Key takeaways
- The Natural Cycles algorithm is more accurate at determining the fertile window than the Rhythm Method and Standard Days Method
- Natural Cycles gives on average more non-fertile days than the other methods
- Natural Cycles is less likely to give a wrong non-fertile day in the fertile window and on the most fertile days
There are several fertility awareness-based methods (FABMs) out there. All FABMs rely on determining the fertile window, however, most FABMs are based on a simplified understanding of the menstrual cycle.
In this study, we set out to investigate how the Natural Cycles algorithm compares to two other FABMs — namely the Rhythm Method and the Standard Days Method — in how accurately they identify the fertile window.
The Rhythm Method (otherwise known as the calendar method) is based on the assumption that ovulation happens around 14 days before the end of the cycle. This method requires the user to track their period for six cycles before it can be used to prevent pregnancy, and the fertile window is then calculated based on the shortest and longest cycle during this time.
The Standard Days Method requires the user to have a regular cycle that is between 26 and 32 days, and cycle day 9 through 18 are always considered fertile.
The Natural Cycles algorithm identifies the fertile window based on temperature, optional ovulation tests and period data, which means that it’s individualized for each user. The start of the fertile window is based on past cycles, taking into account factors such as average ovulation day, cycle length and length of the cycle phases, and how these vary. The end of the fertile window is determined by confirming ovulation based on temperature data. Natural Cycles is effective from day one and does not require the user to monitor their cycle before it can be used, as the algorithm is designed to be conservative until it’s gathered enough data.
For this study, we analyzed data from Natural Cycles users, and we also calculated the fertile window for each user as if they were using the Rhythm Method and the Standard Days Method. Natural Cycles and the Rhythm method were compared using data from 26,626 cycles. For the comparison of Natural Cycles and the Standard Days Method, 16,386 cycles were analyzed. We also wanted to know how the Natural Cycles algorithm performed using basal body temperature (BBT) alone, and using BBT plus ovulation tests.
To compare the accuracy of the fertile window for each method, we looked at how many green (not fertile) days users got in total over 12 cycles. We also calculated how many wrong ‘green days’ were given by each method over 12 cycles, meaning days within the fertile window that were incorrectly marked as not fertile.
Lastly, we calculated how likely each method was to give a wrong green day on each day in the fertile window. This is important because the likelihood of conception varies on the different days in the fertile window — the most fertile day is the day before ovulation, followed by ovulation day, and the day with the lowest chance of conception is the first day of the window.
The results showed that Natural Cycles gave more green days over 12 cycles while also giving fewer wrong green days, compared to both of the other calendar-based methods.
More specifically, the Rhythm method gives no green days during the first six cycles of monitoring. In cycle 7, the amount of green days was 49% and the amount of wrong green days was 0.26%. Both of these fractions decreased until cycle 12, where the amount of green days was 43% and the amount of wrong green days was 0.08%.
The Standard Days Method, gave on average 58% green days and this stayed consistent during all cycles, but the fraction of wrong green days was higher — 1.60% in cycle 1 which decreased to 0.27% in cycle 12, likely because the number of users who were still eligible for this method decreased over time due to the strict restrictions of this method.
In contrast, Natural Cycles gave an average of 44% green days (when BBT only was used) and 49% (when BBT and LH were used) in the first cycle, and this increased to 57% and 61% respectively in cycle 12. The amount of wrong green days was stable over time and was on average 0.12% (BBT only) and 0.07% (LH and BBT).
When looking at the likelihood of wrong green days being given in the fertile window, we found that with the Rhythm Method, the likelihood of a wrong green day was largest on ovulation day (2.46%), followed by the day before ovulation (0.80%). The Standard Days Method also had the biggest likelihood of giving a wrong green day on ovulation day (13.01%) and the day before ovulation (6.90%).
For Natural Cycles, the likelihood of a wrong green day in the fertile window was biggest on the first day of the fertile window (2.03%), and Natural Cycles was less likely than the other two methods to give a wrong green day on the most fertile days. When BBT only was used, the likelihood of a wrong green day on the day before ovulation was 0.31% and on ovulation it was 0.66%, and when BBT and LH was used, the likelihood was 0% for both days.
In summary, this study shows that the Natural Cycles individualized algorithm adapts to each user’s cycle, detecting the fertile window with more accuracy than both the Rhythm Method and the Standard Days Method. Natural Cycles is also less likely to give a wrong green day during the fertile window in general and during the most fertile days in particular.
