It’s Getting Easier to Get FDA Clearance — Just Ask Natural Cycles

Hannah Harris Greene, The Verge: December 19, 2018.


Advocates for health standards are concerned that Natural Cycles, a smartphone contraceptive app recently cleared by the US, has not been sufficiently scrutinized under Food and Drug Administration Commissioner Scott Gottlieb’s leadership. Its US clearance sets a precedent for more under-tested contraceptive apps to appear on the market, advocates fear.

Natural Cycles determines when users are fertile based on basal body temperature, and is the first contraceptive phone app to be cleared by the FDA. While this clearance may appear to increase the number of effective birth control methods available to patients, it has not been tested with nearly the same rigor as other FDA-approved contraceptives like the IUD or the pill, and therefore could be putting patients at risk for unwanted pregnancy and thus endangering their health. And when the FDA cleared the app, Natural Cycles had already raised concerns about unwanted pregnancies in Europe.

The FDA clearance of the app in August is unprecedented, because unlike other types of birth control approved by the FDA, Natural Cycles requires no medical intervention; instead, users take their temperature as soon as they wake up in the morning and feed it into the app — which then uses the information to calculate when the user is fertile. The app was cleared under a process specifically designed for new types of medical devices called De Novo classification. And this clearance may be part of a recent push by policymakers for lower standards. One example is the 21st Century Cures Act, a 2016 law designed to get more drugs and medical devices on the market more quickly. There’s also a laissez-faire attitude when it comes to digital health devices, of which Natural Cycles is one.

The De Novo process required a lot of time and multiple rounds of document submissions when it was first introduced in 1997. In 2017, when the FDA allowed applicants to request a less rigorous, more streamlined De Novo process, more than 30 new devices were cleared this way — more than any other year since the process was introduced. That is, until this year: there have already been 32 De Novo clearances in 2018.

Natural Cycles’ FDA clearance was based on data from more than 15,000 women, who used the app for an average of eight months, according to the FDA’s press release. The company has, so far, published three studies in peer-reviewed journals with external collaborators. The first, published in 2015, was in The European Journal of Contraception and Reproductive Health Care; the second, published in 2016, was in the same journal; and the third, published in 2017, was in the journal Contraception. “Our studies have been conducted in collaboration with scientists from leading research institutions including the Karolinska Institute in Sweden,” says Elina Berglund, CTO and cofounder of Natural Cycles, in an emailed statement.

At the time the FDA cleared Natural Cycles, the app was under investigation for exaggerating its efficacy after 37 users reported unwanted pregnancies at a single hospital in Sweden. (The Swedish Medical Products Agency has since concluded that the app works as it says it does, but required the company to clarify pregnancy risks.) Critics, including Diana Zuckerman, president of the National Center for Health Research, questioned whether the study Natural Cycles submitted was accurate because the company itself conducted it and stood to profit from positive results. It also included no randomized control trial, and the sample group of women was self-selected. “One of the study’s strengths is that it reflects real word usage of the app,” says Berglund in an email. “The results from the study are therefore relevant and applicable to women interested in hormone-free birth control.”

Others, including a former FDA scientist who spoke with The Verge on condition of anonymity, have noted that many factors can influence body temperature, including drinking habits, medication, or illness. Zuckerman agrees that the method might not be ideal for “college students or, you know, somebody whose life is a little bit complicated.”

Nathaniel DeNicola, an OBGYN and co-chair of telehealth at the American College of Obstetricians and Gynecologists, says that contraceptive apps like Natural Cycles could be “very appropriate for a lot of patients, they just need to know what they’re getting into.”

Mobile health applications from fitness trackers to fertility trackers have great potential to benefit patients, but an app that miscalculates users’ fertility window has more serious consequences than one that miscalculates the number of steps they took in a day, DeNicola says. While users avoid hormones, a daily medicine, or an inserted device like an IUD, pregnancy itself is inherently risky. And DeNicola isn’t confident that the study on Natural Cycles reflects the true risk of using the app, partly because it didn’t include any of the women who dropped out for unknown reasons, such as possible pregnancies. “I think it’s certainly safe to assume some of the patients that dropped out did drop out because they were pregnant,” he says. And if you assume all dropouts were due to unwanted pregnancy, that puts the risk at 10 percent, he says. Natural Cycles instead states that its typical user’s pregnancy risk is 7 percent.

“The study includes a sensitivity analysis, which assumes all women lost to follow-up as potentially pregnant,” Berglund says in an email. The difference between DeNicola’s read and Natural Cycles’ read is, essentially, optimism: both numbers appear in the 2016 study.

What’s more, the study wasn’t conducted in the US — and differences between Swedish and US populations may mean some of the data doesn’t translate. “There’s always questions about generalizability,” DeNicola says.

Zuckerman says the Natural Cycles’ efficacy research was closer to an observational study than a clinical trial — the company gathered data through its smartphone app rather than testing women in a clinical setting. And she says the FDA is lowering standards for devices in general.

The lowered standards are part of an FDA and congressional push to remove regulations that some politicians and special interest groups believe interfere with innovation, job growth, and sales of new products, Zuckerman says. She cites the 21st Century Cures Act as part of this push. Supporters of the law included Representative Diana DeGette (D-CO) and Representative Fred Upton (R-MI), who wrote an op-ed about it together, as well as Deloitte and the Advanced Medical Technology Association, which represents over 300 device manufacturers. “This is all part of pressure on the FDA that has nothing to do with its public health mission,” Zuckerman says.

The FDA is under pressure to boost the health care industry, says Erick Turner, a former FDA medical officer. He also views the 21st Century Cures Act as part of that pressure. The law was one of the most heavily lobbied in recent history. Nearly 1,500 lobbyists, including many from pharmaceutical companies, advocated for its passage, and ultimately only five senators voted against it. In his testimony on how the law should be implemented, Gottlieb, the FDA commissioner, states that the “FDA has embraced the concept of least burdensome regulation as clarified and expanded in the Cures Act,” when it comes to devices specifically.

“FDA has certainly taken the ‘least burdensome approach’ in terms of burdens on device makers,” says Zuckerman, “but that clearly means more burden on patients to try to figure out which of these devices are likely to work.”

Gottlieb, who Trump nominated for the position of FDA commissioner last year, has deep ties to the pharmaceutical industry and has written about how “the protection of consumers” should be given less importance and the FDA should instead focus on “[guiding] new medical innovations to market.”

The link between the FDA’s new path and Natural Cycles has even been made by the FDA itself. The Digital Health Innovation Action Plan, which, according to an FDA press release, provides guidance “in how the agency regulates digital health technologies like the Natural Cycles app,” states that the FDA will “not enforce compliance for lower risk mobile apps.” The FDA classified Natural Cycles as a class II or “moderate risk” device.

But Zuckerman and others disagree with the classification. “This device should have been considered a ‘higher risk’ device, since pregnancy can be dangerous for women and also can be financially devastating,” says Zuckerman. The consequences for an unwanted pregnancy are different in Sweden, where Natural Cycles was developed and where any patient can receive an abortion upon request up to 18 weeks into pregnancy, Turner says. But in the US, the availability of abortion is increasingly precarious. Indeed, the reports of unwanted pregnancies that triggered the Swedish Medical Product Agency’s investigation into Natural Cycles all came from women seeking abortions.

Deborah Kotz, a press officer for the FDA, said in an email that “the FDA was aware” of the pregnancy reports in Sweden. “We reached out to the Swedish authorities and felt that the information regarding the pregnancies in Sweden was consistent with our knowledge concerning the pregnancy risks associated with use of this device.” But the Swedish investigation of the device had not yet concluded when the FDA cleared it, and the UK Advertising Standards Authority has since forbidden Natural Cycles from billing itself as a “highly accurate” device.

In order to receive FDA clearance, future contraceptive apps will require “clinical performance testing,” but the guidelines don’t specify a minimum efficacy percentage or types of trial procedures. Zuckerman expects that testing standards will be lower for future contraceptive apps than they were for Natural Cycles, based on the FDA’s track record.

Devices should be tested like drugs, against a control group, Turner says. Natural Cycles was tested against nothing, so there’s no basis of comparison for how well it prevents pregnancy. Turner also worries that Natural Cycles is the result of FDA Commissioner Gottlieb’s stated intention to rely more on “real world evidence.” The evidence Natural Cycles used in its study all came from women who had purchased and downloaded the product and were using it in the real world, not from women being observed in a clinical setting. That’s different from medicine’s gold standard of evidence and a randomized controlled trial, which helps researchers understand what findings come from the product being tested and what might simply be statistical noise.

“I would agree that real world evidence, it can supplement clinical trials, but I’m afraid this is a slippery slope,” says Turner, who is concerned that the FDA might now believe “real world evidence can be a substitute for clinical trials evidence. That seems to be exactly what’s happening.”

Demand for apps like Natural Cycles is increasing: fertility-tracking apps are second in popularity only to fitness apps when it comes to mobile health, says ACOG’s DeNicola. When you consider that about half the population would need a pregnancy app for a window in their lives, that suggests wild popularity, DeNicola says. Given how ubiquitous these apps are becoming, doctors should ask if their patients are using them so that they can offer better information about the health risks, he says.

A study published earlier this year in mHealth showed that over 75 percent of 1,000 women surveyed intended to use fertility-tracking apps to prevent pregnancy in the future. Summer Starling, a clinical researcher at Lexitas Pharma Services and one of the study’s authors, thinks that app developers should focus on building better products for women and that women should be able to consult their doctors about these methods as they would with other contraceptives — instead of simply downloading them. Women want to know that fertility-tracking apps work. “What emerged really clearly from our analysis is just a strong desire among women that we surveyed to say, we really want something grounded in science, we don’t want something that’s pseudoscience,” Starling says. But when a body with the clout of the FDA is altering its standards for what qualifies as science, it’s hard to expect ordinary users to know what apps to trust.

 

See the original post here.