The Pregnancy Risk Assessment Monitoring System–Zika Postpartum Emergency Response Study (PRAMS-ZPER) was conducted in Puerto Rico by PRDH and CDC to assess women’s use of measures to prevent ZIKV infection during pregnancy (11). PRAMS-ZPER, a hospital-based survey of women with a recent live birth, was implemented islandwide in 2 phases: August 28–December 3, 2016, and November 1–December 19, 2017. Hospitals with ≥100 births in the year before the study period were eligible to participate. In 2016, all 36 eligible hospitals participated; in 2017, a total of 30 of 31 eligible hospitals with operating maternity wards during the study period participated.

Delivery dates were randomly sampled for each hospital, and delivery logs were used to identify women who delivered on the sampled dates. Eligibility criteria are described elsewhere (12,13). In 2016, women were approached 24–36 hours after their infants were delivered (12). In 2017, because the environment after Hurricane Maria resulted in early hospital discharges, women were approached within 24 hours postdelivery unless hospital staff recommended otherwise (e.g., mothers who had not yet recuperated). Response rates were 80.6% in 2016 and 94.4% in 2017.

In previous research, we reported 2016 PRAMS-ZPER data on maternal use of ZIKV prevention measures (12) and on the association between provider counseling on condom use to prevent ZIKV infection and self-reported use of condoms with sex partners during pregnancy (13). In this analysis, we report on the association between counseling from prenatal care providers and women wearing protective clothing and using insect repellent during pregnancy to prevent ZIKV infection using data from 2016 and 2017. We assessed the prevalence of wearing protective clothing and using repellent overall and by select maternal characteristics (e.g., education level, marital status; Table 1). We constructed separate multivariable, survey-weighted, logistic regression models for each study year to examine maternal characteristics associated with receiving provider counseling on wearing protective clothing and using repellent to prevent mosquito bites. We constructed separate multivariable models to examine associations between factors identified a priori, including receiving provider counseling, with wearing protective clothing daily and using repellent frequently (defined as use every day in 2016 and use always in 2017) when outside during pregnancy. Each model was further adjusted for maternal characteristics, health district region, and delivery month.

Among 3,806 combined respondents in 2016–2017, nearly all, 99.4%, received prenatal care. Among those who received prenatal care, 87.8% participated in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) during pregnancy, 79.0% were 20–34 years of age, 69.5% had >high school education, and 68.6% were unmarried.

Most women reported receiving provider counseling during pregnancy to wear protective clothing in 2016 (87.1%) and 2017 (79.8%) (Table 1); however, few women reported wearing protective clothing daily during either year (11.3% in 2016 and 7.9% in 2017) (Table 2). In 2016, the prevalence of wearing protective clothing was lower among women ≤19 years (7.8%; adjusted prevalence ratio [aPR] 0.46, 95% CI 0.27–0.77) and 20–34 years (11.1%; aPR 0.65, 95% CI 0.46–0.94) compared with those ≥35 years (17.0%) of age and higher among women with less than a high school diploma (20.2%; aPR 2.03, 95% CI 1.38–2.99) and a high school diploma (13.2%; aPR 1.33, 95% CI 1.01–1.73) compared with those who had more than a high school education (9.9%). In 2017, the prevalence of wearing protective clothing was lower among women 20–34 years compared with ≥35 years of age (7.1% vs 13.5%; aPR 0.52, 95% CI 0.36–0.76) and higher among women with a high school diploma compared with those with more than a high school education (10.9% vs 7.0%; aPR 1.55, 95% CI 1.13–2.14). Although receiving provider counseling varied by age and WIC participation in 2016–2017 (Table 1), wearing protective clothing did not differ by receiving provider counseling during either study year (Table 2). The most common reason women reported for not wearing protective clothing was that it was too hot (>75%).

Most women also reported receiving provider counseling during pregnancy about repellent use in 2016 (92.0%) and 2017 (90.2%); receiving counseling did not differ by maternal characteristics. Frequent repellent use was reported by 45.4% of women in 2016 and 56.9% in 2017. In 2016, frequent repellent use was lower among women ≤19 years (47.6%; aPR 0.82, 95% CI 0.68–0.99) and 20–34 years (43.5%; aPR 0.75, 95% CI 0.67–0.85) compared with those ≥35 years of age (57.9%). During 2016–2017, frequent repellent use was higher among women receiving WIC (46.6%; aPR 1.29, 95% CI 1.07–1.57 in 2016; 58.4%; aPR 1.20, 95% CI 1.07–1.35 in 2017) compared with those not receiving WIC (36.0% in 2016 and 48.5% in 2017). Women who received provider counseling on using repellent were also more likely to report frequent repellent use compared with women not receiving counseling in both 2016 (46.1% vs. 37.3%; aPR 1.23, 95% CI 1.05–1.46) and 2017 (58.9% vs. 38.8%; aPR 1.52, 95% CI 1.29–1.78). The most common reason women reported for not using repellent was forgetting to apply or reapply it (>50%).

Most women reported being counseled by a prenatal healthcare provider during pregnancy on using repellent and wearing protective clothing to prevent ZIKV infection from mosquito bites. Provider counseling about repellent use was associated with a higher prevalence of frequent repellent use. This finding is consistent with our previous analysis of PRAMS-ZPER data, which showed receiving provider counseling was associated with a higher prevalence of condom use to prevent sexual transmission of ZIKV infection during pregnancy (13). In contrast, no significant association was found between receiving provider counseling and wearing protective clothing. Efforts to improve use of other risk-reduction strategies to prevent mosquito bites (e.g., repellent use, removal of standing water, screens on windows) may be beneficial, particularly when barriers, such as hot tropical climates, make wearing protective clothing less feasible. In 2017, the questionnaire was modified to include a question about using repellent on clothing in addition to exposed skin. In addition, changes in conditions after Hurricane Maria may have contributed to the increase in reported repellent use. During the ZIKV outbreak, WIC also implemented efforts to provide participants with targeted education on ZIKV prevention strategies and a prevention kit containing condoms, repellent, a bed net, and larvicide (14), which may partially explain increased use of repellent among WIC participants in our analysis. We found a significant association between WIC participation and frequent use of repellent but were unable to further assess the frequency or type of prenatal education WIC recipients received regarding repellent use. Of note, during 2016 only, women were asked whether they received a WIC Zika prevention kit; 77% reported receiving a kit, demonstrating the broad reach of WIC services related to ZIKV prevention among this study sample.

During prenatal care visits, healthcare providers can help prevent ZIKV infection by counseling pregnant women and their partners about risk-reduction strategies. Provider counseling on repellent and condom use were both associated with increased adoption of practices that reduce the risk of ZIKV infection (13). Findings from this study can be applied more broadly to the prevention of other vectorborne diseases among pregnant women, such as dengue, chikungunya, and malaria (7).