Background
Summary of Council decision:
Two issues were investigated, both of which were Upheld.
Ad description
A paid-for post on Facebook for Natural Cycles, seen on 20 July 2017, stated “Natural Cycles is a highly accurate, certified, contraceptive app that adapts to every woman’s unique menstrual cycle. Sign up to get to know your body and prevent pregnancies naturally”. A video below the text also stated “Natural Cycles officially offers a new, clinically tested alternative to birth control methods”.
Issue
Three complainants challenged whether the following claims were misleading and could be substantiated:
1. “Highly accurate contraceptive app”; and
2. “Clinically tested alternative to birth control methods”.
Response
1. & 2. Natural Cycles provided copies of: their EC medical device certification; a prospective observational study looking at the efficacy of the app using the Pearl Index; two articles on retrospective studies carried out; and a fact sheet that was available on their website.
Natural Cycles referred to the Pearl Index (number of pregnancies per 100 woman years) which was a widely used measure to summarise contraceptive effectiveness. The perfect-use Pearl Index measured the number of unintended pregnancies when the contraceptive method was used as instructed, while the typical-use Pearl Index measured the number of unintended pregnancies when using the contraceptive method in practice.
They said that the app was certified and CE marked and had been marketed as a contraceptive since February 2017. They explained that the certification was based on clinical data which demonstrated the effectiveness of the app as a contraceptive and the clinical claims for the medical device were based on clinical studies that they had carried out. The app was stand-alone software intended to be used for contraception and fertility monitoring and would inform the user whether and when they could get pregnant.
They stated that the clinical findings of the app had also been summarised by an independent group of researchers in the field of contraception and reproductive health. The app was based on the basal body temperature (BBT) method, which was a commonly known variant of fertility awareness based (FAB) methods. The algorithm in the app had been developed by Natural Cycles. Alongside the app the user would require a thermometer.
The app required the input of BBT recordings and the date of menstruation. The underlying technology was a statistical algorithm which returned a red (fertile) or green (non-fertile) day to the user depending on whether or not they would be considered to be at risk of getting pregnant. The algorithm computed the following parameters and their uncertainties: ovulation day; luteal phase; follicular phase; cycle length; and the average temperatures of the different phases. The algorithm assigned green days in a conservative manner, and the number of red days per cycle would be generally greater than the duration of the fertile window. The algorithm would learn from previously recorded cycles and would be able to provide predictions of the user’s fertility status and upcoming ovulation, LH (luteinizing hormone) surge and menstruation days. The current and predicted fertility status of the user was displayed through a status bar, a calendar view and a temperature graph. Statistics of the characteristics of recorded cycles were also displayed and could be shared with physicians. Operating messages based on the user’s specific data were sent out to advise and motivate the user to take measurements as frequently as possible and to repeatedly inform the user to use protection on fertile days. The app used quantitative as opposed to qualitative measurements and its overall goal was to make the use of FAB methods easier to use and more effective.
They stated that based on their clinical evaluation, claims on contraceptive effectiveness were that the app had a method failure rate of 0.5, which was a measurement as to how often the app incorrectly displayed a green day when the user was actually fertile and got pregnant after having unprotected intercourse on that green day. This meant that 5 out of 1000 women who used the app for one year would become pregnant specifically due to a falsely attributed green day.
The app had a perfect-use failure rate of 1.0, which meant that 10 out of 1000 women who used the app for one year and became pregnant would have done so either because (i) they had unprotected intercourse on a green day that was falsely attributed as non-fertile (method failure) or (ii) the user had protected intercourse on a red day but the chosen method of contraception failed. The app had a typical-use failure rate of 6.8, which meant that in total 68 women out of 1000 got pregnant during one year of use due to all possible reasons, including wrongly attributed green days, having unprotected intercourse on red days and failure of contraceptive method used on red days.
They stated that in the clinical trials the app was shown to be highly accurate with a typical-use efficacy rate of 93% and a perfect-use efficacy rate of 99%. They clarified that the app adapted to each individual user and on average, they saw that users had 38% green days in the first cycle, 47% green days in the second cycle and 49% green days in the third cycle. It took roughly three cycles for the algorithm to stabilise the fraction of red and green days. If the cycle was regular, the woman could expect around 70% of green days after the learning phase. This allowed the user to know more about her menstrual cycles and could be used to prevent pregnancies.
2. They clarified that the claim “Clinically tested alternative to birth control methods” was a quote from the news site Business Insider which they considered to be correct as Natural Cycles was clinically tested and intended and certified for contraceptive use and could therefore be used as an alternative to other birth control methods. The claims were backed by scientific evidence, including clinical trials. The app was certified as a class II (b) medical device within the product category “stand alone software for the identification and prediction of the fertile window, for contraceptive or conceptive planning”. They stated that the contraceptive efficacy of the app was evaluated in clinical studies and accepted in peer-reviewed scientific journals. The contraceptive efficacy was comparable to other contraceptives or birth control methods. They referred to a prospective study from 2017 where the purpose was to investigate the contraceptive efficacy of the mobile app by evaluating the perfect- and typical-use Pearl Index. The study included 22,785 women and the typical-use Pearl Index was calculated to be 6.9% confirming the previous clinical study performed in 2016. They stated that the contraceptive efficacy during both typical and perfect use was comparable to other birth control methods.
Assessment
1. & 2. Upheld
The ASA considered that, in the context of the ad, the claim “highly accurate contraceptive app” would be understood by consumers to mean that the app had a high degree of accuracy and was therefore very reliable in being able to prevent unwanted pregnancies. We further considered that the claim “Clinically tested alternative to birth control methods”, presented alongside the “Highly accurate” claim would be understood to mean that the app was a reliable method of contraception which could be used in place of other established birth control methods, including those that were highly reliable in preventing unwanted pregnancies.
The ASA took expert advice on the evidence submitted by the advertiser in support of the claims. We reviewed three published papers based on the accumulated data obtained from the app. We considered that the most notable study was a prospective observational study, which included data input from over 22,000 users of Natural Cycles between August 2014 and August 2016 who had used the app to prevent pregnancies.
We understood that studies of contraceptive effectiveness distinguished between typical-use and perfect-use. Other methods such as barrier methods (e.g. condoms) or oral contraceptives would potentially show a substantial difference between perfect-use and typical-use, whereas long acting reversible contraceptive methods such as IUD’s, implants and injectables, which were highly reliable, and required little user input, would not show much difference between perfect-use and typical-use.
The perfect-use results from the study were in line with estimates for other contraceptive methods but, as with some other contraceptive methods, there were significant differences between the perfect-use results and those for typical-use. The Natural Cycles app required considerably more user input than most forms of contraception, with the need to take and input body temperature measurements several times a week, recording when intercourse had taken place, supplemented with LH measurements, abstention or alternative methods of contraception during the fertile period. The research and evidence showed that 91.7% of users did not become pregnant over the course of 13 cycles based on typical use of the app.
We noted from the studies that the reporting of intercourse was low with only 32% of cycles inputting such data, and that only 9.6% of cycles were considered as perfect-use, where the app had been used precisely as instructed. Given the very low level of perfect-use by users of the app and the significant difference between the effectiveness of the app when in perfect- and in typical-use, we considered that it would be misleading to base an accuracy claim on the perfect-use results and that the relevant data was the level of effectiveness seen in typical-use. We understood that the typical-use effectiveness of the app was comparable to that of some other types of contraception, such as barrier methods, but was significantly lower than the most reliable methods, such as long acting reversible contraceptive methods. Whilst we considered the evidence demonstrated the app could be effective as a method of birth control, we considered that it was misleading to describe it as “highly accurate”.
We considered that in isolation, the claim “clinically tested alternative to birth control methods” was unlikely to mislead. However, when presented alongside the accompanying claim “Highly accurate contraceptive app”, it further contributed to the impression that the app was a precise and reliable method of preventing pregnancies which could be used in place of other established birth control methods, including those which were highly reliable in preventing unwanted pregnancies. Because the evidence did not demonstrate that in typical-use it was “highly accurate” and because it was significantly less effective than the most reliable birth control methods, we considered that in the context of the ad the claim was likely to mislead.
We concluded that the claims “Highly accurate contraceptive app” and “Clinically tested alternative to birth control methods” were misleading.
The ad breached CAP Code (Edition 12) rules
3.1
3.1
Marketing communications must not materially mislead or be likely to do so.
(Misleading advertising),
3.7
3.7
Before distributing or submitting a marketing communication for publication, marketers must hold documentary evidence to prove claims that consumers are likely to regard as objective and that are capable of objective substantiation. The ASA may regard claims as misleading in the absence of adequate substantiation.
(Substantiation)
3.1
3.1
Marketing communications must not materially mislead or be likely to do so.
(Exaggeration) and
12.1
12.1
Objective claims must be backed by evidence, if relevant consisting of trials conducted on people. Substantiation will be assessed on the basis of the available scientific knowledge.
Medicinal or medical claims and indications may be made for a medicinal product that is licensed by the MHRA, VMD or under the auspices of the EMA, or for a CE-marked medical device. A medicinal claim is a claim that a product or its constituent(s) can be used with a view to making a medical diagnosis or can treat or prevent disease, including an injury, ailment or adverse condition, whether of body or mind, in human beings.
Secondary medicinal claims made for cosmetic products as defined in the appropriate European legislation must be backed by evidence. These are limited to any preventative action of the product and may not include claims to treat disease.
(Medicines, medical devices, health-related products and beauty products).
Action
The ad must not appear again in the form complained about. We told Natural Cycles Nordic AB Sweden not to state or imply that the app was a highly accurate method of contraception and to take care not to exaggerate the efficacy of the app in preventing pregnancies.
