We conducted a formative exploratory cross sectional descriptive study using a qualitative approach from February to August 2016. The design enabled the researchers to have a deep understanding of information that couples want during antenatal education . We employed focus group discussions (FGDs), in-depth interviews (IDIs) and key informant interviews (KIIs). Focus group discussions allowed the research team to generate contextualized multiple viewpoints on information needs of couples during antenatal education . In depth interviews provided detailed information on desired topics for couple antenatal education whereas key informant interviews allowed the generation of rich information based on the participant’s expertise [29, 31]. We conducted the following interviews; four focus group discussions among men and women independently plus an extra focus group composed of nurses/ midwives, 13 key informant interviews, 10 in-depth interviews with couples and 10 separate in-depth interviews with men who had been to antenatal clinics before with their spouses. The sample sizes for FGDs, KIIs and IDIs were guided by literature on reaching theoretical saturation [32,33,34]. Furthermore, the sample sizes allowed the researchers to recruit participants for FGDs, IDI and KII based on key demographic variables that were likely to have an impact on participants’ view on content for couple antenatal education [29, 33].
We followed RATS guidelines in presenting the manuscript including the results of the study (see Additional file 1: RATS Checklist).
The study was conducted in Blantyre District situated in the Southern region of Malawi in the following sites: Mpemba, South Lunzu, Queen Elizabeth Central Hospital, Blantyre District Health Office and Kamuzu College of Nursing.
Mpemba and South Lunzu (SL) health centres and their catchment areas are located in the southern and north-eastern parts of Blantyre District respectively. The District is divided into rural and urban settings. The urban setting is regarded as the main industrial city of Malawi as it contains industries and companies which provide employment. Mpemba and SL health centres serve both rural and urban communities. About 100 to 120 new pregnant women report for antenatal care at SL while Mpemba registers 80 to 100 per month. In both facilities antenatal services are provided free of charge. Additionally, antenatal education is an integral part of antenatal care and topics which are frequently discussed during antenatal education include birth preparedness and complication readiness planning focusing on items to prepare and prevention of mother to child HIV transmission (PMTCT). The topic of nutrition for a pregnant woman was discussed sporadically during antenatal education. Few men accompany their wives for antenatal services in both facilities according to anecdotal reports by maternity unit nurse/midwife in-charge at the facilities. Mpemba and SL health centres and their catchment areas were chosen because the centers are semi-urban and the views on content for couple antenatal education participants shared during interviews were likely to represent those of rural and urban populations of Blantyre District.
Queen Elizabeth central hospital
Queen Elizabeth Central Hospital (QECH) is the largest public tertiary hospital in Malawi and functions as the main referral hospital for the Southern Region. It is also a teaching hospital for different health related professions. On average about 100 women report to QECH for antenatal care per month (Personal communication by nurse/midwife in charge). The facility offers antenatal care on paying basis. Few men escort their spouses for antenatal care at the facility. During antenatal education, different topics are discussed with much emphasis on prevention of mother to child HIV transmission, birth preparedness and developing a complication readiness plan, family planning, exercises and nutrition during pregnancy. The hospital was chosen to provide adequate professional nurses/midwives for FGDs, as SL and Mpemba health centres could not.
Blantyre District health office and Kamuzu college of nursing
Blantyre District Health Office is responsible for managing health care services in all the health facilities situated in Blantyre. Kamuzu College of Nursing (KCN) is one of the institutions in the country which trains professional nurses/midwives and has lecturers experienced in midwifery education and practice. These institutions have individuals with different expertise in maternal health, including policy making, and can provide diverse opinions on preferred topics to be covered during antenatal education.
Rangers security service company and plant and vehicle hire engineering services
Rangers Security Service Company (RSSC) and Plant and Vehicle Hire Engineering Services (PVHES) are non-healthcare settings which were used in this study. The former is privately owned institution while the latter is government owned. Most of the employees are unskilled male labourers whose partners access maternity health services from the public health facilities of Blantyre District. The sites were chosen because they had participants with different backgrounds, which broadened the understanding on antenatal education information needs for couples.
Selection and recruitment of female participants for FGDs
Convenient sampling was used to recruit women attending antenatal clinics at SL and Mpemba health centres and the total number of the participants was 34. The principal investigator and a female research assistant approached the women in the waiting areas of antenatal and under five clinics. The women were informed of the study and those willing to participate were requested to remain after the services for informed consent procedures and discussions. Recruiting participants through the clinics provided an opportunity to observe antenatal education sessions and the topics taught to women. Women were divided into two age categories; the first category being those aged 18 to 25 years while the second category being aged 26 years and above. In total, there were 17 women less than 26 years of age and 17 greater than 26 years from Mpemba and South Lunzu.
The inclusion criteria for women to participate were: willingness to participate in the focus group discussions; expectant or had a child under five years during the period of the study; and were from the catchment areas of Mpemba or South Lunzu. Recruiting women with a child under five years meant the participants would be able to remember the childbirth experience. The women also varied in parity to determine similarities and variation on the subject matter. The discussions were conducted in a room at the health facilities. Four focus group discussions were conducted in total which were divided by the two age categories at each health facility.
Selection and recruitment of male participants for FGDs
Purposive sampling was used to recruit men for FGDs with the help of health workers. Purposive sampling provided an opportunity to choose participants who had the knowledge and experience on the subject matter as they were likely to contribute to the study purpose [29, 35].
Variations such as place of identification, number of children, education and age were considered during the recruitment process. The men were classified into two groups: young men between 18 and 25 years and older men above 25 years. The first category comprised 16 participants (six from SL and 10 from Mpemba) and the second category had 19 participants (10 from SL and nine from Mpemba). The inclusion criteria for men were: willing to participate in the focus group discussions, able to give in consent, having an expectant spouse, having a child under five years during the period of the study, resident of the catchment areas of Mpemba or South Lunzu; and were above 18 years old. Just like women, the men were included in this study because they would be beneficiaries of the forthcoming intervention. All the participants who were approached for the FGDs accepted. However, two male discussants, one from SL and one from Mpemba did not appear for the discussions and did not give any reason for their absence. In total, there were four FGDs conducted, one for men ≤25 years of age and another for men > 25 years at each site. We agreed with the male participants on the appropriate date, time and venue for the discussions.
Selection and recruitment of nurses/midwives for FGDs
Nurses/midwives were purposively recruited based on their roles and responsibilities in maternity care. Variations in terms of cadre, years of service and gender were considered in order to have a broader perspective of the content for couple antenatal education. Seven nurses/midwives were selected; two each from Mpemba and SL health centres and; the remaining three were from QECH as Mpemba and SL health centres could not provide the recommended number for a FGD. Three were registered nurses/midwives while four were nurse/midwife technicians and their experience ranged from four to 18 years working in maternity units including antenatal clinics. All were females except one. Nurses/midwives were included as we felt that through their training and experience they would be in a position to propose education content for couples. Although the group was heterogeneous in the sense that it had nurses/midwives from a central hospital whose exposure might be different from nurses/midwives from health centres thereby affecting discussions, the advantage was that data were enriched as it came from different locations from people with different experiences.
Selection and recruitment of key informants
Key informants (KIs) were selected based on their roles and responsibilities in various institutions, which were both health related and non-health related, to gather their perspectives in relation to messages couples should receive when they come for antenatal education sessions. The researchers recruited the key informants. In total, there were 13 KIs and the composition was as follows: one obstetrician, one senior nurse/midwife educator (Head of Department for Maternal and Child health), one senior practicing nurse/midwife (matron), one policy maker (maternal and child health coordinator). Those from non-health institutions were two religious leaders, one Christian and one Muslim, one leader representing the small-scale business community, two group village heads (female and male) and four employers (two from private owned institutions, one from a statutory cooperation and one from a public institution).
The KIs with health backgrounds were chosen as they would provide relevant content for couple antenatal education. Similarly, we assumed that non-health KIs, due to their diverse backgrounds, roles and responsibilities, that their input would not be biased in relation to content for couple antenatal education. For instance, we chose two religious leaders, one Muslim and the other Christian. They were chosen because they represent the two major religious groups in Malawi and would provide a religious point of view. Additionally, the village heads (one male and one female) would suggest topics for couple antenatal education which are likely to be culturally acceptable.. The eligibility criteria for KIs required them to be 18 years or above and currently serving in that particular position for not less than two years. All key informant interviews were conducted at the work place during the convenient time and date agreed upon. The interviews were conducted in a private place.
Selection and recruitment of couples and men who had previously escorted their spouses for antenatal care for IDIs
Participants for IDIs were men who had escorted their wives to the antenatal clinic previously, three years prior the study and couples who were expectant or had a child under five years during the period of the study and were from the catchment areas of Mpemba or South Lunzu. Men who had escorted their wives were chosen in order to ascertain their opinions as they were already motivated to be involved in antenatal care and would therefore be in a better position to share relevant antenatal education information couples may need. Couples were chosen because this study sought information for couples, therefore their input as couples could be of great value. Both the men and couples were aged 18 years and above. With the assistance of health workers, the participants were recruited through the health facilities and most of them were recruited from the catchment areas, as with the male discussants. All the participants who were approached for the IDIs consented to the interviews. Interviews were conducted at agreed times and venues. The participants varied in terms of age and number of children. In total there were 10 IDIs for couples and 10 men. Observing variations during recruitment allowed us to identify differences and similarities among the groups of participants in relation to the subject matter, which in this study was the content for couple antenatal education.
Data collection for FGDs
Two research assistants (RAs), a female registered nurse/midwife and a male social scientist, were recruited. The RAs were recruited as they have knowledge on maternal health, male involvement and the research process. In addition, the presence of a male research assistant might have helped male participants to openly share during FGDs for men. The research assistants received training to familiarize themselves with the study, its background, aim and their roles and responsibilities in this process of data collection.
The participants were given information regarding the purpose of the study and why they were chosen including the time each interview would take which was 60 to 90 min. To facilitate their understanding about the study, the participants were given an information sheet and were asked to read it carefully. For those who could not read, the information sheet was read to them. They were informed that they could ask questions for clarification and that their participation was voluntary.
Additionally, it was communicated to the participants that the discussions will be digitally recorded for accuracy and completeness of data, which all participants accepted. All participants signed an informed consent form demonstrating their acceptance to participate in the study.
All focus group discussions except the male FGDs were moderated by one of the researchers (MCC) while the research assistants filed verbal and nonverbal behaviors of participants during discussions and recorded the conversations. The male FGDs were moderated by the male research assistant and MCC was the note taker. The participants in the FGDs were given numbers for identification during the discussions. All FGDs, except with nurses/midwives, were conducted in the local language, Chichewa. The FGD for nurses/midwives was conducted in English, the language of instruction in schools and work places. A pretested unstructured discussion guide was used and had one broad question as follows: What type of information should be discussed with expectant couples during antenatal education sessions? Probes were used during the discussions to obtain detailed and clear information about antenatal education information needs for the couples. For example, participants were asked to suggest the content and depth of discussion concerning pregnancy, childbirth and the postpartum period (see Additional file 2: Focus group discussion guide). The unstructured discussion guide generated rich information as prior information on the subject matter did not influence the interview, rather the participants’ narrations structured the interview [29, 36]. After each interview, a summary of important points regarding antenatal information needs was provided to the participant as means for verifying what had been discussed during the discussions . Additionally, one of the researchers (MCC) and the moderator met to review and plan for the next focus group discussion after each FGD.
Data collection for IDIs and KIIs
One of the researchers (MCC) conducted all the in depth and key informant interviews using the pretested unstructured interview guides which had one broad question and probes just like the discussion guide (see Additional file 3: In depth interview guide and Additional file 4: Key informant interview guide). Each interview took 40 to 90 min and none of the participants refused to have their voices recorded. All IDIs and some KIIs were conducted in Chichewa. After each interview, a summary of important points regarding antenatal information needs was provided to each participant as means for verifying what had been discussed during the interviews .
Trustworthiness of the data
To ensure trustworthiness of the data, we considered credibility, dependability, conformability and transferability. Piloting of the research instruments and inclusion of direct quotes in the results section enhanced credibility and dependability. We achieved conformability by triangulating data as information collected from multiple sources and methods help in confirming emerging issues [29, 36]. The study considered parity of the participants, which enhanced transferability of the study as views came from individuals with a variety of experiences. In addition, the participants were drawn from semi-urban settings, which would mean we captured views from urban and rural settings.
The recorded information, field notes and transcripts were kept in a lockable cupboard accessible by the researchers only. The computer with the data had a password known by one of the researchers (MCC). Soon after data collection, the researchers transcribed the recorded data verbatim and the Chichewa transcripts were then translated into English. Recorded data were transcribed verbatim before cleaning and anonymising to remove any participant identifying details.
Data analysis was done simultaneously with data collection to allow the researchers refine the subsequent interviews. The data were analyzed using thematic content analysis frame work. Braun & Clarke [37:79] describe thematic analysis as “a method for identifying, analyzing and reporting patterns (themes) within data not necessarily on dependent on quantifiable measures but rather on whether it captures something important in relation to the research question ”. Conversely, content analysis, apart from identifying, analyzing and reporting themes, patterns of words used and their frequency are also regarded important as frequent occurrence may indicate significance particularly in areas where little is known about a particular phenomenon [29, 35]. In this study, thematic content analysis was significant as it involved exploring views from different stakeholders on the content for couple antenatal education, an area with little information. Deductive and inductive approaches were employed to code the data. The former focused on literature and study objectives while the latter was data driven. The data analysis was guided by six stages according to Braun & Clarke which are familiarizing with the data, generating initial codes, searching for themes, refining themes, naming themes and producing the report .
Familiarization with the data
All the scripts for the focus group discussions, in-depth interviews and key informant interviews were read once by MCC against the recorded information to get the sense of the data. Initial ideas related to the objective of the study were written down. Thereafter, identification numbers for each script were written on small pieces of paper and casted in a box. An independent person picked four papers from the box as the researcher wanted four transcripts for further familiarization with the data. The four chosen transcripts were as follows: couple in depth interview no 1 (CID 01), older male focus group discussion from South Lunzu, key informant from PVHES (Key informant No 11) and a younger female focus group discussion from Mpemba.
Generating initial codes
We coded a transcript (couple IDI CID 01) by reading through line by line. All items relating to the same topic were coded to similar nodes. Coding was done inductively and deductively.
Searching for themes
Codes that were identified in stage two above were reviewed by an independent researcher who also coded a clean copy of the script which MCC coded. Agreement was done on the codes to be used for all the transcripts with the independent researcher, co-authors: ASM and EMC. The codes identified were used to code the data and emerging codes were included as well in the process. The coded data was categorized and common themes were identified.
Reviewing themes & defining and naming themes
We reread the data to identify a coherent pattern and to see if the data fitted into each theme identified. The first author verified the themes by checking on them against the audio taped data. Similarities and differences were noted across the data set at this stage.
Producing the report
The report outlined the overarching theme, subthemes and categories.
We obtained ethics approval from the College of Medicine Research and Ethics Committee (COMREC) Certificate No P.11/151821. Permission was obtained from individual heads of institution where participants were drawn as follows: Queen Elizabeth Central Hospital (QECH) Blantyre District Health Office, Blantyre District Commissioner, Kamuzu College of Nursing, Blantyre Campus, Private Vehicle Hire Engineering Services (PVHES) and Rangers Security Company. Additionally the researchers ensured the participants were aware they could withdraw without reprimands at any time. Confidentiality and anonymity of the participants were observed by conducting interviews in a private room and using codes for identification of the participants. Participants in focus group discussions were told that the researchers were not in control of the information that may be disclosed outside of the discussion by participants within the group itself. The participants were also informed that the information collected might be published while maintaining confidentiality as agreed.