These will deal with the research design, study population, the methods of data collection, processing, analysis and the limitations to the study.
An explorative qualitative design using in depth interviews was employed to explore the perceptions and experiences of male companionship during delivery [5] The study was carried out at a missionary private not for profit hospital providing quality care (St Francis Hospital Nsambya).
The study population comprised of all men of eighteen years and above who accompanied their spouses during labour and were willing to participate in the study. It is a missionary hospital located in the periurban area.
Its catchment population is approximately 250,000 with 7000 deliveries per year. Antenatal and maternity services at public health units in Uganda are provided free of charge. The Hospital, being a nongovernmental not- for- profit institution under Kampala Archdiocese, charges a fair user fee for these services.
Interviews were done for men who had accompanied their spouses during delivery until data saturation was reached. This was a point where no new information was gotten. Sixteen men were interviewed.
Purposive sampling was used to select participants who have characteristics with the potential of providing rich, relevant, and diverse data. The participants were approached in person, and none declined to participate.
Written consent was obtained from each of the men whose partner had normal vaginal delivery or caesarean section with a live birth in the hospital.
In depth interviews (IDIs) were the main data collection methods used in the study. Two (2) trained data collectors were used in the study. A pretested semi structured interview guide was used. I the principal researcher with the research assistant carried out the interviews. Interviews were conducted in English or Luganda, a local dialect in the periurban setting, depending on the language the respondent was comfortable with. Follow-up questions were asked and the clarification of points which arose was also sought.
One of the research assistants moderated and guided the interview while the other took notes and did the tape recording. The interviews were tape-recorded with the permission of the participants; in addition, the interviewer took notes. Codes were written on the interview guides and each recording was started by first mentioning the code on the interview guide to ensure data collected could be analyzed as belonging to that interviewee.
Notes were taken immediately after each interview. These notes covered the initial interviewee reactions, including the first analytical reflections from the interview content, and any useful observations that could not be captured by digital recording. Notes were taken on the demeanor of the respondent, his body language and mood, and any informal conversation that took place before or after the interview. Each interview ranged from 30 to 45 min.
The guide used covered the main areas on male companionship during labour. Questions were raised and participants were encouraged to express their opinion openly. Probes were made for clarification. Point of saturation was deemed to have been reached if no new information was obtained.
Validity in qualitative research is the extent to which observations reflect phenomenal of interest [6].
The principal researcher is a male doctor who conducted the interviews together with trained research assistance. The principal researcher was working at the Hospital while the research assistant in school of public health at a University. Data collection tools were pre tested and the semi structured interview guide was piloted to test the understanding of the questions.
Hence to improve credibility in data collection, clarification was sought during and after each interview. Triangulation was done to improve validity by interviewing women whose male partners participated in the study to collaborate their information.
Reliability in qualitative inquiry refers to the consistency of the research findings, through the interview process, the data transcription and analysis [6]. A semi structured interview guide was used to make sure that all participants answered the same questions. This improved consistency during interviews and build up of codes and themes. The data was transcribed and rechecked collaborating with the notes taken during interviews for gestures and motions, improving consistency.
Repeated listening to tapes of interviews was done however in absence of participants during analysis. Repeat interviews were done for 2 randomly selected participants to correlate with the initial interviews.
The taped interviews were transcribed verbatim after repeatedly listening to the recordings and the resulting texts analyzed by using Content Analysis technique. Broad themes were extracted from the transcripts and then progress to identifying coded themes. In establishing themes, consideration was given to statements of meaning that were present in most of the relevant data. In an attempt to ensure the credibility of the findings, an independent experienced coder was used to verify or corroborate the themes extracted from the data. The data was analyzed simultaneously with data collection. This allowed the researchers to progressively focus the interviews and observations, and to decide how to test the emerging conclusions. The transcripts were entered into ATLAS.ti for analysis. I developed a codebook based on the major themes of the study. The major themes were transformed in tree nodes and free nodes. Based on the codebook, I developed verified independently coded texts from the transcriptions.
The emergent themes and sub-themes are discussed below, supported and illuminated by respondents’ quotes. After exhaustive analysis of the interviews, it was possible to identify the following themes: feeling about attending child birth, responsibilities during child birth, positive experiences and negative experiences about child birth.
This research was approved by the Hospital’s Institutional Review Board (IRB) - IRC/PRJ/010/14/05. The data collection was carried out between October 2014 and January 2015, using semi-structured interviews, after being approved by the IRB of the fore mentioned institution and after the participants signed a Consent Form. To ensure confidentiality of participants who gave written consent, codes were used on the form instead of their names. In addition, only codes were written on interview transcripts.
The results cannot be generalized to all populations in and outside Uganda. We do not claim that these findings apply to all men who accompanied their partners during pregnancy and child birth.
The men who took part were married. Further exploration with men of different ethnicities, relationship status, and ages would be beneficial to expand on and compare with the findings in our study. The pre conception of the principal investigator from biomedical background training may affect the interpretation of information obtained from the interview.