Knowledge and Prevention Practices among U.S. Pregnant Immigrants from Zika Virus Outbreak Areas

INTRODUCTION

An epidemic of the Zika virus (ZIKV) has been declared in many parts of Central and South America as well as the Caribbean.¹ After it was found that ZIKV infection was associated with microcephaly and other neurological birth defects among infants born to infected mothers in South America in early 2016, the World Health Organization declared a Public Health Emergency of International Concern.² In April of the same year, the Centers for Disease Control and Prevention (CDC) declared that ZIKV infection is a cause of microcephaly and other fetal brain defects.³ Although ZIKV is transmitted by mosquitoes, most cases in the United States have occurred as a result of individuals traveling to outbreak regions.⁴ There are also cases where transmission occurred through sexual intercourse with an infected person.⁵ As of December 7, 2016, there have been 185 mosquito-borne cases transmitted in the continental United States and 4,389 travel-associated cases reported.⁶

Even though ZIKV is now circulating in the United States, only a few studies have examined the knowledge level of pregnant women in America, including those at higher risk of exposure through their own travel or a partner who traveled to an outbreak area. Data from the currently published surveys are limited by their methodology and narrow scope and few have reported data for pregnant women. Moreover, most have not been published in peer-reviewed journals, but instead appeared in the lay press. These articles suggest low levels of knowledge in the United States. For example, a national poll of 1,004 households in March 2016 found that 40% of Americans had heard “only a little” or “nothing at all” about the ZIKV. Only 57% of those aware of ZIKV knew it could be spread through sexual intercourse with an infected person.⁷ An online survey of 300 U.S. citizens showed that ZIKV concerns are not stopping Americans from traveling abroad.⁸ Analysis of a ZIKV twitter chat hosted by the CDC in February 2016 demonstrated that participants were most concerned about how the ZIKV spreads and ZIKV infection-related fetal abnormalities.⁹

Few of the studies on ZIKV have focused on pregnant women, even though they may be at risk of delivering a child with severe health problems if infected. The largest study that included pregnant women surveyed 1,275 households in which someone was pregnant or considering becoming pregnant in the next 12 months.¹⁰ About 25% were not aware of the link between ZIKV infection and microcephaly and had little knowledge about how it was transmitted, prevented, or what the risk of symptoms were in infected individuals.¹⁰ A second study of 99 pregnant women found that 31% were unaware of the U.S. advisory to avoid travel to active ZIKV outbreak regions and nearly half did not know there was active ZIKV transmission in areas where they had traveled.¹¹ Neither study included information on women’s attitudes or their prevention practices other than limiting travel.

To address these gaps in the literature, we conducted a clinical survey of pregnant women attending prenatal clinics in southeast Texas in the summer of 2016, soon after the link between ZIKV infection and fetal brain defects was announced. These clinics serve a large number of women who immigrated to the United States from Central or South America and thus are at higher risk of ZIKV infection than women born in the United States as they are more likely to travel to countries with ongoing outbreaks or have partners who traveled there. The survey assessed knowledge, attitudes, and preventive practices related to ZIKV infection. By the time it was administered, the CDC had issued guidelines recommending that pregnant women limit their travel to countries with active outbreaks and that providers test pregnant women who did travel to one of these areas.³ To assess whether women originally born in countries with active ZIKV outbreaks had different views or practices that may affect their risk of infection, we stratified our findings by place of birth (born in an area with active ZIKV outbreak versus born in the United States). This information is needed to develop interventions that address the specific needs of high-risk groups.

MATERIALS AND METHODS

Pregnant women 18–50 years of age attending one of five reproductive health clinics in southeast Texas were invited to participate in an anonymous paper survey, which was available in English or Spanish, between June 14 and July 21, 2016. At the time of this study, information on ZIKV was not being actively distributed in the clinics. These clinics provide obstetric care to low-income women, almost 80% who have an annual family income < $30,000. Research assistants approached women in the waiting room before their clinic visit who had been preidentified as being eligible on the clinic census (currently pregnant and between 18 and 50 years of age). After confirming their eligibility, their interest in completing the self-administered survey was assessed before it was handed to them. Those who agreed to participate were allowed to select a gift for their baby valued at ≤ $5 as reimbursement for their time and effort. To ensure that participants completed the survey only once, study personnel maintained a cumulative database containing the clinic numbers of all women who had previously been approached across all clinics and compared it daily to those scheduled for a visit; previously approached women were not asked to participate a second time. The University of Texas Medical Branch Institutional Review Board approved all procedures and protocols for this study, including a waiver of written consent.

Questions were adapted from those published in the World Health Organization’s Resource Pack on ZIKV,¹² the National Health Interview Survey,¹³ the First Nations’ Knowledge of and Protection from the West Nile virus survey,¹⁴ and the Ipsos poll on ZIKV conducted for Reuters.¹⁵ Reading level of English questions was a Flesch–Kincaid grade level of 7.4. Self-reported demographic information including age, ethnicity, education level, country of birth, and gestational age were collected. This survey assessed respondents’ 1) knowledge of ZIKV, 2) past and future travel to outbreak regions, 3) exposure to mosquitoes, and 4) preventive practices against mosquito bites. We defined outbreak areas as any country in which local ZIKV transmission was declared as of July 2016 (when the survey was administered). Outbreak areas represented in our population included Mexico, all of Central America, Brazil, Colombia, Puerto Rico, and the Caribbean Islands.

Statistical analyses were conducted using Stata Statistical Software, release version 14.1 (StataCorp LP; College Station, TX). A two-sided P value < 0.05 was considered significant. Descriptive analyses included χ² and Fisher’s exact tests for bivariate analyses of categorical data and two-tailed t-tests for comparison of continuous data. Multivariable logistic regression models that controlled for age, education, race/ethnicity, and number of years lived in the United States were used to compare all significant results found in bivariate analyses. For those born in the United States, the number of years in the United States was equal to their age.

RESULTS

A total of 639 women were included in the final analysis. Women were excluded if they declined to participate, had incomplete surveys, or if they were not born either in the United States or in areas with ZIKV outbreaks (Figure 1

• Download Figure
• Download figure as PowerPoint slide

Inclusion of pregnant women in a survey about Zika who attended a reproductive health clinic in southeast Texas from June 14 to July 21, 2016. Percentages in flowchart are based on the total number of eligible women invited to participate (N = 876). ZIKV outbreak areas included Mexico, all of Central America, Brazil, Colombia, Puerto Rico and the Caribbean Islands. ZIKV = Zika virus. This figure appears in color at www.ajtmh.org.

Citation: The American Society of Tropical Medicine and Hygiene 97, 1; 10.4269/ajtmh.17-0062

Data from this survey indicated that there were several gaps in knowledge about ZIKV, and that women wanted to know more about it. Almost all women reported they had heard of ZIKV, but more than half of all participants wanted more information on ZIKV symptoms (58%), prevention measures (49%), or consequences of infection during pregnancy (42%; results not shown in table). U.S.-born women most often reported they trusted their physician for ZIKV information, whereas those from ZIKV outbreak areas indicated they trusted their family (Table 2

A small proportion of women from outbreak areas thought there was a cure for ZIKV infection. About half of all women knew about the CDC’s recommendation against traveling to outbreak areas during pregnancy, but had low knowledge about which regions had confirmed mosquito-borne outbreaks. Less than half had discussed travel restrictions with their health-care provider. Most women were aware that ZIKV was spread by mosquitoes and could be transmitted during pregnancy from a mother to her child. The majority did not know their risk of being infected by ZIKV or that it could be transmitted through sexual contact with an asymptomatic partner. Only half of the women understood the importance of using condoms to prevent transmission during pregnancy, with those born in the United States less likely to know this fact (55% versus 44%). Moreover, most did not know how often those infected by ZIKV were symptomatic. Few women knew the approximate risk of transmitting ZIKV to their fetus during pregnancy.

Almost half of surveyed women spent ≤ 2 hours a week outdoors, although a higher proportion of U.S.-born women spent ≥ 2 hours outdoors per week than women born in ZIKV outbreak countries (Table 3

More than half of pregnant women born in countries with ZIKV outbreaks were concerned about the safety of using repellent during pregnancy, whereas less than a third of U.S.-born women had this concern (Table 4

In multivariable models that controlled for age, education, and number of years that each participant spent in the United States, a number of significant differences were observed between women born in outbreak areas versus those born in the United States (Table 5

DISCUSSION

In this study, we found that women born in ZIKV outbreak areas had greater knowledge than those born in the United States of ZIKV’s signs and symptoms and associated birth defects. This difference could also have been due to more extensive provider counseling, or due to discussions with family still living in their birth countries. However, women from outbreak areas still had a number of important knowledge gaps. Over half were not aware of several signs and symptoms of ZIKV infection or did not know that infection was usually asymptomatic. Furthermore, although they were more likely than U.S.-born women (44%) to know that condoms could prevent transmission of the virus, only 55% knew this fact. This has important implications as pregnant women may not take precautions if they have sexual intercourse with a partner who could be infected, but does not have symptoms.

We found that 40% of the sample that was born in the United States and 24% from outbreak areas would keep a ZIKV infection secret if they or a family member contracted it. This implies that the disease carries a perceived stigma for infected individuals and their families, particularly for those born in the United States. One possible reason for this finding could be the extreme negative reaction of the public to U.S. residents, who developed Ebola virus infections during their travel to Africa in 2014.¹⁶ Although stigma is associated with other infectious diseases, such as human immunodeficiency virus, concerns about newly emerging infectious diseases are more comparable to each other as news about their outbreaks are often accompanied by controversy, scattered information about policy, and threats of isolation or quarantine.¹⁷ Education, particularly in U.S. states that are vulnerable to ZIKV outbreak, is needed to combat stigma of these types of diseases, as less educated people are at risk of failing to be diagnosed or may not share information with their sexual partners. This could be especially detrimental to a pregnant woman’s fetus, if the infected individual engaged in unprotected sexual intercourse.

Many of the study participants stated they would agree to be vaccinated against ZIKV, if it becomes available. Women born in outbreak areas favored vaccination over those born in the United States, possibly because they were more aware of potential consequences for their infants, if infected. In addition, the majority agreed that pregnant women should be offered ZIKV testing, demonstrating they would want to know if their infant was at risk. However, few would agree to an amniocentesis (defined as a test that inserts a needle through your belly to find out if your baby was infected) to diagnose fetal ZIKV infection. Some may have felt that this would be of little benefit since they knew there was no treatment or cure. It is also possible that some respondents did not understand that an amniocentesis would not harm the infant. Our findings are consistent with prior reports that acceptance of amniocentesis is significantly lower in Hispanic than Caucasian populations.¹⁸ In contrast to a study conducted in Latin America, women receiving care in these Texas clinics stated they would not seek an abortion if diagnosed with ZIKV infection.¹⁹ This makes it even more critical to educate women who live in states that may experience an outbreak on how to avoid this infection, as the birth of an infant with severe, permanent disabilities would likely impact the entire family.

Women born in countries with ZIKV outbreak reported less frequent use of mosquito repellant. This is concerning, as repellent is one of the few ways available to prevent infection and mosquito-borne circulation. This may have resulted from their concerns about its cost or its safety during pregnancy as almost half of the women surveyed were concerned that repellant could harm their infant. Moreover, 20% stated that its cost was prohibitive, demonstrating a need for programs to help pregnant women easily obtain repellant at a reduced cost or free of charge. These programs could be offered at clinics that see a high proportion of underserved patients, such as those in which this study was conducted. In fact, Texas announced a program in November 2016 that includes a provision for Medicaid benefits for mosquito repellant as a direct consequence of the threat of a ZIKV outbreak.²⁰

We found that low-income Texan women also need education on the type of mosquito repellant to use. The CDC strongly recommends using repellants that contain DEET in addition to Picardin, IR3535, oil of lemon eucalyptus, and 2-undecanon during pregnancy to protect against ZIKV infection. However, most women in this study were not aware of the active ingredients in the repellant they used nor did they know that DEET is safe to use. In fact, the U.S. Environmental Protection Agency placed DEET in the second lowest of the agency’s toxicity categories (Category III) and labeled it as a Group D carcinogen (not classifiable as a human carcinogen).²¹ Its safety during pregnancy has been echoed by the CDC and obstetrics/gynecology organizations.^22,23 In addition, 71% of the sample did not use permethrin even though the CDC recommended that pregnant women treat their clothing with this repellant/insecticide.²¹ Thus, most pregnant women in this study infrequently used these two simple techniques recommended by the CDC to avoid mosquito bites during pregnancy. This demonstrates that education on the type, safety, and importance of mosquito repellants and avoidance methods is urgently needed for those attending clinics that serve low-income pregnant women.

Another concerning finding is that more women born in outbreak areas reported that their family and friends were the source they trusted for accurate information on ZIKV, rather than doctors or the government. This has clinical implications as physicians and clinics are often the main provider of accurate information about diseases. However, women from both groups stated that their preferred method of learning about ZIKV was face-to-face communication with their health provider suggesting that they would accept physician education as a first-line technique to learn more about this particular disease. Women in these clinics may feel that it is most appropriate to learn about ZIKV directly from their provider.

Our study has some limitations. The sample was limited to low-income pregnant women in southeast Texas and was not representative of the general U.S. population. There was no information about whether the pregnant women had had previous visits for prenatal care, which limits the possibility that they had an opportunity to be exposed to their provider’s information about the virus. In addition, a cross-sectional study cannot be used to infer causality. The strengths of this study include a large sample of pregnant women from diverse backgrounds. Furthermore, we were able to sample women born in countries with active outbreaks who may be at higher risk of this disease than those born in the United States.

Overall, our study has important implications for the medical community. We found that pregnant women have many misconceptions about ZIKV and need more information on how to recognize disease resulting from infection. Those from outbreak areas are in urgent need of education on prevention practices, including the use of condoms if their partner has traveled to an outbreak area and use of repellents containing DEET or other proven and safe ingredients whenever they are outdoors. Treating clothing with permethrin is also an underutilized practice. Finally, the stigma of ZIKV infection needs to be addressed to allow early diagnosis and decrease the risk of transmission.