Evaluating Pregnancy Rates in Fertility Awareness-Based Methods for Family Planning: Simulated Comparison of Correct Use to Avoid,Method-Related,and Total Pregnancy Rates

Abstract

Fertility awareness-based methods (FABMs), also known as natural family planning (NFP), enable couples to identify the days of the menstrual cycle when intercourse may result in pregnancy (“fertile days”), and to avoid intercourse on fertile days if they wish to avoid pregnancy. Thus, these methods are fully dependent on user behavior for effectiveness to avoid pregnancy. For couples and clinicians considering the use of an FABM, one important metric to consider is the highest expected effectiveness (lowest possible pregnancy rate) during the correct use of the method to avoid pregnancy. To assess this, most studies of FABMs have reported a method-related pregnancy rate (a cumulative proportion), which is calculated based on all cycles (or months) in the study. In contrast, the correct use to avoid pregnancy rate (also a cumulative proportion) has the denominator of cycles with the correct use of the FABM to avoid pregnancy. The relationship between these measures has not been evaluated quantitatively. We conducted a series of simulations demonstrating that the method-related pregnancy rate is artificially decreased in direct proportion to the proportion of cycles with intermediate use (any use other than correct use to avoid or targeted use to conceive), which also increases the total pregnancy rate. Thus, as the total pregnancy rate rises (related to intermediate use), the method-related pregnancy rate falls artificially while the correct use pregnancy rate remains constant. For practical application, we propose the core elements needed to assess correct use cycles in FABM studies.

Summary

Fertility awareness-based methods (FABMs) can be used by couples to avoid pregnancy, by avoiding intercourse on fertile days. Users want to know what the highest effectiveness (lowest pregnancy rate) would be if they use an FABM correctly and consistently to avoid pregnancy. In this simulation study, we compare two different measures: (1) the method-related pregnancy rate; and (2) the correct use pregnancy rate. We show that the method-related pregnancy rate is biased too low if some users in the study are not using the method consistently to avoid pregnancy, while the correct use pregnancy rate obtains an accurate estimate.

Short Summary

In FABM studies, the method-related pregnancy rate is biased too low, but the correct use pregnancy rate is unbiased.

Keywords

Correct use fertility awareness fertility awareness-based methods method effectiveness natural family planning perfect use pregnancy rates simulation study

Effectiveness, or the ability of a family planning method to prevent pregnancy, is one of the key decisional parameters for persons or couples choosing a method and the main factor in clinical counseling conversations (Gavin et al. 2017). The majority of family planning studies are designed to assess two parameters: (a) proportion of couples becoming pregnant during the normal use of a method, called the typical use pregnancy rate and (b) the proportion of couples becoming pregnant when the method is used as instructed to avoid pregnancy, generally called the perfect use or correct use pregnancy rate (Trussell 2011). In this article, we prefer and use the term correct use, because the term “perfect” might imply an unachievable standard. However, an older measure sometimes is still sometimes used, which has often been called the method-related pregnancy rate, or method effectiveness. In the medical literature, the cumulative proportion of couples with pregnancy is usually referred to as the pregnancy rate. Usually pregnancy rates (i.e., cumulative proportions) are calculated for twelve months or thirteen cycles of use. The mathematical reciprocal of the pregnancy rate is usually referred to as effectiveness; for example, if the pregnancy rate is 10 percent, the effectiveness is 90 percent (Hilgers 1984).

FABMs of family planning, also referred to as natural family planning (NFP), are a class of family planning methods wherein a woman observes symptoms or signs that reflect cyclic hormonal changes in the body, and interprets these with specific algorithms or rules to identify times in the reproductive (menstrual) cycle when pregnancy can occur, commonly referred to as the fertile window (Manhart et al. 2013; Simmons and Jennings 2020). Using this information, couples may time sexual intercourse according to their desire to achieve or avoid a pregnancy (Stanford and Porucznik 2017). Separate from family planning, couples may also use FABMs for medical evaluation and management (Duane et al. 2022). Sometimes FABMs are used in combination with other methods to avoid pregnancy, which can be considered dual method use, or a fertility awareness combined approach. Some studies consider using specific additional methods with FABMs consistent with correct use (Jennings et al. 2019) while others do not (Arevalo et al. 2004). In the models we use throughout this article, we evaluate the use of FABMs alone without the use of additional methods such as condoms or withdrawal. A significant proportion of people of reproductive age may be interested in FABMs (Leonard et al. 2006; Jackson et al. 2016; Bull et al. 2019), and there is a need for accurate information about pregnancy rates during use.

In a recent systematic review of FABMs used to avoid pregnancy, among fifty-three studies, only eight reported an appropriately calculated correct use pregnancy rate. Most of the remaining studies reported a method-related pregnancy rate, and it was sometimes erroneously called a perfect use pregnancy rate (Peragallo Urrutia et al. 2018). Trussell and colleagues previously showed that method-related pregnancy rates were biased downward in relation to correct use (“perfect use”) pregnancy rates (Trussell 1991); however, the difference between the method-related pregnancy rate and the correct use pregnancy rate is still often not appreciated, and the factors that may artificially decrease the method-related pregnancy rate have not been evaluated quantitatively.

We conducted a series of lifetable simulations to quantitatively assess the relationship between the method-related pregnancy rate, the correct use (“perfect use”) pregnancy rate to avoid pregnancy, and the total pregnancy rate. Our simulations also demonstrate the relationship between the correct use pregnancy rate to the typical use pregnancy rate. For practical application, we also discuss the principal elements required to assess correct use cycles, to calculate the correct use pregnancy rate.

Methods

Because FABMs can be used to avoid pregnancy or to try to conceive, it's necessary to distinguish different types of use (Lamprecht and Trussell 1997; Stanford and Porucznik 2017). For the purposes of this simulation study, and to capture the full range of possible uses of FABMs in family planning, we define three types of use of an FABM: correct use to avoid, targeted use to conceive, and intermediate use (Table 1). Correct use to avoid requires consistent behaviors to identify the fertile window according to the instructions of the specific FABM being used and to avoid intercourse during the fertile window. This has been shown to be much more likely to occur when there is consistently strong intention to avoid pregnancy (Fehring et al. 2013). Targeted use to conceive, which may also be called correct use to conceive, involves identifying the fertile window and having sexual intercourse targeted to the “most fertile” days or “center” of the window (Lynch, Jackson, and Buck Louis 2006; Gnoth et al. 2003; Mu and Fehring 2014; Bouchard, Fehring, and Schneider 2018). Intermediate use can involve multiple different behaviors across a spectrum between correct use to avoid and targeted use to conceive, which may correlate to a wide spectrum of motivations and intentions, including different levels of openness to pregnancy (Bachrach and Newcomer 1999). This is important, as pregnancy intentions may change over time depending on one's life circumstances and ambivalence around pregnancy is associated with higher unintended pregnancy rates (Geist et al. 2021; Higgins, Popkin, and Santelli 2012). Of note, the term “intermediate use” is not synonymous with the term “imperfect use,” because the concept of “imperfect use” requires a stated intention to avoid pregnancy. For the purposes of our simulation, we use the term “intermediate use” to capture the full spectrum of possible use of an FABM, rather than requiring an explicit intention to avoid pregnancy. In the context of FABMs, intermediate use may include having intercourse on the “edges” of the fertile window (i.e., when the probability of pregnancy is slightly increased but not maximal), either knowingly or unknowingly.

Table 1.

Example intentions, behaviors, and plausible probability of pregnancy for each cycle during the use of a Fertility Awareness-Based Method (FABM) for family planning, or natural family planning (NFP).

Type of use	Intentions	Circumstances	Identification of fertile window	Timing of sexual intercourse	Probability of pregnancy per cycle
Correct use to avoid	Strong intention to avoid	Supportive of use consistent with intention	Careful, consistent	None in identified fertile window	Very low 0.001–0.002
Intermediate use*	Possible ambivalence	Possible challenges or barriers to use consistent with intention	May be imprecise and/or inconsistent	Occasional in fertile window, especially at “edges”	Moderate 0.03–0.07
Targeted use to conceive	Intention to conceive	Supportive of use consistent with intention	Consistent	Targeted to most fertile days	High 0.25

*As defined in this article, “intermediate use” includes a full spectrum of any type of use other than correct use to avoid or targeted use to conceive. This is a broader concept than “imperfect use,” which presumes that the intention is to avoid pregnancy.

We established plausible parameters for each type of use, based on day-specific probabilities of conception reported in multiple prior studies. For example, Colombo and Masarotto (2000) calculated the probability of clinically recognized pregnancy of 0.26 for intercourse on the most fertile day of the cycle (two days prior to estimated ovulation), which corresponds reasonably well to maximum per cycle probabilities of pregnancy in many other studies (also occurring one to two days prior to estimated ovulation) (Lynch, Jackson, and Buck Louis 2006; Faust et al. 2019). With multiple acts of intercourse in the fertile window, the probability of pregnancy rapidly approaches that of intercourse on the most fertile day (Stanford and Dunson 2007). The maximum probability of pregnancy per cycle has been reported in different populations without known subfertility ranging from about 0.20 to 0.40 (Dunson, Colombo, and Baird 2002; Faust et al. 2019). We set per cycle probability of pregnancy during targeted use to conceive at 0.25 (Table 1).

With regard to intermediate use, Colombo and Masarotto (2000) calculated day-specific probabilities of conception of 0.07 five days prior to estimated ovulation, and 0.03 six days prior to ovulation, and 0.01 seven days prior to ovulation. Very similar probabilities from five to seven days prior to ovulation have been found in other studies (Lynch, Jackson, and Buck Louis 2006; Dunson, Colombo, and Baird 2002; Lum et al. 2016; Faust et al. 2019). Day-specific probabilities of conception can vary by age, parity, and other factors of the population being studied (Dunson, Colombo, and Baird 2002; Sundaram, McLain, and Buck Louis 2012; Mikolajczyk and Stanford 2006). In our simulations, we varied the probabilities of pregnancy per cycle for intermediate use from 0.03 to 0.07 (Table 1).

For correct use to avoid pregnancy, we consider that several FABM studies have found correct use pregnancy rates (cumulative proportions) at thirteen cycles ranging from 0.01 to 0.03 (Peragallo Urrutia et al. 2018). We chose per cycle pregnancy rates of 0.001 to 0.002, which results in correct use pregnancy rates at 13 cycles in this range (Table 1).

The definition of the relevant numerators and denominators for different types of pregnancy rates (or cumulative incidence of pregnancy) for FABMs are summarized in Table 2, including the numerators and denominators for each. For the purposes of this article, we focus on the method-related pregnancy rate and the correct (or “perfect”) use of pregnancy rate. Both metrics have the same numerator, pregnancies that occur during correct use. For the denominator, the method-related pregnancy rate includes in the denominator all cycles (or months) in the study. In contrast, the denominator for the correct use pregnancy rate is all cycles with correct use to avoid pregnancy. These criteria for the numerator and denominator are independent of the statistical approaches that may be used for analysis. While it is considered best practice to use a life table for the cumulative probability of pregnancy by 13 months, another commonly used approach is the Pearl Index, which is calculated by the formula: (number of pregnancies × 1300)/(number of cycles); sometimes 1,200 is used instead of 1,300 (Trussell and Portman 2013). In our simulations, we use a life table.

Table 2.

Summary of numerators and denominators for different types of cumulative incidence of pregnancy (“pregnancy rates”) in studies of fertility awareness-based methods (FABMs) or natural family planning (NFP).

Measure	Numerator	Denominator
Method-related pregnancy rate*	Pregnancies during correct use to avoid	All cycles (or months)
Correct (“perfect”) use pregnancy rate (to avoid pregnancy)	Pregnancies during correct use to avoid	Cycles with correct use to avoid
Total pregnancy rate	All pregnancies, regardless of intention (avoid or conceive)	All cycles (or months)
Typical use pregnancy rate (to avoid pregnancy)	All pregnancies occurring with intention to avoid	All cycles (or months) with intention to avoid

*The definition of “method-related pregnancy rate” in this table is what has been used in studies of FABMs or NFP; often reported as “method effectiveness,” that is, 100 percent minus the method-related pregnancy rate.

Note: Any of these measures can be calculated with Pearl Rates, or with life table or similar statistical methods (e.g., Kaplan–Meier). A standard time period is up to thirteen consecutive cycles or one year, although any time period can be used.

For the major parameter of interest for this simulation, we systematically varied the proportion of cycles that were characterized by intermediate use (from 5 percent to 95 percent). This range of intermediate use is plausible across different populations. A few studies of FABMs have systematically reported the proportion of cycles with correct versus incorrect use among those seeking to avoid pregnancy: the proportion of cycles with “imperfect use” in other reports has been reported as low as 13 percent (Trussell and Grummer-Strawn 1991), to as high as 76 percent (Jennings et al. 2019). Because our definition of intermediate use does not require a stated intention to avoid, the proportion of intermediate use cycles could be higher, depending on the population.

We also conducted simulations in which we systematically varied the proportion of cycles characterized by targeted use to conceive (from 5 percent to 20 percent). The proportion of cycles with targeted use to conceive has not been reported in FABM studies; rather use to conceive has usually been excluded from studies of FABMs to avoid, and then studied in isolation as a separate issue (Gnoth et al. 2003; Mu and Fehring 2014; Bouchard, Fehring, and Schneider 2018). In FABM studies that assess pregnancy intentions at the beginning of each cycle, these cycles are excluded from typical and correct use pregnancy rates (Lamprecht and Trussell 1997; Frank-Herrmann et al. 2007). Further, only a few FABM studies have published total pregnancy rates (that include intended pregnancies which would presumably result from targeted use to conceive, included with other types of pregnancies) (Howard and Stanford 1999; Fehring et al. 2013).

We made three simplifying assumptions for modeling purposes. First, we assume that all cycles include sexual intercourse. Second, we set a per cycle probability of attrition from observation of 0.01. Third, we assume that mean fecundability among couples remains constant across the simulation. Fourth, we used a steady-state assumption, namely that the proportion of cycles in each category of use remained the same at the beginning of each cycle of the simulation. We also explored simulations with alternate assumptions with similar results, described in the discussion.

We expressed all results of life table simulations as pregnancy rates (cumulative proportion pregnant) after 13 consecutive cycles.

Results

In Figure 1, we illustrate the relationship between correct use pregnancy rates, method pregnancy rates, and total pregnancy rates (cumulative proportion for thirteen cycles), under the following parameters: probability of pregnancy per cycle with correct use to avoid 0.001; probability of pregnancy per cycle with intermediate use, 0.05; no cycles with targeted use to conceive; attrition per cycle 0.01. As the proportion of cycles with intermediate use is varied from 5 percent to 95 percent, the method-related pregnancy rate drops from 1.2 percent for 5 percent of the cycles with intermediate use to 0.1 percent for 95 percent of the cycles with intermediate use. However, the correct use pregnancy rate remains constant, at 1.3 percent. Meanwhile, as the method-related pregnancy rate drops, the total pregnancy rate rises.

Figure 1.

Simulated lifetable cumulative proportions of women pregnant (“pregnancy rates”) after 13 cycles of use of a Fertility Awareness-Based methods under the following fixed conditions: probability of pregnancy per cycle with correct use to avoid 0.001; probability of pregnancy per cycle with intermediate use, 0.05; no cycles with targeted use to conceive; attrition per cycle 0.01. The proportion of cycles with intermediate use is varied from 5% to 95%.

A similar pattern is shown in Figure 2, with the probability of pregnancy per cycle with correct use to avoid 0.002, and otherwise the same parameters as Figure 1. As the proportion of cycles with intermediate use is varied from 5 percent to 95 percent, the method-related pregnancy rate drops from 2.4 percent for 5 percent of the cycles with intermediate use to 0.1 percent for 95 percent of the cycles with intermediate use. However, the correct use pregnancy rate remains constant, at 2.6 percent. As before, while the method-related pregnancy rate drops, the total pregnancy rate rises.

Figure 2.

Simulated lifetable cumulative proportions of women pregnant (“pregnancy rates”) after 13 cycles of use of a Fertility Awareness-Based methods under the following fixed conditions: probability of pregnancy per cycle with correct use to avoid 0.002; probability of pregnancy per cycle with intermediate use, 0.05; no cycles with targeted use to conceive; attrition per cycle 0.01. The proportion of cycles with intermediate use is varied from 5% to 95%.

Finally, increasing the proportion of cycles with targeted use to conceive also lowers the method-related pregnancy rate, as illustrated in Figure 3. In this simulation, the probability of pregnancy per cycle with correct use to avoid is 0.001; the probability of pregnancy per cycle with intermediate use, 0.05; the probability of pregnancy per cycle with targeted use to conceive 0.25; the proportion of cycles with intermediate use is fixed at 0.30; attrition per cycle 0.01. As the proportion of cycles with targeted use to conceive is varied from 0 percent to 20 percent, the method-related pregnancy drops from 0.9 percent to 0.6 percent, while the total pregnancy rate rises.

Figure 3.

Simulated lifetable cumulative proportions of women pregnant (“pregnancy rates”) after 13 cycles of use of a Fertility Awareness-Based methods under the following fixed conditions: probability of pregnancy per cycle with correct use to avoid 0.001; probability of pregnancy per cycle with intermediate use, 0.05; probability of pregnancy per cycle with targeted use to conceive 0.25; proportion of cycles with intermediate use 0.30; attrition per cycle 0.01. The proportion of cycles with targeted use to conceive is varied from 0% to 30%.

Discussion

Using life table simulations, we have demonstrated quantitatively how the method-related pregnancy rate (cumulative proportion pregnant at 13 cycles) is reduced inversely to the proportion to cycles with any kind of intermediate use (or with targeted use to conceive). However, the correct use pregnancy rate remains unaffected. The apparent drop in the method-related pregnancy rate can be substantial, e.g., a relative drop to 50 percent or less of the correct use pregnancy rate. These results reinforce that the correct use pregnancy rate (also called “perfect use” pregnancy rate) provides the most accurate information about the “best possible” or “maximum possible” effectiveness (minimum pregnancy rate) of an FABM, information which is important for women, couples, clinicians, policymakers. In contrast, the method-related pregnancy rate, or method effectiveness, as it has been widely used, is misleading.

We conducted these simulations in relation to the total pregnancy rate, rather than the typical use pregnancy rate. We chose this approach in order to include the full range of possible uses of an FABM in relation to family planning. The typical use pregnancy rate describes the pregnancy rate experienced overall by a group of users intending to avoid pregnancy, while the total pregnancy rate describes the pregnancy rate experienced by all users, including those intending to avoid, those intending to conceive and those who may be ambivalent or “trying to whatever” (Fisher 2016). However, with slightly different assumptions, we can apply the exact same modeling to a typical use pregnancy rate. For example, if we assume that everyone in the simulations represented in Figures 1 and 2 was consistently intending to avoid pregnancy, then the “proportion of cycles with intermediate use” in the simulation becomes by definition exactly the same as the “proportion of cycles with imperfect use.” In this alternative scenario, the “total pregnancy rate” is by definition the same as the “typical use pregnancy rate,” because there are no cycles where there is any ambivalence about intention to avoid. However, the simulated result is the same.

Similarly, as shown in Figure 3, the more cycles there are in the denominator with targeted use to conceive, the lower the apparent method effectiveness rate. In the most general terms, the more cycles there are in the denominator with any type of use other than correct use to avoid, the more the method-related pregnancy rate is artificially lowered relative to the correct use pregnancy rate. Thus, any studies that include a substantial proportion of cycles with imperfect use to avoid, and/or other kinds of intermediate use cycles, and/or cycles with targeted use to conceive, will also have artificially lower method-related pregnancy rates.

While we have selected plausible pregnancy rates for each of the types of cycles that we defined (correct use to avoid, intermediate use, and targeted use to conceive), other pregnancy rates could be entered into the simulation. As suggested by the comparison of the simulations shown in Figure 1 and 2, across many other possible pregnancy rates, the absolute correct pregnancy rate and the total pregnancy rate will change, but the pattern remains the same for the method-related pregnancy rate: as more cycles are added to the denominator that are not correct use cycles to avoid pregnancy, the method-related pregnancy rate is artificially decreased.

The fundamental principle to assessing the correct use pregnancy rate is that correct use needs to be assessed for each cycle of use. Preferably, intention of use (to avoid pregnancy, to conceive, or undecided) is also assessed prospectively for each cycle. What constitutes correct use to avoid is fundamentally whether the fertile window has been adequately identified (according to the method), and whether any genital intercourse or contact has occurred within the fertile window. The signs and symptoms needed to identify the fertile window may vary for specific FABMs. With this in mind, we propose general recommendations for identifying the correct use in each cycle in Table 3. The specific application of these recommendations will vary depending on the specific FABM used, as well as the judgment of the investigators.

Table 3.

Key study design considerations to assess correct use cycles to avoid pregnancy in fertility awareness-based method (FABM) studies

Method type	Factors to assess in study design and conduct
All FABM methods	Are pregnancy intentions assessed at the beginning of each cycle? What type of training/education do the participants receive on the FABM under study? Is there an assessment of the participants’ correct identification of the fertile window according to the FABM under study? Is the possible concomitant use of other methods of family planning explicitly assessed, including condoms, spermicide, or withdrawal? Does the definition of correct use for the study include or exclude the concomitant use of other methods of family planning, that is, FABM combined use with another method, and which ones? Do participants record all acts of intercourse or genital contact? How is this incentivized and evaluated? What is the frequency and timing of participant recording of their observation of fertility signs and sexual contact (e.g., daily records, weekly records, or other)? Is there potential for social desirability bias in the information recorded and reported by participants in the study?
Issues with method-specific application	Is there an assessment of concomitant use of methods of identifying the fertile window outside of those of the FABM under study? (e.g., do some couples in a study of a method based on cervical fluid also use basal body temperature or test urine LH outside of study protocol?)

Method type

Factors to assess in study design and conduct

All FABM methods

Are pregnancy intentions assessed at the beginning of each cycle?

What type of training/education do the participants receive on the FABM under study?

Is there an assessment of the participants’ correct identification of the fertile window according to the FABM under study?

Is the possible concomitant use of other methods of family planning explicitly assessed, including condoms, spermicide, or withdrawal?

Does the definition of correct use for the study include or exclude the concomitant use of other methods of family planning, that is, FABM combined use with another method, and which ones?

Do participants record all acts of intercourse or genital contact? How is this incentivized and evaluated?

What is the frequency and timing of participant recording of their observation of fertility signs and sexual contact (e.g., daily records, weekly records, or other)?

Is there potential for social desirability bias in the information recorded and reported by participants in the study?

Issues with method-specific application

Is there an assessment of concomitant use of methods of identifying the fertile window outside of those of the FABM under study? (e.g., do some couples in a study of a method based on cervical fluid also use basal body temperature or test urine LH outside of study protocol?)

While FABMs differ regarding the signs that individuals may need to learn to track and the rules for use, they all require some level of training and behavioral cooperation to use correctly (Sinai and Arevalo 2006; Simmons and Jennings 2020). Thus, the typical use effectiveness of FABMs is dependent on several ongoing user behaviors, which in turn are influenced by intention and motivation. For most FABMs that have calculated a correct use pregnancy rate, it is less than five per 100 woman-years; for several methods, it is less than two pregnancies per 100 woman-years. Typical use pregnancy rates from the same studies range from two to twenty-two pregnancies per 100 woman-years; for all but one FABM, the absolute difference between the typical use and correct use pregnancy rates is more than 6 per 100 woman-years (Peragallo Urrutia et al. 2018; Manhart et al. 2013). Therefore, when a method-related pregnancy rate is reported instead of a correct-use pregnancy rate, this yields an artificially low estimate of the achievable pregnancy rate with correct use (i.e., an artificially high estimate of maximum possible effectiveness to avoid pregnancy). The typical use pregnancy rate, as defined in Table 2, is also an important metric for users and clinicians, which provides critical information about the actual use of a method with the intention to avoid pregnancy, and facilitates comparisons between different methods of family planning. However, it is not the focus of this article.

The simulations we reported in this article used a steady-state assumption, namely that the pregnancy probabilities per cycle for each category of use (correct use, intermediate use, and targeted use to conceive), and the proportion of cycles in each category of use, were the same at the beginning of each cycle. This assumption implicitly assumes ongoing transitions of some couples from correct use to intermediate use at the beginning of each cycle; such transitions are plausible for a group of FABM users. However, we also conducted additional simulations based on the assumption that couples never transition from their category of use at the beginning of the simulation, except to pregnancy; that is, couples starting with correct use to avoid always having correct use until pregnancy, and likewise for couples with intermediate use or targeted use to conceive, respectively. The results were very similar for the correct use pregnancy rate, and had the same inverse relationship between the proportion of cycles with intermediate use and the method-related pregnancy rate; however, under these assumptions, the total and typical use pregnancy rates were lower (data not shown).

In conclusion, we have quantitatively demonstrated that as the total pregnancy rate rises (related to any kind of intermediate use), the method-related pregnancy rate falls artificially, while the correct use pregnancy rate remains constant. Since the correct use (perfect use) pregnancy rate remains unaffected by the proportion of intermediate use cycles (or cycles with targeted use to conceive), it is therefore the appropriate measure for estimating the best possible effectiveness (lowest possible pregnancy rate) of an FABM or NFP method. We have also proposed the key elements to determine the correct use in each cycle.

Footnotes

Acknowledgments

We thank Madeline Mullholand for her assistance with the figures.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Joseph B. Stanford

Biographical Notes

Joseph B. Stanford, MD, MSPH, is Professor, Vice Chair for Research, and Director of the Office of Cooperative Reproductive Health, Division of Public Health, Department of Family and Preventive Medicine, University of Utah School of Medicine. His interests include reproductive epidemiology, restorative reproductive medicine, fertility awareness and natural family planning, women’s health, and the periconceptional and prenatal origins of health and disease. He has served on committees for the Eunice Kennedy Shriver National Institute of Child Health and Human Development, and for the Food and Drug Administration. He is past president of the American Academy of FertilityCare Professionals and serves as a board member for FertilityCare Centers of America and for the International Institute for Restorative Reproductive Medicine.

Marguerite Duane, MD, MHA, MSPH, a board-certified family physician, is co-founder and Executive Director of FACTS about Fertility, fertility awareness-based methods (FABMs) education. She also serves as an Adjunct Associate Professor at Georgetown University, Duquesne University, and the University of Utah. Dr. Duane currently serves as the CMA State Director for DC, and is the past president of the St. Giuseppe Moscatti DC CMA Guild. She has served on the board of the American Academy of Family Physicians (AAFP) and the Family Medicine Education Consortium (FMEC).

Rebecca Simmons, PhD, is Assistant Research Professor; Department of Obstetrics and Gynecology, University of Utah School of Medicine. She previously was Assistant Professor and Senior Research Officer at the Institute for Reproductive Health, Georgetown University. She has experience leading large research and implementation studies in the United States and internationally. Her work is specifically directed toward improving family planning access and method choice.

References

Arevalo

Jennings

Nikula

Sinai

. 2004. “Efficacy of the New TwoDay Method of Family Planning.” Fertility and Sterility 82 (4): 885–92.

Bachrach

C. A.

Newcomer

. 1999. “Intended Pregnancies and Unintended Pregnancies: Distinct Categories or Opposite Ends of a Continuum?” Family Planning Perspectives 31 (5): 251–2.

Bouchard

T. P.

Fehring

R. J.

Schneider

M. M.

. 2018. “Achieving Pregnancy Using Primary Care Interventions to Identify the Fertile Window.” Frontiers in Medicine 4: 250. https://doi.org/10.3389/fmed.2017.00250.

Bull

Rowland

Lundberg

Berglund-Scherwitzl

Gemzell-Danielsson

Trussell

Scherwitzl

. 2019. “Typical Use Effectiveness of Natural Cycles: Postmarket Surveillance Study Investigating the Impact of Previous Contraceptive Choice on the Risk of Unintended Pregnancy.” BMJ Open 9 (3): e026474. https://doi.org/10.1136/bmjopen-2018-026474.

Colombo

Masarotto

. 2000. “Daily Fecundability: First Results From a New Data Base.” Demographic Research 3 (5).

Duane, Marguerite, Joseph B. Stanford, Christina A. Porucznik, and Pilar Vigil. 2022. “Fertility Awareness-Based Methods for Women's Health and Family Planning.” Frontiers in Medicine 9. https://doi.org/10.3389/fmed.2022.858977.

Dunson

D. B.

Colombo

Baird

D. D.

. 2002. “Changes with Age in the Level and Duration of Fertility in the Menstrual Cycle.” Human Reproduction 17 (5): 1399–403.

Faust

Bradley

Landau

Noddin

Farland

L. V.

Baron

Wolfberg

. 2019. “Findings from a Mobile Application-based Cohort Are Consistent with Established Knowledge of the Menstrual Cycle, Fertile Window, and Conception.” Fertility and Sterility 112 (3): 450–57.e3. https://doi.org/10.1016/j.fertnstert.2019.05.008.

Fehring

R. J.

Schneider

Barron

M. L.

Pruszynski

. 2013. “Influence of Motivation on the Efficacy of Natural Family Planning.” MCN: The American Journal of Maternal/Child Nursing 38 (6): 352–8. https://doi.org/10.1097/NMC.0b013e3182a1ecc0.

10.

Fehring

R. J.

Schneider

Raviele

Rodriguez

Pruszynski

. 2013. “Randomized Comparison of Two Internet-supported Fertility-awareness-based Methods of Family Planning.” Contraception 88 (1): 24–30. https://doi.org/10.1016/j.contraception.2012.10.010.

11.

Fisher

Simcha.

2016. “If NFP works, why does she have ten kids?” Sicha Fisher, July 28, 2016. https://www.simchafisher.com/2016/07/28/if-nfp-works-why-does-she-have-ten-kids/.

12.

Frank-Herrmann

Heil

Gnoth

Toledo

Baur

Pyper

Jenetzky

Strowitzki

Freundl

. 2007. “The Effectiveness of a Fertility Awareness Based Method to Avoid Pregnancy in Relation to a Couple's Sexual Behaviour During the Fertile Time: A Prospective Longitudinal Study.” Human Reproduction 22 (5): 1310–19.

13.

Gavin

L. E.

Ahrens

K. A.

Dehlendorf

Frederiksen

B. N.

Decker

Moskosky

. 2017. “Future Directions in Performance Measures for Contraceptive Care: A Proposed Framework.” Contraception 96 (3): 138–44. https://doi.org/10.1016/j.contraception.2017.06.001.

14.

Geist

Everett

B. G.

Simmons

R. G.

Sanders

J. N.

Gawron

L. M.

Myers

Turok

D. K.

. 2021. “Changing Lives, Dynamic Plans: Prospective Assessment of 12-Month Changes in Pregnancy Timing Intentions and Personal Circumstances using Data from HER Salt Lake.” PLoS One 16 (9): e0257411. https://doi.org/10.1371/journal.pone.0257411.

15.

Gnoth

Godehardt

Frank-Herrmann

Freundl

. 2003. “Time to Pregnancy: Results of the German Prospective Study and Impact on the Management of Infertility.” Human Reproduction 18 (9): 1959–66.

16.

Higgins

J. A.

Popkin

R. A.

Santelli

J. S.

. 2012. “Pregnancy Ambivalence and Contraceptive Use among Young Adults in the United States.” Perspectives on Sexual and Reproductive Health 44 (4): 236–43. https://doi.org/10.1363/4423612.

17.

Hilgers

T. W.

1984. “The Statistical Evaluation of Natural Methods of Family Planning.” Int Rev Nat Fam Plann 8 (3): 226–64.

18.

Howard

M. P.

Stanford

J. B.

. 1999. “Pregnancy Probabilities during Use of the Creighton Model Fertility Care System.” Arch Fam Med 8 (5): 391–402.

19.

Jackson

A. V.

Karasek

Dehlendorf

Foster

D. G.

. 2016. “Racial and Ethnic Differences in Women's Preferences for Features of Contraceptive Methods.” Contraception 93 (5): 406–11. https://doi.org/10.1016/j.contraception.2015.12.010.

20.

Jennings

Haile

L. T.

Author

R. G.

Spieler

Shattuck

. 2019. “Perfect- and Typical-Use Effectiveness of the Dot Fertility App over 13 Cycles: Results from a Prospective Contraceptive Effectiveness Trial.” The European Journal of Contraception & Reproductive Health Care 24 (2): 148–53. https://doi.org/10.1080/13625187.2019.1581164.

21.

Lamprecht

Trussell

. 1997. “Natural Family Planning Effectiveness: Evaluating Published Reports.” Advances in Contraception 13 (2-3): 155–165.

22.

Leonard

C. J.

Chavira

Coonrod

D. V.

Hart

K. W.

Bay

R. C.

. 2006. “Survey of Attitudes Regarding Natural Family Planning in an Urban Hispanic Population.” Contraception 74 (4): 313–7. https://doi.org/10.1016/j.contraception.2006.05.075.

23.

Lum

K. J.

Sundaram

Buck Louis

G. M.

Louis

T. A.

. 2016. “A Bayesian Joint Model of Menstrual Cycle Length and Fecundity.” Biometrics 72 (1): 193–203. https://doi.org/10.1111/biom.12379.

24.

Lynch

C. D.

Jackson

L. W.

Buck Louis

G. M.

. 2006. “Estimation of the Day-Specific Probabilities of Conception: Current State of the Knowledge and the Relevance for Epidemiological Research.” Paediatric and Perinatal Epidemiology 20 (Suppl 1): 3–12.

25.

Manhart

Michael D.

Author

Lind

April

Sinai

Irit

Golden-Tevald

Jean

. 2013. “Fertility Awareness-based Methods of Family Planning: A Review of Effectiveness for Avoiding Pregnancy Using SORT.” Osteopathic Family Physician 5 (1): 2–8. https://doi.org/10.1016/j.osfp.2012.09.002.

26.

Mikolajczyk

R. T.

Stanford

J. B.

. 2006. “Measuring Fecundity with Standardised Estimates of Expected Pregnancies.” Paediatric and Perinatal Epidemiology 20 (Suppl 1): 43–50.

27.

Fehring

R. J.

. 2014. “Efficacy of Achieving Pregnancy with Fertility-Focused Intercourse.” MCN: The American Journal of Maternal/Child Nursing 39 (1): 35–40. https://doi.org/10.1097/NMC.0b013e3182a76b88.

28.

Peragallo Urrutia

Polis

C. B.

Jensen

E. T.

Greene

M. E.

Kennedy

Author

J. B.

. 2018. “Effectiveness of Fertility Awareness–Based Methods for Pregnancy Prevention.” Obstetrics & Gynecology 132 (3): 591–604. https://doi.org/10.1097/AOG.0000000000002784.

29.

Simmons

R. G.

Jennings

. 2020. “Fertility Awareness-Based Methods of Family Planning.” Best Practice & Research Clinical Obstetrics & Gynaecology 66: 68–82. https://doi.org/10.1016/j.bpobgyn.2019.12.003.

30.

Sinai

Arevalo

. 2006. “It's All in the Timing: Coital Frequency and Fertility Awareness-based Methods of Family Planning.” Journal of Biosocial Science 38 (6): 763–77. https://doi.org/10.1017/S0021932005027227.

31.

Stanford

J. B.

Dunson

D. B.

. 2007. “Effects of Sexual Intercourse Patterns in Time to Pregnancy Studies.” Am J Epidemiol 165 (9): 1088–95. https://doi.org/10.1093/aje/kwk111.

32.

Stanford

J. B.

Porucznik

C. A.

. 2017. “Enrollment, Childbearing Motivations, and Intentions of Couples in the Creighton Model Effectiveness, Intentions, and Behaviors Assessment (CEIBA) Study.” Front Med (Lausanne) 4: 147. https://doi.org/10.3389/fmed.2017.00147.

33.

Sundaram

McLain

A. C.

Buck Louis

G. M.

. 2012. “A Survival Analysis Approach to Modeling Human Fecundity.” Biostatistics, 4–17.

34.

Trussell

1991. “Methodological Pitfalls in the Analysis of Contraceptive Failure.” Statistics in Medicine 10 (2): 201–220.

35.

Trussell

2011. “Contraceptive Failure in the United States.” Contraception 83 (5): 397–404. https://doi.org/10.1016/j.contraception.2011.01.021.

36.

Trussell

Grummer-Strawn

. 1991. “Further Analysis of Contraceptive Failure of the Ovulation Method.” American Journal of Obstetrics and Gynecology 165 (6 Pt 2): 2054–2059.

37.

Trussell

Portman

. 2013. “The Creeping Pearl: Why Has the Rate of Contraceptive Failure Increased in Clinical Trials of Combined Hormonal Contraceptive Pills?” Contraception 88 (5): 604–10. https://doi.org/10.1016/j.contraception.2013.04.001.