A twenty-two year old Brazilian woman gives birth to her second daughter and is surprised in the delivery room. Her baby girl’s head looks drastically different from her first daughter’s. What happened? she asks her obstetrician. She did not have any ultrasounds in her third trimester, so the diagnosis of microcephaly is a surprise. However, she does recall a few days at the end of her first trimester when she had a mild fever and a rash. She also recalls many mosquito bites throughout her pregnancy. Her doctor tells her that she was probably infected with the Zika virus. Can her doctor tell her that Zika caused the microcephaly? Have scientists proven a causative connection?
Scientists encounter the problem of proving causation every day. As we all learned in Intro to Statistics, correlation does not equal causation. Ice cream sales spike at the same time as shark attacks- but we all know that eating ice cream does not cause shark attacks. In 2015 Brazil experienced an epidemic of infections with Zika virus, concurrent with a sudden uptick in cases of microcephaly. Can scientists now conclude that Zika causes microcephaly? The scientific community has come to the conclusion that yes, Zika causes microcephaly. To make this conclusion confidently, thought leaders used a set of criteria called Shepard’s 7 Criteria. Published in 1994, Shepard’s 7 Criteria were established to allow scientists to draw a causal connection between an exposure and a clinical outcome. Here is how scientists fulfilled the seven criteria for Zika-associated birth defects:
1. Exposure must occur at a critical time during prenatal development. It stands to reason that the timing of infection with critical periods of neurodevelopment should impact the results. Using case series, researchers were able to demonstrate more severe infection (like microcephaly) when infection occurred earlier in pregnancy, during the most critical periods for fetal brain development.
2. Two or more high-quality epidemiological studies must report consistent findings. Two epidemiological studies on birth defects and Zika focused on the two most recent outbreaks: French Polynesia and Brazil. The study in French Polynesia showed a 1-13% chance of microcephaly if a woman was infected with Zika in her first trimester (although there were only 8 cases studied). The Brazilian study found a 29% risk of abnormalities on prenatal ultrasound for women with Zika virus infection during pregnancy. Although these studies do not match exactly, the connection between an increased risk of microcephaly and Zika virus infection is consistent. In the months to come, new epidemiological studies will likely further support these findings. (Although, arguably, we are still awaiting more evidence to truly fulfill this criteria.)
3. The microorganism must produce a specific syndrome. In the case of Zika virus, a distinct syndrome is emerging including microcephaly, redundant scalp skin, ocular abnormalities and blindness, joint contractures, clubfoot, and other central nervous system abnormalities.
4. A rare environmental exposure must be associated with a rare birth defect. Microcephaly is extremely rare; it occurred in about 6 in 10,000 infants in the United States. Prior to these epidemics, Zika was also extremely rare. After the Zika epidemics in Brazil and French Polynesia, the occurrence of cases of microcephaly far exceeded previous reports of this very rare defect.
5. Studies in laboratory animals must demonstrate teratogenicity. Last week, reports of an experiment with macaque monkeys demonstrated brain abnormalities in fetuses of monkeys infected with Zika virus during pregnancy.
6. The exposure and anomaly must be biologically plausible. Researchers have firm evidence for a biologically plausible pathway for Zika to disrupt neurodevelopment. First, two similar viruses, rubella and cytomegalovirus, have long been known to causes similar syndromes. Second, Zika virus has been identified in the amniotic fluid of affected pregnancies and the brains of affected infants. Third, animal and in vivo studies have shown that Zika virus is capable of infecting nerve cells.
7. The agent must act in an unaltered state in an experimental system. This seventh criteria is for medication or chemical exposures, so it’s not relevant to Zika virus.
Scientists, including those at the CDC, are confident that Zika virus causes microcephaly. Now we are tasked with learning how to stop this growing epidemic and keep pregnant mothers and infants safe. We want to hear from you. Which criteria is most convincing of causation to you?