Pregnant and Postpartum Women’s Experiences and Perspectives on the Acceptability and Feasibility of Copackaged Medicine for Antenatal Care and PMTCT in Lesotho

Objective. To improve PMTCT and antenatal care-related service delivery, a pack with centrally prepackaged medicine was rolled out to all pregnant women in Lesotho in 2011. This study assessed acceptability and feasibility of this copackaging mechanism for drug delivery among pregnant and postpartum women. Methods. Acceptability and feasibility were assessed in a mixed method, cross-sectional study through structured interviews (SI) and semistructured interviews (SSI) conducted in 2012 and 2013. Results. 290 HIV-negative women and 437 HIV-positive women (n = 727) participated. Nearly all SI participants found prepackaged medicines acceptable, though modifications such as size reduction of the pack were suggested. Positive experiences included that the pack helped women take pills as instructed and contents promoted healthy pregnancies. Negative experiences included inadvertent pregnancy disclosure and discomfort carrying the pack in communities. Implementation was also feasible; 85.2% of SI participants reported adequate counseling time, though 37.8% felt pack use caused clinic delays. SSI participants reported improvement in service quality following pack introduction, due to more comprehensive counseling. Conclusions. A prepackaged drug delivery mechanism for ANC/PMTCT medicines was acceptable and feasible. Findings support continued use of this approach in Lesotho with improved design modifications to reflect the current PMTCT program of lifelong treatment for all HIV-positive pregnant women. This was a mixed method, cross-sectional study. Structured exit interviews (SI) and in-depth semistructured interviews (SSI) with pregnant and early postpartum women were conducted from December 2012 to May 2013. The study was conducted in six purposively selected districts out of the ten districts in Lesotho. Three districts were those selected by the MOH to first implement the mother/baby copackage. Three comparable districts with later initiation were selected to match the three geographical settings of Lesotho (Lowlands, Foothills, and Highlands) represented in the initial implementing districts. Within each district, health facilities (HF) were randomly selected as study sites using the probability proportional-to-size (PPS) method to ensure all HF had the same probability of being selected. SI were conducted in all 31 study sites while SSI were conducted in two study sites per district (highest volume hospital and health center) for a total of 12 sites. HIV-positive and HIV-negative women who were attending a subsequent ANC visit and 6-week and 14-week postnatal care (PNC) visits at study sites were eligible for participation in interviews. Health care workers (HCW) introduced the study and assisted in linking potential participants with study staff. Study staff attempted to conduct SSI with HIV-positive women who did not return to ANC or early PNC to capture their experiences with the copackage and any influences it may have had on their health-seeking behavior. We used the routine PMTCT program follow-up system to identify and trace women who had missed visits. When these women were found, the community or health worker introduced them to the study and invited them to speak with study staff at a convenient time and location. A sample of 196 women in each of the three groups (HIV-negative, HIV-positive on prophylaxis, and HIV-positive on ART) was targeted. We estimated 85% of women would find the intervention both acceptable and feasible, based on the MPP evaluation [15]. Assuming an expected proportion of 0.85, the large sample normal approximation was used to calculate a two-sided 95% confidence interval around the observed proportion with a margin of error of 0.05 to determine a needed sample size for SI of 196. The target sample per HF was determined using PPS based on routinely collected antenatal attendance program data. Women were selected through a random process by which they were consecutively referred and screened on days when study staff were present, until either the sample size at each HF was reached or the data collection period ended. HF targets were further divided into six subgroups of roughly equal numbers based on all possible combinations of visit type (ANC, PNC) and type of copackage received (Pack 1, Pack 2, and Pack 3). Once a subgroup target was reached, women in that particular group were no longer eligible. Trained maternal and child health (MCH) study nurses explained the study, obtained and documented verbal informed consent, and conducted SI using data collection instruments specifically designed according to the type of visit and type of copackage received. All interviews were conducted in Sesotho. SI involved closed-ended questions, including a series of seven copackage acceptability (e.g., size, convenience, and design) and eight copackage feasibility (e.g., counseling, understanding, and clinic flow) statements. Women were asked to indicate whether they agreed or disagreed or had no opinion for each statement. They were also asked to describe what they liked/disliked about the copackage and their positive/negative experiences related to its use. Interviewers selected precoded responses that best fit the women’s answers and documented other responses that could not be classified. SSI guides covered similar topics, but their open-ended nature allowed interviewers to probe more deeply. These interviews were intended to capture richer and more complex information than the SI. As such, a smaller approximate sample size range was estimated in order to reach saturation of theme: 9–15 HIV-positive women (with approximately equal targets for women on prophylaxis and ART) and 4–6 HIV-negative women per HF for a total of 108–180 and 48–72, respectively, with subgroup targets by visit type (ANC, 6-week PNC, 14-week PNC). SSI study staff were different from those conducting SI and were trained in qualitative research methods. Consecutive recruitment of women on days when study staff were present was similar to the SI process. If SI and SSI recruitment were taking place at the same HF on the same day, women were first approached to participate in the SSI; if they declined, they were asked whether they were willing to participate in the shorter SI. SSI were audio recorded and were simultaneously transcribed and translated into English. Transcripts were reviewed by the study coordinators and/or investigators once translated. Data from close-ended questions in SI and SSI were entered into a Microsoft Access database (2007) using a double data entry and verification system. Descriptive statistics were calculated and reported for key characteristics of the study participants and agreement with acceptability and feasibility statements and reported experiences. Means and standard deviations were reported for all continuous variables (Table 1) and frequencies and percentages were reported for all categorical variables (Tables ​(Tables11 and ​and22 and Figures ​Figures22 and ​and3).3). Differences in (1) the level of agreement with acceptability and feasibility statements and (2) the experience of either positive or negative consequences as a result of pack receipt among HIV-negative women, HIV-positive women on prophylaxis, and HIV-positive women on ART were examined with generalized estimating equations (GEE) using the binomial distribution with the logit link for dichotomous outcomes or the multinomial distribution with the cumulative logit link for ordinal outcomes. Select significant findings are highlighted in the text. Compound symmetry and independent working correlation structures were considered to account for the clustering of women in multiple facilities. Score tests were used to test the proportional odds assumption, and the Tukey-Kramer method for P value adjustment was used to account for multiple comparisons. All statistical tests were two-sided and the level of statistical significance was set at 0.05. All data were analyzed in Washington, DC, USA, using SAS/STAT software version 9.3 (Cary, North Carolina). Percentage of structured interview participants that agree with the copackage acceptability statements by HIV status group. (Possible responses to statements included agree, disagree, and no opinion. “No opinion” responses ranged from 0.4 to 2.0% of the three groups combined for each statement, except the instruction sheet statement, in which 13.7%, 24.8%, and 22.0% of HIV-negative, HIV-positive on prophylaxis, and HIV-positive on ART women responded with “no opinion,” because they did not have or did not use the sheet.) Percentage of structured interview participants that agree with the copackage feasibility statements by HIV status group. (Possible responses to statements included agree, disagree, and no opinion. For most statements, “no opinion” responses ranged from 0.6 to 7.0% of the three groups combined for each statement. The exceptions were “all questions about the pack were answered” in which 12.8%, 10.5%, and 8.5% of HIV-negative, HIV-positive on prophylaxis, and HIV-positive on ART women, respectively, responded with “no opinion,” at first visit and 16.1%, 13.2%, and 12.7%, respectively, responded with “no opinion,” at this visit (if interviewed in ANC) or subsequent visits (if interviewed in PNC) because they did not have questions.) Characteristics of study participants. ∗Excluded from the percentage calculation. Women’s positive and negative experiences with the pack reported in structured interviews. SSI transcripts were imported into MAXqda (V10). A codebook was created based on the research objectives and variables of interest. Data were coded by a team in the US and Lesotho and reviewed for consistency by one investigator. Textual data were carefully reviewed to identify recurrent patterns and themes and draw conclusions connected to the study questions. Data were summarized through descriptive, text-based summaries and data display matrices. Both deductive codes based on research questions and inductive codes derived from the data were utilized. This study was approved by the George Washington University Institutional Review Board (IRB), the Baylor College of Medicine Lesotho IRB, and the Lesotho MOH ethical review committee. All participants underwent a verbal informed consent process in Sesotho using an IRB-approved verbal consent text with documentation of consent by study staff.