Contraception can be a charged topic for many reasons. It’s highly personal, of course. Frequently it’s political. And the stakes, should your method of choice fail, are obviously very high indeed.
Which perhaps explains why there hasn’t been a whole lot of innovation in the space — unless you count the prospect of a male pill which nonetheless still hasn’t arrived on the market.
None of these challenges have deterred European startup Natural Cycles from pushing to establish what it bills as a “digital contraceptive” into the market as another method for couples to consider — assuming they’re comfortable having unprotected sex (which of course means no protection against the risk of STDs).
Last July the team announced a $6M Series A round. Today they’re reupping with a $30M Series B led by EQT Ventures fund, with participation from existing investors Sunstone, E-ventures and Bonnier Growth Media (the VC arm of privately held Swedish media group, the Bonnier Group).
Back in February the product also gained certification in the EU as a contraception — giving a regional boost to its ambition to disrupt the market for contraception choices.
Their subscription product, which now has 500,000+ users across ~160 countries, is a fertility tracking app fed by algorithms working off-of individual data inputs, including daily temperature measurements via a body basal thermometer (this device is included in the subscription price if users sign up for a year).
The product aim’s to predict on which days a woman is fertile and which she’s not. A user sees a calendar view in the app with certain days of the month colored red — to signify she shouldn’t have unprotected sex that day unless she does want to try to get pregnant.
Days when the algorithm judges there’s no risk of pregnancy are colored green — to signify the system is confident that it’s okay for the woman to have unprotected sex and not risk pregnancy.
For Natural Cycles to work women need to measure their body temperature with a two decimal basal thermometer each morning (ideally), and do so before they get up — recording the data in the app. It then uses these readings as an indirect measure of hormone levels, and uses that to calculate each woman’s fertility (factoring in other relevant factors such as sperm survival rates).
The algorithms powering the product function by seeking to understand probabilities and uncertainty around each individual, says co-founder Dr Elina Berglund, who previously worked as a particle physicist at CERN.
“What the algorithm is really trying to assess is to understand how well do we know this woman? Because every women is unique and her cycle is unique, and you need to apply statistical scrutiny to understand when can you definitely say that if you have unprotected intercourse today it will not lead to pregnancy,” she tells TechCrunch.
“And that is all about probabilities. So you have to combine the probability of sperm survival, how well you know the ovulation pattern of this women, how well you see the temperature pattern for this particular woman to detect do you really know that she has ovulated and how well do you assess when she has ovulated and to be able to predict that.
It’s all about understanding the uncertainties around the data and what this woman’s unique cycle looks like.
“So it’s all about understanding the uncertainties around the data and what this woman’s unique cycle looks like.”
A woman’s body temperatures can of course fluctuate for all sorts of reasons — such as sickness or even drinking alcohol. While the menstrual cycle can also vary in length, month to month, or be otherwise irregular for all sorts of lifestyle-related or health reasons.
Berglund says Natural Cycles’ algorithms are designed to interpret and accommodate these sorts of individual differences and fluctuations without a reduction in the product’s overall efficacy exactly because it’s engineered to take such uncertainties into account when predicting a woman’s fertile/non-fertile days. (Though the website FAQ does warn the product may not be suitable for women with very irregular cycles.)
The algorithm also adapts if users do not input their temperature every day, as they are supposed to in the ‘perfect use’ scenario.
“The accuracy rate of course takes all this into account,” she says. “There’s no women that takes her temperature every day. On average she takes her temperature 73 per cent of the days. But of course you have a large spread on that — some women only take it 40 per cent of the days and some do it to 95 per cent. And this method failure rate is based on all that real live data. So it’s not based on women that take their temperature on 100 per cent of the days.”
There is also an initial learning period for the app, with each woman starting off being given more red days while the algorithms wait to be fed more data. Gradually the user will get more green days — but obviously there will always be some red days per month.
“The very first cycle there’s usually quite a lot of red days, so you only get 40 per cent green days on average the first cycle, and then it quickly ramps up,” says Berglund. “Usually after nine cycles you reach a point where you don’t get more green days on average. But this is just the average so it looks very, very different from woman to woman.”
In one published study of the product’s efficacy, which was carried out by Natural Cycles utilizing existing users of the product as study subjects (and following up, she says, with 99 per cent of those who quit using it to try to determine if they got pregnant), they say it demonstrated 93 per cent efficacy with “typical use” — although the self-selecting nature and structure of this study does not achieve the scientific ‘gold standard’ of a randomized control trial.
Nor has Natural Cycles carried out a randomized control trial comparing the product against other contraceptive methods. So users wanting to know how Natural Cycles stacks up against the pill, for example, or vs using condoms do not yet have any way to know that.
And because the product remains untested at the core standard used for scientific research there has been some criticism of the efficacy claims Natural Cycles are making.
“I have a number of concerns about the current claim of 93 per cent typical use efficacy,” says Victoria Jennings, director of the Institute for Reproductive Health at Georgetown University, who has been involved in several efficacy studies for fertility awareness-based contraception methods (she has also worked directly on research involving a rival fertility awareness app, called Dot). “As the published paper indicates, there is a significant amount of missing data, and drop-out rates are quite high. The study was not designed according to the standards of an efficacy trial, so it seems inappropriate and misleading to offer this claim.
“The authors themselves state that the results cannot be compared to efficacy of other contraceptive methods but the purpose of a trial is to be able to do exactly that. A study should be designed so that it can be comparable. What we have here is much more in line with post-marketing research, which is useful in understanding the real-world experience of a large number of people, but not the basis for efficacy claims.”
“Any [fertility awareness] app that makes efficacy claims should have been examined in a standard efficacy study,” she adds.
Asked about whether it will be using the Series B funding to carry out a randomized control trial, Berglund tells us Natural Cycles would like to do this to back up its efficacy claims. She also says it did previously, a year or more ago, apply to carry out a comparative randomized control trial against the contraceptive pill. But its application was rejected by the local medical product agency, in Sweden, for “several reasons” — including that “they didn’t see the purpose for such a study”.
It’s now hoping to carry out a comparative study in future against a condom, which she says she hopes “might be easier to recruit women who are willing to just use condoms or use condoms together with Natural Cycles”.
“We do wish to do it still,” she adds. “We would now try a second round but instead of doing it against the contraceptive pill, because that seemed to be too controversial, we would do it against a condom.”
That said, there’s no firm timeline for the study at this point — beyond Berglund saying: “It will take several years to carry out this trial.” So users wanting to see the product’s claims backed up by standard efficacy research have longer to wait.
“We’re happy that so far we’ve done the largest study ever published concerning natural birth control, so we do feel that as a startup, as a company, we do perform a lot of research,” she adds. “But we do want to achieve even more — and we realize also that this comparison trial is the one thing that is missing.”
Another criticism which some women’s reproductive health experts we spoke to have directed at the product is that there’s a risk of it being misleading — given its relative complexity to use and the need for alternative contraception on red days (or else that users abstain from sex).
Berglund rejects this view, arguing that they are providing users with clear guidance and information via the app, including on relative pregnancy risks for using different alternative contraception methods.
“We try to be very clear in our communication that if you do use withdrawal [as a contraceptive method] on red days then your pregnancy probability is much higher — and we inform them about that inside the app. Because we ask them what they use, and they tell us,” she says.
“We’re really trying working on being very transparent… Who is best for, who is not best for. We have an 18-year-old limit, so no one below 18 can use the app. So I think we’re very good at providing also an effective alternative to women but also being very transparent about the pros and cons of our method to the women.”
“The app is just a tool for women to use how they see fit,” she adds. “It puts the information in their hands, understanding their body when they’re fertile, when they’re not fertile. But, in the end, it’s up to them to use this tool how it suits them and their lives.”
Who are the users?
At this point the average age of a Natural Cycles user is 29-years-old. While the renewal rate after the first year’s subscription currently stands at 60 per cent.
“Most women are between 25 and 35, they’re in a stable relationship, they’re often thinking about that they want to have children in a few years and they don’t want to take hormones anymore,” says Berglund. “Most of our users do come from hormonal contraception — and they’re thinking I’ll use this method in the mean time to also understand how my body works.”
The product can also be used by women to help them try to get pregnant, though she says that’s currently a minority of users (circa 25 per cent).
“We often see that after about two years [users] then switch to use the app to get pregnant, and then they get pregnant very quickly because they already know their body and the app knows them as well,” she adds.
Another feature gives women an incentive to continue using Natural Cycles for the first three months after they get pregnant — as the app offers a miscarriage alert related to risks of hormone levels dropping during pregnancy.
The algorithm can also prompt a user to take a pregnancy test — based on its calculations of their hormone levels (so could figure out they’re pregnant before they do).
The plan for the Series B funding is to spend on clinical research and product development — including growing the team with additional recruits from medicine and science. (Berglund says Natural Cycles is currently 40 people, including seven physicists PhDs, with the goal to grow headcount to between 100 and 120 staff in a year’s time.)
The research portion of the funding won’t just be going towards conducting the aforementioned, eventual randomized control study; the team also wants to investigate various aspects of women’s reproductive health and fertility to feed its product development.
“We’ve published three clinical studies to date and we have a fourth one on the way that is measuring what affects time to pregnancy and infertility among our users that are trying to conceive but we want to do even more than that,” she says. “We have quite an ambitious research plan, looking at acceptability — like which contraceptive method is suitable for which type of person and when in their lifetime.”
International expansion is another focus for the funding. The US market is Natural Cycles’ third biggest at this point, after its home market of Sweden and then the UK — where Berglund says the product is seeing its fastest growth.
“We really want to establish ourselves as a global leader in digital contraception — so expansion to other markets is key,” she adds.
Its European business undoubtedly got a big boost when the product gained certification as a contraception in the European Union. Although this certification is specific to the region, and while Berglund confirms the company has applied for the equivalent accreditation in the US, via the FDA, it’s now having to wait and see if that’s successful.
And she concedes the wait for an FDA stamp of approval on a contraceptive app could take some time.
“Since this is an innovative and new product it can take quite a lot of time. I think it’s been 15 years since the last time FDA approved a medical device for contraception. So we do believe it can take a while but we have started the process,” she notes, adding: “We have submitted all of our documentations and our clinical data etc so we have done that work already on our side — but of course we have to wait and see what they come back with.”
There is perhaps inevitably a degree scepticism when loud marketing claims are attached to new product which — at least in part — is leveraging an existing approach that isn’t generally considered as reliable as accredited (and tested) in-the-market alternatives. Especially when the company in question hasn’t yet done the standard science required to robustly underpin its claims.
Though, on the fertility awareness-based tracking point at least, Natural Cycles claims its algorithms are doing more than just the standard calendar method of tracking fertile days. And are thus, it says, more accurate than a person could be if they tried to track their fertility on their own.
“It’s very important not to do this analysis by hand because many studies have shown that even if you are good at applying simple rules in your own temperature tracking, 25 per cent of the women will end up pregnant in a year,” claims Berglund.
Although she also says the original idea for the business came after she and her husband had been seeking a natural contraception method — and she developed the original algorithm to use on herself, so was effectively manually tracking her own fertile days (albeit with a physics PhD to underpin her methodology).
“We care about our users,” she adds. “We wouldn’t have released [the product] — and we wouldn’t have been certified for contraception if we wouldn’t have done this [comparison with the standard calendar method algorithm] either.
“Part of the certification is a lot of documentation and studies that are not published but are reviewed by the oversight body that gave us the certification for contraception and has approved our marketing claims that our algorithm’s method failure rate is just 0.5 per cent per year.”
At the end of the day contraception is a personal choice and existing options do already vary greatly in accessibility, complexity and more besides. Natural Cycles’ bet is that a sizable proportion of couples will be open to an app-wrapped addition joining their mix.
This article was updated to correct the date when Natural Cycles obtained EU certification; it was in February 2017, not August
Comment