A socio-ecological approach to understanding the factors influencing the uptake of intermittent preventive treatment of malaria in pregnancy (IPTp) in South-Western Nigeria

Malaria in pregnancy (MiP) remains a key cause of poor maternal and neonatal health outcomes, particularly in the African region. Two strategies globally promoted to address MiP require pregnant women in malaria-endemic regions to sleep under insecticide-treated bed nets (ITNs) and take at least three doses of intermittent preventive treatment (IPTp) during pregnancy. Yet, several multilevel factors influence the effective uptake of these strategies. This study explored the factors for the poor uptake of IPTp and use of ITNs in lower socioeconomic communities in Nigeria. We conducted semi-structured interviews (SSI) and focus group discussions (FGD) with a total of 201 key stakeholders in six communities in Ogun State, South-Western Nigeria. Twelve SSIs were conducted with traditional birth attendants (TBAs), faith-based birth attendants and healthcare providers operating in public health facilities. Community leaders (7), pregnant women (30) and 20 caregivers were individually interviewed. Sixteen FGDs were conducted with multi- and first-time pregnant women grouped by location and pregnancy experiences. A thematic approach was used for data analysis. At the individual and social levels, there is a high general awareness of MiP, its consequences and ITNs but low awareness of IPTp, with type of antenatal care (ANC) provider being a key factor influencing access to IPTp. The choice of ANC provider, which facilitates access to IPTp and ITNs, is influenced by the experiences of women, relatives and friends, as well as the attitudes of ANC providers and community perceptions of the type of ANC providers. Concurrent use of multiple ANC providers and ANC providers’ relationships further influence acceptability and coverage for IPTp and ITN use. At the health sector level, there is low awareness about preventive malarial strategies including IPTp among TBAs and faith-based birth attendants, in contrast to high IPTp awareness among public healthcare providers. The findings highlight several factors that influence the utilisation of IPTp services and call for greater synergy and collaboration between the three groups of healthcare providers towards enhancing access to and acceptability of IPTp for improving maternal and child outcomes. This was a multi-site cross-sectional qualitative study, using semi-structured interviews (SSI) and focus group discussions (FGDs) to collect primary data from 201 participants in six semi-urban and rural communities in Ogun state between February and March 2019. Initial study questions which guided data collection were developed using grounded theory [12, 13] as part of a larger study on determinants of malaria prevention during pregnancy in Ogun state, Nigeria. Conducted following the consolidated criteria for reporting qualitative research [14] (S1 File), all data was collected as digitally recorded audio files by 14 Research Assistants (RAs) with at least two years contextual research experience. The authors developed the topic guides (S2 File) for data collection based on existing literature on MiP interventions in Africa and Nigeria specifically and drawing on their own conceptual experiences working on maternal and child health issues in the study areas. Interviews and discussions were conducted in private locations, usually identified by the participant (s) at his/her convenience on a scheduled date after verbal and written/thumb printed consent was sought from each participant. Ethical approval for the study was received from the Ethics Review boards of Babcock University (BUHREC056/19) and the Ogun State Ministry of Health (HPRS/381/290). The research team had no prior contact with study participants. The study was conducted in three semi-urban and rural communities within Ogun State, southwest Nigeria. The state borders Lagos State to the south, Oyo and Osun states to the north, Ondo to the east and the Republic of Benin to the west. Abeokuta is the capital and largest city in the state. Ogun state occupies an area of 16,981 km2 and is divided into three geopolitical zones (East, West and Central), with 20 administrative units known as local government areas (LGAs); each LGA consists of several administrative wards. Three LGAs (Ijebu-Ode, Sango-Ota, Odeda) were randomly selected from each geopolitical zone after which one semi-urban and one rural community was purposively selected from each LGA. From Ijebu-Ode LGA, Porogun was selected as the semi-urban community and Itamapako as the rural community. Sango was selected as the semi-urban community and Ketu as the rural community in Sango-Ota LGA. In Odeda LGA, the semi-urban community selected is Obantoko, while the rural community is Odeda. The urban areas consist of many public and private health institutions, with many of the community members engaged in the formal sector. There is primary, secondary, and tertiary level of healthcare services provided within the semi-urban communities. However, traditional birthing homes which are often managed by traditional birth attendants (TBAs) also exist. In contrast, rural areas consist of mostly petty-traders and farmers. A few government offices are situated in rural areas with limited primary and secondary schools available. Only primary healthcare facilities are often available within rural communities, usually managed by a nurse and mid-wife. There are many TBAs in rural areas with high patronage from the community. SSIs were used to explore in-depth, perceptions and experiences regarding uptake of ITN and IPTp in six rural and semi-urban communities in Ogun West, Central and East districts. We purposively selected public healthcare providers (4), TBAs (4) and faith-based birth attendants (4) based on popularity and accessibility within selected communities, to explore access and awareness surrounding IPTp in a pluralistic health sector. We further purposively contacted community leaders (7) and caregivers, mainly, family members (20), through participating pregnant women (30), to enable us to explore the influence of community and family networks on IPTp access and uptake. A sample size of between 20 to 30 interviews in qualitative studies permits data saturation where the key themes are addressed and additional interviews do not add new themes [14–16] and so a predetermined sample size of 69 SSIs was considered sufficient to reach data saturation. SSIs were conducted in the local language (Yoruba) and/or English with each interview ranging from 30 to 65 minutes. Sixteen FGDs were conducted with 132 pregnant women attending ANC, grouped by location and ANC provider (TBA, faith-based birth attendants and public healthcare providers) to gain insights into their collective pregnancy experiences with malaria prevention strategies. RAs recruited seven to eleven pregnant women at ANCs for FGDs with discussion time ranging between 40 to 76 minutes. Study information was given verbally and information sheets clearly outlining the purpose of the research presented to participants. Two trained RAs moderated the discussions in the local language using a topic guide and took notes. Because of the iterative nature of the study, data collection, translation, transcription and analysis were concurrent to enable the exploration of emerging themes. Six native speakers transcribed digitally recorded interviews and discussions into Yoruba, which were then translated into English. Co-authors reviewed all transcripts and translations to ensure the quality and accuracy of the translation. Transcripts were not returned for participant crosschecking due to low literacy rates. A thematic approach was used in data analysis with the initial coding framework generated in QRS Nvivo 11 pro by the first three authors using pre-identified themes derived from the initial topic guide with emerging themes. Transcripts were coded in constant comparison and codes reviewed for contextual relevance. Patterns and linkages between quotes, codes, themes and existing literature were explored in-depth to identify areas of convergence and divergence.