A few weeks ago, the Boston tech firm Whoop made a startling claim about its fitness trackers: It said the colorful wristbands can detect a heartbeat change in pregnant women seven weeks before labor begins, indicating that wearers may deliver a baby prematurely.
LeBron James and Michael Phelps have worn Whoop straps to monitor their heart rates, breathing, and sleep, but this announcement had broader implications. Whoop had discovered a novel biomarker for a global public health hazard, the firm said, and it began promoting the tracker for “pregnancy coaching” on social media, its website, and other platforms.
“We have an indication about whether or not [pregnant women are] going to have a premature baby or if they’re going to have a full-term baby,” said Emily Capodilupo, Whoop’s senior vice president of data science and research, in a video on Instagram. She was among five researchers from Whoop and the West Virginia University School of Medicine who reported the discovery in a yet-to-be published study of 241 pregnancies.
Will Ahmed, founder and CEO of the 10-year-old company, rhapsodized about the findings in the video. “To be able to tell women for the first time, ‘Hey, we’re seeing things that indicate you need to talk to your doctor,’ I mean, that’s an unbelievable discovery,” he said.
Now, however, those breathless claims are drawing criticism as being premature and scientifically shaky. The complaints came on several fronts ― from a former Whoop employee who worked on the pregnancy feature, an obstetrician-gynecologist at the West Virginia medical school who coauthored the study, and two prominent physicians with expertise in preterm births.
Keely Byron served as product manager for the feature but left Whoop in October before it was unveiled, in part because she believed the company rushed the effort and relied on flimsy science. She shared her concerns with a supervisor in her exit interview, she said, but they went unheeded. On Nov. 9, Whoop issued a news release headlined, “Whoop Identifies Novel Pregnancy Digital Biomarker to Screen for Premature Birth.”
“I was surprised to see how deceptive Whoop’s Instagram and direct-e-mail marketing on the pregnancy coaching feature is,” Byron, who now works as a contract employee for the drug maker Genentech, told the Globe. “A layperson would very reasonably look at these ads, and their takeaway would be that Whoop can help them identify if they are at risk for a premature delivery, which is not what the product does at all.”
Dr. Shon Rowan, the West Virginia OB-GYN and only physician among the study’s authors, said the discovery that an expectant mother’s heartbeat changes seven weeks before delivering both full-term and premature babies is tantalizing. But Whoop’s marketing claims have made him feel “a little leery,” he said. The study was posted online in November but has yet to withstand the scrutiny of peer review, a requirement for publication in a scientific journal. “From a scientific standpoint, I don’t think we’re there yet,” Rowan said.
Whoop said in a series of e-mails that the firm stands by its research and product. As for the company’s claims about the pregnancy feature, Whoop insisted that the firm qualified its discovery, saying it “has potential” and “could” lead to screening for preterm births.
“We do not make claims in the marketing campaign that the device ‘can predict premature births,’” the company said. Whoop said it plans to do more research and that for now, expectant mothers can use the feature to see their fitness trends “contextualized” with their pregnancies.
Whoop added that the feature explicitly says the firm “does not provide medical advice and its services are not intended to be a substitute for advice from your doctor or other medical professionals.”
A number of physicians have raised doubts in recent years about whether tech firms’ claims for wearable devices ― a $52 billion global market last year, by one estimate ― are leaping ahead of medical science. Just Monday a cardiologist at the University of California at San Francisco who was among the first to explore the potential of smartwatches to detect potentially dangerous irregular heart rhythms wrote in the journal Circulation that such products need more study to be considered reliable.
For Whoop, the controversy comes at a sensitive time. The privately held firm has generated considerable buzz and in 2021 had an estimated market value of $3.6 billion. Unlike makers of other fitness trackers, Whoop uses a subscription model; customers pay $30 a month, which includes the screen-less wristband and access to a smartphone app. But in July, Whoop announced it had laid off 15 percent of its employees due to uncertainty in the tech market, leaving it with roughly 500 workers.
Cofounded by Ahmed a decade ago when he was a student at Harvard University, where he was captain of the men’s varsity squash team, Whoop sells wristbands that, among other things, measure “heart rate variability,” or HRV, the tiny variations in time between beats of the heart. If someone’s heart rate is 60 beats a minute, according to Whoop’s website, the heart doesn’t actually beat once every second. Within that minute, there may be, say, 0.9 seconds between two beats and 1.15 seconds between two others.
HRV may appear to be a function of the heart, but it actually originates in the nervous system and is an important measure of fitness, according to Whoop. People with greater heart rate variability, says the company, are typically in better shape and more prepared to exert themselves. The band also tracks resting heart rate, or RHR, another important measure of conditioning, Whoop says. Generally speaking, a lower resting heart rate is a good sign.
Rowan said he turned to Whoop a few years ago for a small study on the benefits of exercise in pregnant women. With a grant from West Virginia University, he paid Whoop about $20,000 for wristbands that were worn by 18 women during and immediately after pregnancies, he said. The fabric bands have sensors and collect health data as people wear them round the clock. The data are automatically uploaded to a Whoop app.
All 18 women carried their pregnancies to term. (A pregnancy is considered full term at 37 weeks, according to the American College of Obstetricians and Gynecologists.) Given the stress that a growing fetus places on a woman, Rowan said, he wasn’t surprised that the volunteers’ cardiovascular health ― as measured by HRV and RHR ― deteriorated during pregnancy. But something unexpected happened seven weeks before delivery, he said. The trend reversed itself, and their fitness, as evidenced in HRV and RHR data, rapidly improved before delivery and afterward.
“Why would it reverse?” said Rowan, lead author of the study written by five West Virginia University researchers and recently published in the journal BMC Pregnancy and Childbirth. “I’ve spoken to a few people that I think are amazing minds on the physiology of pregnancy. Nobody can come up with an answer.”
After that study was completed, Whoop launched a second study, this one conducted by Rowan, a university colleague, and three researchers from the company. The firm collected retrospective data of 241 women who wore the wristband during their pregnancies, including 21 whose babies were born prematurely. The study showed that regardless of whether the babies were born on time or prematurely, the mothers’ cardiovascular conditioning deteriorated until seven weeks before the onset of labor, when it rapidly improved. So if a woman noticed the reversal in, say, her 25th week of pregnancy, it could indicate her baby would be born prematurely, at 32 weeks.
Whoop called the sudden reversal the “Whoop Inflection Point” and said the dynamic “has the potential to save lives.” Roughly 10 percent of US babies are born prematurely, according to experts. Such births make up the second-leading cause of infant deaths after birth defects, and often contribute to health problems in surviving babies.
Many physicians would welcome an affordable, reliable, and noninvasive way to screen expectant mothers for impending premature births. Currently, they can perform an ultrasound in the second trimester to see if a woman has a shortened cervix, which increases the risk of preterm labor. Other physicians take a sample of vaginal fluid to check for a protein that helps the amniotic sac stick to the uterus; if present in the fluid, it can signal a likely preterm labor.
A wearable that could screen for preterm births might need to be cleared by the US Food and Drug Administration, according to Frances Miller, a professor at Boston University’s law school and expert on the FDA. But the burgeoning wearable market occupies a gray zone in agency regulations, she said, and “the FDA can’t possibly keep up.” Whoop said it would do what the FDA required if the feature needed agency clearance.
Two prominent OB-GYNs with expertise in premature births said it will take far more than the unpublished study of 241 pregnancies before they would recommend the Whoop wristband.
Dr. Michael Gravett, a professor at University of Washington Medicine, said the study, while intriguing, didn’t address a number of factors known to increase the likelihood of premature births, including the race, ethnicity, and socioeconomic status of patients. Researchers, he said, “have to account for the multi-factorial nature of preterm birth.”
Dr. Thomas McElrath, who runs the preterm birth clinic at Brigham and Women’s Hospital, said that not only does the study need to be peer reviewed, it needs to be repeated in a different group of pregnant women to make a compelling case that the heartbeat change can signal premature birth.
Meanwhile, another OB-GYN cited in Whoop’s press release last month is walking back her comments a bit.
Dr. Elizabeth Cherot, chief medical officer of New Jersey-based Axia Women’s Health, was quoted as saying the Whoop study was “incredibly significant” in light of the problem of premature births. Eight days later, Cherot was named chief medical officer of the March of Dimes, a prominent nonprofit group seeking to prevent premature births.
In a recent Globe interview, Cherot acknowledged that the study hadn’t been peer reviewed and had other shortcomings. She said she wasn’t endorsing the product, just the research effort to find an innovative, noninvasive screening tool.
“I should have said I’m incredibly excited,” she said.
Jonathan Saltzman can be reached at firstname.lastname@example.org.