Gravid and COVID
As early as February, when SARS-CoV-2 was spreading across the country but long before the virus was declared a national emergency, Sonia Hernández-Díaz, a professor of epidemiology at the Harvard T. H. Chan School of Public Health (HSPH), realized the importance—and, at the time, the total absence—of research on the effects of coronavirus infection on mother and fetus during pregnancy. Hernández-Díaz studies how medications affect pregnant women, and she knew that this group is a special concern for doctors trying to understand and respond to COVID-19, the disease caused by the virus. Under any circumstances, respiratory problems exacerbate the immense physical stress that pregnancy can impose. And during outbreaks of earlier viruses, such as SARS-CoV-1 and influenza, pregnant women have been considered a vulnerable population because the immune response can sometimes cause the body to attack its own cells, in what is called a “cytokine storm,” which is in turn associated with miscarriages and preterm delivery. Hernández-Díaz saw an urgent need to discover whether COVID-19 worsens with pregnancy.
She began a study that asks women to answer questions about their health, pregnancies, and babies’ health. (Currently, 700 participants are enrolled; Hernández-Díaz plants to enroll about 10,000 women altogether.) The work, which will follow participants through 90 days after delivery, will take at least two years to complete. That pace may seem slow, acknowledges Hernández-Díaz, but she notes that “without [the urgency of] COVID, the information we are trying to produce would take decades. Now we are all putting all our energy and love” into getting informative results into the hands of “women and families and practitioners as soon as possible.” To that end, her team plans to release preliminary findings as the study progresses, especially if they discover anything that is urgently actionable.
Designing the research, Hernández-Díaz realized that COVID-19 would affect pregnant women not only through the virus itself, but also the circumstances of living through a pandemic: decreased access to prenatal care, upended childbirth plans, heightened anxiety, economic stress, isolation from friends and family, and so on. She reached out to her colleague Karestan Koenen, a professor of psychiatric epidemiology at HSPH, and together they designed a survey that would provide a baseline snapshot of the study cohort. During 10 days in June, more than 7,500 participants from 68 countries filled out an online survey about their mental health.
“What is most striking is the high prevalence of distress,” which is consistent pre- and post-partum, and across countries, Koenen wrote in an email discussing the survey. The data are still being analyzed. More than 70 percent of the women reported clinically significant depression or anxiety, and upwards of 40 percent screened positive for post-traumatic stress disorder—rates several times higher than pre-COVID-19 studies suggest is the norm for pregnant women. More than half reported high levels of loneliness. Media exposure and worries about the pandemic and pregnancy, in particular, correlated with clinical distress. “Our results highlight that equal attention needs to be made to women’s mental health while protecting their physical health,” Koenen wrote. “We know that maternal mental health during pregnancy and post-partum has long-term effects on offspring development—so protecting maternal mental health is critical.”
Despite the efforts of scientists around the world to study the relationship between pregnancy and COVID-19, collecting data and arriving at reliable conclusions is challenging. One problem is sample size: “Even when we have a large proportion of the population infected,” Hernández-Díaz explains, “only a small proportion are women of reproductive age. Of those, only a small proportion are pregnant.” And many researchers rely on already-burdened clinicians to voluntarily report pregnant women with COVID-19, making data-gathering more difficult. Hernández-Díaz’s study’s instead uses an online registry to collect information directly from women, rather than going through physicians, thereby circumventing some of these challenges. This approach also increases the sample size, and potentially could lead to unique research results.
For months, Hernández-Díaz says, the medical community has had to base its conclusions on a patchwork of findings from around the world: a handful of cases from a hospital in one country, a few dozen from another. These studies lack control groups and significant sample sizes, but imperfect data are necessary during a pandemic—doctors and public-health practitioners need all the information they can obtain to optimize treatment for their patients.
One study in particular bears out her point about the difficulty of parsing the information available right now. In late June, the Centers for Disease Control and Prevention (CDC) published the largest study to date of pregnant women with COVID-19, with more than 8,000 participants. The research found that pregnant women with COVID-19 were more likely to be hospitalized, admitted to an ICU, and receive ventilation. Media headlines across the country erupted, exclaiming that the CDC found pregnant women with COVID-19 had an elevated risk of severe illness—contradicting the prevailing medical consensus.
In fact, after analyzing the available papers relating to pregnancy and COVID-19, she says, medical professionals have concluded that “there is no reason to think that pregnant women have more severe infection.” (She notes that mother-to-fetus transmission of the coronavirus is unlikely but has been reported in a few instances.Physicians have also documented increased rates of premature births in some countries, but are unsure about the cause.) The increased hospitalizations in the CDC report, Hernández-Díaz says, don’t imply worse outcomes for pregnant women. The CDC researchers counted childbirth-labor hospitalizations toward the total, but even under normal circumstances pregnant women are hospitalized more frequently because of delivery. It is also unclear, she adds, whether other statistics in the CDC analysis imply more severe symptoms: if two women, one pregnant and one not, are displaying the same symptoms, a doctor is more likely to put the pregnant woman on a ventilator.
“[The CDC report] made the news,” Hernández-Díaz says. “And most of us in the field were like, ‘What are they saying? There is no reason to say that.’” But the study is still valuable, she says—just not for the reasons most frequently reported. And even though COVID-19 symptoms for pregnant and non-pregnant women are similar, she agrees that pregnant women should be considered a vulnerable population simply because of the “stress test” their bodies go through: respiratory complications during labor, for instance, can become much more serious. “The same symptoms,” she says, “are not the same symptoms.”
The misinterpretation of the CDC study illustrates how keenly the public and medical professionals alike are grasping for information in the midst of crisis and uncertainty. Scientists are trying to balance the urgent need for data against the scientific mandate to conduct careful and responsible research—hence the abundance of small case studies with no reference group. These papers are important, but must be interpreted with caution. “Because of the [public health] emergency, the journals are publishing very quickly, sometimes without peer review,” Hernández-Díaz explains. “They’re publishing everything.” As always, she notes, there is an incentive for academics to publish—it is good for their careers, and she believes most do so with good intentions. But with so few cases of infected pregnant women available to analyze, reviews—rather than new studies—have proliferated. “I have to screen through 2,000 papers to find the 10 that are providing information.”
Her own study is part of an effort by an international research consortium that involves 80 investigators from across the globe who study pregnancy and COVID-19. “The plan is to put all of the cohorts together,” Hernández-Díaz says. “We will harmonize the collection so that we can put some numbers together.” Combining their data, she hopes, will soon produce stable estimates and accurate conclusions that can help treat soon-to-be mothers and infants around the world.