Even after thousands of studies, questions still linger about one of the most common diseases unique to pregnancy: preeclampsia. This disorder leads to dangerously high blood pressure in about 5 percent of U.S. pregnancies, with significantly higher rates in Black women. And it is becoming more common.

The only known cure is delivery, which creates a serious dilemma: the longer that people with preeclampsia remain pregnant, the sicker they get—but the longer a fetus gestates, the healthier it will be at birth. It is challenging for doctors to predict how intense a case will be in order to make individual treatment decisions.

A study in NEJM Evidence offers a way to predict whether pregnancy-related high blood pressure will deteriorate into severe preeclampsia, which can cause organ failure, vision loss and stroke. The new research focused on the balance between two pregnancy proteins, says lead author Sarah Kilpatrick, chair of obstetrics and gynecology at Cedars-Sinai Medical Center in Los Angeles. A high level of the protein PIGF (placental growth factor), which stimulates placental growth, is good. A high level of the protein sFlt-1 (soluble fms-like tyrosine kinase) is bad—and is known to rise well before a patient shows signs of preeclampsia. But high sFlt-1 numbers alone don’t predict progression to a serious case.

To determine how the proteins’ levels align with sickness, Kilpatrick’s team studied 1,014 racially and geographically diverse high-risk patients between 23 and 35 weeks of gestation. All had already been admitted with high blood pressure to one of 18 urban or suburban hospitals.

The team found that when the ratio of the “bad” to the “good” protein was 40 or more, patients had a high chance of developing severe preeclampsia within two weeks. When the ratio was under 40, that chance was less than 5 percent. Kilpatrick suggests doctors could monitor a high-risk patient at home if the ratio is, say, 10, whereas someone with a ratio of 100 could be transferred to a hospital that can manage preeclampsia complications as well as a preterm baby. “For an incredibly complicated disease, that’s really the clinical benefit of knowing this risk,” Kilpatrick says. “To me, that’s probably the most important thing.”

Right now there is no fast, decisive preeclampsia test, says William Grobman, a specialist in maternal-fetal medicine at the Ohio State University who was not involved in the research. Although this study doesn’t remove all uncertainty, he says, it offers clinicians “a better understanding of who is going to deteriorate with that condition in a given period of time.” Grobman looks forward to seeing if the results can be replicated. Kilpatrick, meanwhile, views this study as a key step toward FDA approval for such a test.