Malaria intermittent preventive treatment in Nigeria: a qualitative study to explore barriers

This study was conducted in Kano state, located in North Western Nigeria. It is the second most populous state in the country with an estimated 13.4 million people. The state was selected for the study due its high level of malaria burden compared to other regions [12]. For example, malaria prevalence in the state is approximately 32.4% [11]. In addition, maternal mortality ratio in the region is high, i.e., about 1025 per 100,000 births [15].

We employed snow ball and purposive sampling strategy to select participants for face-to face in-depth interview and Focus Group Discussions (FGDs). Interviews were conducted with 14 malaria stake holders at the national and state level. The maximum variation sampling approach was used to capture responses from varied participants, thereby enhancing the comprehensiveness of our findings [16]. Participants, mainly, policy makers in malaria prevention, health workers, malaria experts and community heads were recruited from three local governments in the states. This includes Nassarawa, Fagge and Kano municipal respectively. Six FGDs were conducted – 3 FGDs were held with pregnant women and 3 FGDs with married men. Each focus group consisted of about 8–10 participants. Data for health workers and pregnant women was gathered across three hospitals – all stationed at different communities. This enabled us to capture wide range of perspective from participants irrespective of their socioeconomic status and geographical location. The selected hospitals were: 1) Aminu Kano Teaching Hospital, 2) Abdullahi Wase Specialist hospital, and 3) Murtala Mohammed Specialist Hospital. Interviews were conducted by FM. The three FGDS for married men and the in-depth interviews for the community heads were conducted in the communities within the three local governments unlike the other interviews conducted in the hospital. Community contact persons assisted in recruiting the eligible participants for the FGDs. The FGDs were conducted in a venue arranged by the community heads (Table 1).

All interviews were audio recorded. Field notes were also taken where necessary. Separate pre-tested semi- structured interview guides were used for data collection during both the key informant interviews and FGDs. Interview questions covered: challenges of IPTp policy implementation, attendance of women for ANC, IPTp distribution in ANC units, accessibility of IPTp in the communities, knowledge about adverse effects of malaria during pregnancy, facilitators and barriers of IPTp usage. Interviews lasted between 20 and 45 min – when saturation was reached [17].

Data was transcribed verbatim. The conventional content analysis approach was used to interpret meaning from the content of data due to the inductive nature of this qualitative design [18]. Some of the transcripts were translated from local language to English and verified by experts to enhance accuracy. Transcripts were analyzed by FM and SN using a coding scheme developed from the topics. Data were thoroughly examined by researchers, and relevant perceptions were coded as concepts. Codes related were then summarized to form categories. Next, RM and MP developed the categories, sub-category and the themes. AA, HS and the whole research team ensured that the themes were linked to research questions and study objectives. The themes were compared and discussed to enhance understanding of data. MAXQDA 10 software was used for data management and analysis.

To enhance the rigor of the study, several researchers independently assigned pre-specified codes to the data [19]. First, five interviews were double-coded by FM. The remaining interviews were double-checked by SN. Next, the entire research team thoroughly discussed the findings for completeness and accuracy. Following that, some data were returned to the interviewees to check the correctness of interpretations. Interviews and data analysis were done at the same time. This enabled us to explore alien responses that emerged during interviews [16].

The study captured responses from experts with wide range of experiences and background, thereby enhancing the comprehensiveness of our findings. Again, we used various data collection strategies including in-depth interviews, focused group discussions and review of secondary data, which help enhance methodological rigor. Our research team comprises of different researchers with varied backgrounds including epidemiologists, health economist and policy makers. This enabled us to address reflexivity, i.e., individual beliefs, judgements and practices that might have influenced the interpretation of our findings. Still, the study was not without limitation as participants might have had underlining rationale to the responses due to their political and social affiliations. Hence, our findings should be interpreted with caution.