681 - Effect of training on rates of identification of suspected maternal and neonatal infections: A secondary analysis of the Azithromycin-Prevention in Labor Use Study (A-PLUS) pilot phase

Publication Number: 3658.681

Amy Mackay, University of Alabama School of Medicine, Birmingham, AL, United States; Norma Pugh, RTI International, RTP, NC, United States; Musaku Mwenechanya, University Teaching Hospital, LUSAKA, Lusaka, Zambia; Elwyn Chomba, University Teaching Hospital, Lusaka, Lusaka, Zambia; Vivek V. Shukla, University of Alabama at Birmingham, Birmingham, AL, United States; Denise C. Babineau, RTI International, Durham, NC, United States; Jennifer Hemingway-Foday, RTI International, Durham, NC, United States; Elizabeth McClure, RTI International, CHAPEL HILL, NC, United States; Janet Moore, RTI International, Research Triangle Park, NC, United States; Waldemar Carlo, University of Alabama, Birmingham, AL, United States

Background: Maternal and neonatal infections are a major cause of mortality worldwide, disproportionately affecting those in low-resource settings. Prevention, early diagnosis, and effective management of infections are key to reducing morbidity and mortality.

Objective: To determine if additional training in recognizing maternal and neonatal infection increased identification of infection

Design/Methods: We divided the multi-site A-PLUS pilot phase data into four periods, based on site-specific enrollment proportions. Maternal and neonatal characteristics and infections were compared across periods and by infection status using Pearson's chi-squared, Kruskal-Wallis, or Type III F-tests. We estimated unadjusted and adjusted infection rates and risks for each outcome using generalized linear mixed models with binomial or negative binomial distributions. We adjusted maternal rates for maternal education, primiparous status and prolonged labor and neonatal rates for Apgar score < 7 at 5 minutes.

Results: Baseline characteristics across periods differed by maternal education, primiparous, prolonged labor (≥ 18 hours), and Apgar score < 7 at 5 minutes (Table 1). Adjusted number of maternal infections per 100 moms within 6 weeks after delivery did not differ across periods: 4.4 (95% CI 2.2-8.8), 4.6 (95% CI 2.2-9.0), 3.8 (95% CI 1.8-7.6) during periods 2-4 with period 1 as reference 4.7 (95% 2.3-9.3), (overall adjusted p-value = 0.49). Adjusted number of neonatal infections per 100 neonates did not differ during periods 2-4: 6.6 (95% CI 3.3-12.8), 7.0 (95% CI 3.6-13.5), and 5.9 (95% CI 2.9-11.4), overall adjusted p-value = 0.64, with period 1 as reference 6.5 (95% CI 3.3-12.5). Baseline differences by infection status included region of residence, labor induction, prolonged rupture of membranes, birth weight, and Apgar score < 7 at 5 minutes (Table 2). Mothers with infections before or at the time of an unplanned medical visit had an adjusted relative risk of 1.52 (95% CI 1.39-1.64, p-value < 0.0001) of unplanned medical visits within 6 weeks after delivery compared to mothers without any reported infection. We observed similar findings in neonates: 1.46 (95% CI 1.34-1.58, p-value < 0.0001) for unplanned medical visits within 6 weeks after birth compared to those without any reported infection.

Conclusion(s): Identification of maternal and neonatal infections were stable across periods after initial training. Infection rates aligned with published data. Infections were associated with increased unplanned medical visits within 6 weeks after delivery or birth, suggesting high-risk participants were identified.

Table 1: Maternal and neonatal characteristics by period
*C = chi-square p-value, K = Kruskal-Wallis p-value, F = Type 3 Fixed Effect p-value

Figure 1A: Rate of any reported maternal infection during pregnancy by site and period, Figure 1B: Rate of any reported maternal infection within 6 weeks of delivery by site and period, Figure 1C: Rate of any reported neonatal infection within 6 weeks after birth by site and period


Table 2: Baseline characteristics among those with any reported infection and those with no reported infections
*C = chi-square p-value, W = Wilcoxon p-value, F = Type 3 Fixed Effect p-value