Menu
Menu
Menu
Menu
Background: Advance provision of misoprostol to women during antenatal care aims to achieve broader access to uterotonics for the prevention of postpartum hemorrhage. Studies of this community-based approach usually involve antenatal education as well as timely postpartum follow-up visits to confirm maternal and neonatal outcomes. The MamaMiso study in Mbale, Uganda sought to assess the feasibility of conducting follow-up visits in the postpartum period following advance provision of misoprostol for postpartum hemorrhage prevention. MamaMiso recruited women during antenatal care visits. Participants were asked to contact the research team within 48 h of giving birth so that postpartum follow-up visits could be carried out at their homes. Women’s baseline and delivery characteristics were collected and analyzed with respect to follow-up time (‘on time’ ≤ 7 days, ‘late’ > 7 days, and ‘lost to follow up’). Every woman who was followed up late due to a failure to report the delivery was asked for the underlying reasons for the delay. When attempts at following up participants were unsuccessful, a file note was generated explaining the details of the failure. We abstracted data and identified themes from these notes. Results: Of 748 recruited women, 700 (94%) were successfully followed up during the study period, 465 (62%) within the first week postpartum. The median time to follow up was 4 days and was similar for women who delivered at home or in facilities and for women who had attended or unattended births. Women recruited at the urban hospital site (as opposed to rural health clinics) were more likely to be lost to follow up or followed up late. Of the women followed up late, 202 provided a reason. File notes explaining failed attempts at follow up were generated for 164 participants. Several themes emerged from qualitative analysis of these notes including phone difficulties, inaccurate baseline information, misperceptions, postpartum travel, and the condition of the mother and neonate. Conclusions: Keeping women connected to the health system in the postpartum period is feasible, though reaching them within the first week of their delivery is challenging. Understanding characteristics of women who are harder to reach can help tailor follow-up efforts and elucidate possible biases in postpartum study data.
From May to October 2012, a study of self-administered misoprostol for PPH prophylaxis (MamaMiso) was implemented in Eastern Uganda enrolling 748 women. The main clinical findings are reported elsewhere [13]. Trained clinical staff recruited pregnant women into MamaMiso during antenatal care visits at four participating health facilities (one regional hospital and three health centers). The regional hospital was a tertiary facility located in the center of town, serving a population of about 4–5 million people from 14 districts and staffed by OBGYN specialists. The health centers were rural facilities staffed primarily by clinical officers and midwives. Women were only recruited to the study if they lived within one of the pre-specified 200 villages near to the recruiting centers. During antenatal care visits, women with estimated gestational ages 34 weeks or higher who did not plan to travel after their delivery were informed about the study aims and procedures (details reported elsewhere). Women who consented to be involved in the study were asked for contact information including up to two telephone numbers and detailed directions to their place of residence (including a sketched map when possible). Baseline demographic data and reproductive histories were collected as well as an assessment of hemoglobin (Hb) levels. Women were provided with a neck purse containing the study medication (either misoprostol or placebo), instructions for its proper use, and a phone number to call to report their delivery. Women were asked to report their deliveries immediately either directly to the study team or to their village health teams (VHT) who would subsequently notify the study team. VHTs are composed of community volunteers who provide health information and linkages to health services. Women were informed that a postnatal follow-up visit would be conducted at home 3-5 days after delivery, a period selected both to capture the nadir of the hemoglobin reading and to minimize poor recall of labor and delivery. In the three health centers, a VHT Coordinator was present at enrollment and linked the participants to VHTs who assisted with delivery notification. To incentivize timely notification of deliveries, VHTs were informed that they would receive 10,000 Ugandan Shillings (~ 4 USD) for the timely report of a delivery. The incentive was split between the women and VHT if both were involved in the delivery notification and a smaller incentive was provided for deliveries reported after 5 days. Efforts were made to follow up women, even beyond the target time period of 3–5 days postpartum in order to confirm the well-being of the mother and newborn. Upon successful follow up, women were provided a small gift of soap. While the study aimed for follow up between 3 and 5 days, for the purposes of this data analysis a cut-off time of 7 days postpartum was used to make it comparable to other studies and research protocols (UNICEF and WHO recommend neonatal home based care within the first week of life [15]). Women were considered “on-time” if follow up was achieved within 1 week of the delivery date, “late” if they were successfully followed up 8 days or more after the delivery, and “lost” if the study team was never able to conduct a full follow-up interview. Some women who were “lost” were contacted by phone, and limited information was gathered on them. Patient characteristics and clinical outcomes were then analyzed by these ‘time to follow up’ categories. Nearly all delivery notifications were received by the Research Fellow, mostly by phone. The Research Fellow then coordinated follow up by assigning a study staff member (midwife or nurse) to locate the woman, typically by using a motorcycle taxi to drive to the participant’s home. Follow up consisted of an in-person interview using a standardized case report form and the assessment of hemoglobin levels, using a Hemocue® handheld device (Hemocue, Angelholm, Sweden). Information regarding the details of the delivery, medications used, and health status of the mother and baby were collected. Unused medicines or empty foil packs were also retrieved during the visit. When follow-up difficulties arose early on, a system of keeping extensive file notes was put into place. Each woman who did not report her delivery on time was asked why, and when attempts by staff to follow up participants were unsuccessful, the details were noted. This analysis seeks to understand the challenges of community-based follow up in the context of a placebo-controlled randomized trial of antenatal distribution of misoprostol to women for self-administration to prevent postpartum hemorrhage. Bivariate analyses of background and follow-up characteristics were conducted by follow-up time and multivariate logistic regression was used to predict follow-up category. Data were abstracted from the file notes and reviewed for initial and emergent themes using a general inductive approach. Themes were identified through an iterative process, and data were independently coded by two researchers (LF and MB).
N/A
