarisation with the data. Quality appraisal of studies was concomitantly carried out. From the table, the study findings were further examined and using thematic analysis, major themes were coded under pre-defined categories in the analytical framework by Jacobs et al. [11] (Table 1). Emergent themes were also aggregated under the label ‘other barriers’. The underlying methodology applied within the thematic analysis approach was the essentialist or realist method, which is based on the experiences, meanings and the reality of participants [19]. Subsequently, relevant information was drawn out and relationships between the studies were more comprehensively described under the discussion, to enhance interpretation of the review data. Greater attention was paid to differences and similarities between studies as regards the settings, populations, outcomes of interest, methodological approaches and how these might have been reflected in the results. Finally, a critical reflection on the narrative synthesis process was undertaken as regards the quality of the primary studies reviewed and strengths and limitations of this systematic review. In addition, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist was followed to enhance the quality of reporting [20] (Additional file 4).

The current review was conducted as part of a PhD (by publication) study by MKN. While MKN conducted the data screening and extraction, quality assessment and data synthesis, she did so under the supervision of MCO and TVM, two experienced researchers. All authors had primary responsibility for the development of research questions and study design, and for the intellectual content of the paper. MCO and TVM provided direction and supervision for all aspects of the work and ensured that questions relating to the accuracy or integrity of the work were investigated and resolved. They critically reviewed the methods and made revisions, as required. They also reviewed and discussed the findings and reviewed and edited each draft of the final document.