Background: Stillbirth and neonatal mortality are very high in many low-income countries, including Malawi. Use of family planning to encourage birth spacing may optimize outcomes for subsequent pregnancies. However, reproductive desires and influences on birth spacing preferences of women who have experienced a stillbirth or neonatal death in low-resource settings are not well understood. Methods: We conducted a qualitative study using 20 in-depth interviews and four focus group discussions with women who had experienced a stillborn baby or early neonatal death to explore attitudes surrounding birth spacing and potential interventions to promote family planning in this population. Qualitative data were analyzed for recurrent patterns and themes and central ideas were extracted to identify their core meanings. Results: Forty-six women participated in the study. After experiencing a stillbirth or neonatal death, most women wanted to wait to become pregnant again but women with living children wished to wait for longer periods of time than those with no living children. Most women preferred birth spacing interventions led by clinical providers and inclusion of their spouses. Conclusions: Many influences on family size and birth spacing were noted in this population, with the most significant influencing factor being the spouse. Interventions to promote birth spacing and improve maternal and neonatal health in this population need to involve male partners and knowledgeable health care providers to be effective.
This was a qualitative study using in-depth, semi-structured interviews and focus group discussions. Approval was obtained from the National Health Sciences Research Committee of Malawi (Protocol #1354) and the University of North Carolina School of Medicine Institutional Review Board (#14–2677). Women gave written informed consent at the time of enrollment in the language of their choice (Chichewa or English). The study population was recruited from Bwaila Hospital, a district government hospital in Lilongwe, the capital city of Malawi. Bwaila has approximately 15,000 deliveries annually, of which 2900 are preterm. Between 80 and 110 cases of birth asphyxia (a portion of which results in stillbirth or neonatal death) occur each month on the labor ward. Inclusion criteria for the study were: 1) current admission to the postpartum ward at Bwaila Hospital; 2) delivery of a stillborn over 28 weeks gestation or with a birthweight ≥1000 g, or delivery of a liveborn infant weighing ≥1000 g with a neonatal demise in the first 7 days of life; 3) ability to speak Chichewa (the local language) or English fluently; and 4) age 18–45 years old. We recruited 60 women with and without living children from prior pregnancies at a 1:1 ratio from the postnatal wards. A demographic form was completed for women who consented to be part of the study. This form collected information about age, number of other living children, HIV status, marital status, completed education, and occupation. Data on access to a working phone and roof type were collected to assess socioeconomic status. HIV testing is performed on all Malawian women during antenatal care unless they opt out. HIV status was determined by verifying the participant’s health passport (a government-issued personal medical record booklet kept by the patient) with the participant’s permission at time of enrollment. Enrolled women were then contacted and traced 4–8 weeks later to either participate in an in-depth interview or a focus group discussion. All in-depth interviews and focus group discussions were conducted by the same experienced bilingual researcher (M.T.). Twenty in-depth interviews took place in participants’ homes or another private setting and four focus group discussions (of 6–8 participants each) took place in a private conference room on the campus of Kamuzu Central Hospital in Lilongwe, Malawi. Interviews and focus group discussions were audiotaped, transcribed and translated into English. All transcriptions and translations were completed by the same researcher (M.T.). Accuracy of the translations was verified by two other bilingual members of the research team (A.B. and G.H.). Our analysis approach was to use content analysis to compare the birth spacing intentions of women who did and did not have living children. The interview and focus group discussion guide (Additional files 1 and 2) focused on several domains, two of which are relevant to this analysis: 1) birth spacing plans and influences, and 2) acceptable educational interventions to promote birth spacing and family planning among women with poor obstetric outcome. Focus group and in-depth interview guides were used to ensure that all critical topics were discussed, but the interviewer was given license to cover topics in a manner that facilitated flow and rapport. A specific aim of the focus group discussions was to facilitate brainstorming about potential birth spacing interventions, whereas the in-depth interviews focused more on individual and social influences on birth spacing that may be too personal to share in a group setting. For each domain, results were analyzed separately for women with and without living children prior to the stillbirth or neonatal death to examine the role this plays on reproductive desires. Previous qualitative exploration in this field has demonstrated that the minimum threshold for data saturation can be reached within 20 in-depth interviews and four focus group discussions [14–16]. Transcripts of completed interviews were independently analyzed by two of the investigators (A.B. and D.K.). A code dictionary was developed in an iterative process based on identified domains, and this dictionary was assigned to sections of the text using qualitative software NVivo® 10. Recurrent themes and sub-themes were identified based on these initial codes, and any discrepancies were resolved through discussion. Matrices and tables were used to organize the data and display these to facilitate analysis that integrated both in-depth interviews and focus group discussions based on the conceptual domains determined a priori.