Background: Twenty years after acknowledging the importance of joint responsibilities and male participation in maternal health programs, most health care systems in low income countries continue to face challenges in involving men. We explored the reasons for men’s resistance to the adoption of a more proactive role in pregnancy care and their enduring influence in the decision making process during emergencies. Methods: Ten focus group discussions were held with opinion leaders (chiefs, elders, assemblymen, leaders of women groups) and 16 in-depth interviews were conducted with healthcare workers (District Directors of Health, Medical Assistants in-charge of health centres, and district Public Health Nurses and Midwives). The interviews and discussions were audio recorded, transcribed into English and imported into NVivo 10 for content analysis. Results: As heads of the family, men control resources, consult soothsayers to determine the health seeking or treatment for pregnant women, and serve as the final authority on where and when pregnant women should seek medical care. Beyond that, they have no expectation of any further role during antenatal care and therefore find it unnecessary to attend clinics with their partners. There were conflicting views about whether men needed to provide any extra support to their pregnant partners within the home. Health workers generally agreed that men provided little or no support to their partners. Although health workers had facilitated the formation of father support groups, there was little evidence of any impact on antenatal support. Conclusions: In patriarchal settings, the role of men can be complex and social and cultural traditions may conflict with public health recommendations. Initiatives to promote male involvement should focus on young men and use chiefs and opinion leaders as advocates to re-orient men towards more proactive involvement in ensuring the health of their partners.
The study was carried out in the East and West Kassena-Nankana Districts in the Upper East Region; the poorest region of Ghana. The Kassena-Nankana District (KND) has a population of about 150,000 with a population growth rate of less than 1% [24, 25]. There are two major ethnic groups – the Kassena and the Nankani. The two ethnic groups largely share a homogeneous socio-cultural system although there are some minor differences in cultural practices and festivals. Polytheism is common with animism co-occurring with various denominations of Christianity and Islam. Illiteracy is high 55% within the KND especially among women [26]. The district is largely patriarchal and male dominance is ensured through marriage customs such as the patrilineal inheritance and polygamy. Individuals live in relatively isolated compounds made up of physically linked household units headed by males with absolute authority over members. This gives men control over the use of family resources and decisions about health seeking [27]. Villages which are made up of several scattered compounds are headed by traditional chiefs. Political governance is through a district assembly with elected representatives from sections within the villages. Ten paramount chiefs and several divisional and sub-chiefs with their elders, zealously guard the cultural traditions of the people. As gate keepers, chiefs play a critical role in implementing health programs within their villages. Maternal health programs therefore require the support of the traditional hierarchy in order to facilitate men’s understanding of women’s health and support the empowerment of women to make decisions regarding their health. In terms of health care, public health facilities have been strategically located throughout the district to improve geographical access to services. The community has access to 33 community health compounds with resident community health nurses, six health centres, two public and four private clinics [28, 29]. The only hospital in the district is located in the urban centre and serves as a referral facility for all cases in need of emergency obstetric care. Financing of health services in the district is largely through a district mutual health insurance scheme and all maternal healthcare services are provided free within accredited health facilities. This research was part of a bigger study which used data from a demographic surveillance system run by the Navrongo Health Research Centre to identify severe maternal morbidities within the community. The surveillance system which covers the study district, collects routine data on marriages, pregnancies, births, deaths and migrations every 120 days [25]. The surveillance platform also has an address system that facilitates the tracing of research participants in the community. The study was therefore sited within the surveillance system in order to facilitate the tracing of severe maternal morbidities. A qualitative research approach was used to gain an understanding of men’s perception of their role in maternal health. Opinion leaders who were mostly male and health professionals who have direct contact with community members shared their views through focus group discussions and in depth interviews respectively. The bigger study was carried out as a PhD project [22, 23, 30]. The main study obtained local explanations for severe maternal morbidities from interviews with traditional healers and traditional birth attendants and used that explanation to screen over 900 women who recently gave birth in order to identify those who suffered severe maternal complications based on the local definition. Women who suffered severe maternal complications responded to an audit tool with an open narrative section that enabled to collection of qualitative data. The results from the audit which had a community focus were shared with opinion leaders within the research setting and the reactions of the community leaders regarding the role of men in maternal health are shared in this paper. Specifically, the opinion leaders offered their views on the role of men in delaying access to an appropriate place of care, lack of support for their partners during pregnancy, delays in initiating antenatal care and challenges with complying with drug regimes and nutritional requirements. Communities were purposively selected based on their relative isolation from the district hospital. This was to explore the influence of geographical access to the referral centre for emergency obstetric care. In all, 10 communities beyond a 15 km radius of the district hospital were selected. Selection of opinion leaders (chiefs, elders, assembly members, leaders of women groups) was done through the village chiefs. The study team requested each chief to invite between 10 and 12 opinion leaders to discuss issues related to maternal health. The Ghana Health Service is implementing a primary health care policy which makes community participation led by the Chiefs, an integral part of the implementation strategy. Beyond that, the Navrongo Health Research Centre (NHRC) has been carrying out research in the study district for over 25 years and as part of the requirements for initiating any research project, researchers have to conduct community entry activities which involves meetings with chiefs and their elders to discuss the research project and to seek permission to implement the study in their communities. Also, each of the chiefs has been involved in research at some point during the period of NHRC’S existence but this is rare. Apart from the current study, the only other study that directly involved opinion leaders as respondents was one that involved the lead author. That study determined the appropriateness of community engagement strategies as used by the research Centre [31]. In-depth Interviews (IDIs) were held with health professionals whose work demands regular interaction with opinion leaders or the provision of direct maternal health services to community members. Four members of the district health management team (public health nurses and District directors of health) and four staff of the health centres. These categories of professionals were considered key stakeholders in maternal health in the various villages. We also interviewed eight community health officer-midwives. Community health officer-midwives are nurses who have been trained to proficiency in midwifery skills and placed in health compounds within the community to offer basic obstetric care to women. The student researcher who has over 14 years experience in conducting qualitative research, assisted in conducting the 16 in-depth interviews with the health workers. The researcher also moderated ten Focus Group Discussions (FGDs) with the opinion leaders. The discussions with opinion leaders were conducted in the local languages – Kasem or Nankani – while the interviews with the health workers were in English (Additional file 1). Each focus group discussion lasted about 2 hours and the in-depth interviews, 1 hour. All the interviews and discussions were audio recorded and labeled with unique identifiers. For ethical reasons, quotes from the in-depth interviews are attributed generically to “health workers”. The study guides were pre-tested to ensure that they elicited the right information to meet the research objectives. Data collection lasted from March to November 2012. The discussions focused on gender specific concerns that were raised in the larger study. Male support for women over the duration of pregnancy and the opinions of males on broader gender dynamics during pregnancy were explored. Specifically, men’s enduring role in making decisions related to place for delivery or health care when complications occur were discussed. Based on the sampling approach, only 10 communities qualified to participate in the study. Considering that we had an average of 10 people participate in each focus group discussion, it meant about 120 individuals participated in the focus groups. This was in addition to the 16 IDIs with the health workers. Although there is currently no consensus on how saturation is reached for one to stop interviewing, we think the number of interviews we conducted were enough for us to achieve saturation. Creswel suggests 20 to 30 interviews [32] while Morse suggests 30 to 50 [33] but they both failed to explain why these number of interviews and not any other. In studies with a high level of homogeneity like the current study, a sample of six interviews is deemed sufficient to enable the development of meaningful themes and useful interpretations [34]. All the focus group discussions were transcribed into English. In transcribing the discussions, culture specific terminologies were retained and transcribed verbatim in the local languages. Transcripts and field notes were imported into QSR NVivo 10.0 software [35] for thematic analysis. Initial coding was based directly on the research questions. Themes were derived inductively by reading every sentence in all transcripts, identifying answers to repeated questions and naming them, and segmenting the data into similar groupings to form preliminary categories of information about male involvement in maternal health. Segments of texts in the respondents’ own words and expressions relating to the themes were extracted and labeled. Subsequent coding was based on emergent themes. Patterns, similarities and differences in these codes and themes were examined. This was to ensure that negative and deviant views were considered in the research report. Observations and field notes were coded together with the main transcripts and used to provide further contextual information.