Socio-cultural and economic barriers, and facilitators influencing men’s involvement in antenatal care including HIV testing: a qualitative study from urban Blantyre, Malawi

Background: Male partner involvement in antenatal care (ANC) is associated with positive maternal and neonatal outcomes. However, only a handful of men attend ANC with their partners. This study aimed to understand the underlying barriers and facilitators influencing men’s ANC attendance including HIV testing in Blantyre, Malawi. Methods: Data were collected during a formative qualitative study of a cluster-randomised trial. Six focus group discussions (FGDs) with 42 men and women and 20 in-depth interviews (IDIs) were conducted at three primary health centres in urban Blantyre, Malawi. FGD participants were purposively sampled with IDI participants subsequently sampled after FGD participation. Thematic analysis was used to analyse the data. Results: The economic requirement to provide for their families exerted pressure on men and often negatively affected their decision to attend ANC together with their pregnant partners despite obvious benefits. Peer pressure and the fear to be seen by peers queueing for services at ANC, an environment traditionally viewed as a space for women and children made men feel treated as trespassers and with some level of hostility rendering them feeling emasculated when they attend ANC. Health system problems associated with overall organization of the ANC services, which favours women created resistance among men to be involved. An association between ANC and HIV testing services discouraged men from attending ANC because of their fear of testing HIV-positive in the presence of their partners. The availability of a male friendly clinic offering a private, quick, supportive/sensitive and flexible service was considered to be an important incentive that would facilitate men’s ANC attendance. Men described compensation to cover transport and opportunity cost for attending the clinic as a motivator to attending ANC services and accepting an HIV test. Conclusion: Peer and economic influences were the most influential barriers of men attending ANC and testing for HIV with their pregnant partners. Addressing these socio-economic barriers and having a male friendly clinic are promising interventions to promote male ANC attendance in this setting. This was an analysis of primary data from a formative qualitative study of a multi-arm, multi-stage cluster randomised trial from a study called Partner Assisted Self-testing and Linkage (PASTAL ISRCTN18421340) [25]. The study was conducted between 8 August 2016 and 30 June 2017 at three primary health centres (PHCs) of Zingwangwa, Ndirande and Bangwe located in urban Blantyre, Malawi. Of the three PHCs, Zingwangwa had the highest ANC throughput and offered ANC to first time attendees on Friday only, whereas Bangwe and Ndirande offered it on Monday only. Pregnant women attending ANC for the first time as well as male partners of ANC attending women, not necessarily as couples participated in focus group discussions (FGDs) and in in-depth interviews (IDIs). Qualitative data were collected through six focus group discussions (FGDs) involving a total of 42 ANC attendees and male partners, and 20 in-depth interviews (IDIs) conducted with ANC attendees and male partners who had participated in the FGDs. FGDs were used to enable researchers elicit collective views that represented community perceptions around the barriers and facilitators of male partners’ attendance to ANC through group discussions. The follow-on IDIs built on the FGDs and solicited more intimate-personal views using different although somewhat similar set of questions. This approach strengthened the quality of the data collected because data from FGDs provided a global view whereas data from IDIs provided a much finer view and potentially eliminated bias from group level influences in responses. Thus, combining FGDs and IDIs allowed triangulation of data through constant comparison of the results. Purposive sampling was employed with the aim of recruiting ANC attending pregnant women to participate in the FGDs and as a way of reaching out to their partners. A research staff member worked with the nurse on duty to identify and approach pregnant women for inclusion in the study. To be eligible, women had to be 18 years and above, and visiting ANC for the first time. Sampling was implemented during pregnant women’s routine visit to the ANC at the selected PHC (Fig. 1). While ANC attendees waited in a group at a waiting bay to access ANC services, the study team utilised the routine early morning health education talks to introduce themselves and the study by providing oral information on its purpose and the importance of their participation. Recruitment aimed to have between 8 to 12 participants for the FGDs and from each completed FGD, 4 participants for the IDIs. After accessing the ANC services, women were then approached individually and requested to voluntarily participate in the study. Women who showed interest were requested to provide an informed consent and recruited for the FGD which took place immediately after receiving the ANC service. Description of recruitment for focus group discussions and in-depth interviews Women who participated in FGDs were given a letter inviting their male partners to participate in the study. The partner’s phone number was obtained from the women during the group discussion. Male partners were contacted via a phone call after 5 o’clock in the evening as this was suggested by most women to be the most convenient time as they would have clocked off from work or business. During the phone call, eligibility screen was administered: aged 18 and above and being available in the catchment area in the next 28 days. Male partners who verbally consented were then informed about participating in FGD and IDIs, venue and time for the interviews. All male partners and women who participated in FGDs were eligible to participate in IDIs. Therefore, all FGD participants were approached and asked for oral consent to participate in IDIs. A random selection of the final IDI participants was drawn from all FGD participants who orally consented to participate in IDIs. Written consent was obtained from all participants before completing an IDI. All FGDs and IDIs were conducted either at the health centre for women or at a convenient place in the community (e.g. school) as preferred by participants. FGDs and IDIs were done in the local language (Chichewa) and were led by social scientists (DS and MK) who are both fluent in the local language of use. Participants received ~US$5 for their time. Data were mainly collected by two researchers, a senior social scientist (MK) with a PhD led FGDs while a research assistant (DS) with a BSc took the field notes during the FGD’s and led the IDIs (Fig. 2). Research participants had no relationship with the researchers although women had interacted with the researchers before engaging the male participants. A semi-structured question guide was used during the FGDs (Additional file 1) and the IDIs (Additional file 2) with embedded and flexible probes to prompt additional discussion and correct course of discussion. Both the FGD and the IDI guides focused on perceptions of the ANC service, perceptions of HIV testing by male partners during ANC, and perceptions of HIV self-test kits delivered to male partners at home by a pregnant woman. Notable differences between the FDG and the IDI guide were the individualized nature of the IDI questions and probes around whether or not the individual would accept to distribute (pregnant woman) or receive (male partner) HIV self-test kits provided during ANC. Consolidated criteria for reporting qualitative research (COREQ) checklist All FGDs and IDIs were recorded with digital recorders and the recorded qualitative data were transferred onto a computer, before translation and transcription. The FGD’s were in-between 1 h to 1 h 30 min, whilst the IDI’s lasted a maximum of 30 min. The FGD and IDI participants also completed a short questionnaire to collect their demographic data and these were analysed quantitatively (see Fig. 5 in results section). Apart from the participants and interviewers, there was no one present in the data collection rooms. Characteristics of all study participants by sex. Reproduced with permission from Choko et al. JIAS 2017 Nvivo version 10 was used for data organization and analysis following translation and transcription of the six FGDs and 20 IDIs. A coding framework was developed separately for FGDs and IDIs based on the question guides. Thematic analysis was used to analyse qualitative data [26]. Data analysts (MK, DS, AC) familiarised with the data by reading and re-reading transcripts to look for implied meanings and try to see the data in context [26]. Inductive and deductive coding was done by developing an initial coding framework based on the research objectives and also reading transcripts multiple times while attaching labels to data on the basis of meanings that the researcher discerned in the data [26]. Data were conceptualised by constant comparison of points raised by different people within and across FGDs and IDIs. Themes were developed by grouping several categories/codes that represented a unified subject/topic (Fig. 3). Data are presented as a descriptive narrative with quotes used to support each emerging theme. Summary of data analysis process The overall analysis was guided by the social identity theory which was introduced by social psychologists Henri Tajfel and John Turner in the 1970’s [24]. The theory is based on the idea that people categorise themselves based on the social groups to which they belong (Fig. 4). The theory has three processes namely; social categorization, whereby people are defined based on social categories rather than their individual categorization. The second process is social identification where people tend to behave in the way that they believe their group should behave. The third is social comparison where people compare themselves to another group in terms of having a higher or lower social standing. A number of factors were at play in the male decision-making process to go to ANC and test for HIV and these were critiqued in light of the social identify theory in this study. Most factors involved men categorising themselves in different groups affecting their decision to attend ANC and have an HIV test. Components of the theory one’s self. Source: Modified from Social Identity theory Tajfel (1971) British Journal of Social and Clinical Psychology