Behavioral Methods of Contraception

Behavioral Methods of Contraception

Anita L. Nelson, MD

Diana Crabtree Sokol, MD, MSc


Behavioral methods of contraception, such as fertility awareness-based methods (FABMs), coitus interruptus (withdrawal), and a range of other sexual practices, appeal to couples who might, for religious, personal, or medical reasons, wish to avoid contraceptive methods that require medication or devices. Behavioral methods also appeal to those couples who, at least temporarily, lack access to modern methods. Compared to other contraceptives, these methods are associated with the largest gaps between failure rates with correct and consistent use and those with typical use. Patient education and motivation are particularly critical to success with these methods.1

However, when considering behavioral methods, we often forget to include the lactational amenorrhea method (LAM), which is more effective than FABM and withdrawal and is specific to the postpartum woman. LAM is vitally important for short-term and long-term maternal and child health. In this chapter, we devote an entire section to postpartum breast physiology and LAM.

FABMs utilize various techniques to detect at-risk days in the woman’s cycle when unprotected intercourse is most likely to result in pregnancy. During those identified at-risk days, couples can choose to abstain from intercourse altogether (periodic abstinence) or to use other contraceptive methods of birth control, usually barriers or withdrawal.

However, we stress that abstinence alone is not a behavioral method of contraception. Abstinence is a behavior—a lifestyle choice. Conspicuously, women, men, or couples who choose abstinence are not sexually active and, therefore, do not need contraception. It is important to note that people that choose abstinence may use the method intermittently and not continually; thus, these persons may still need education about contraceptive options. Data from the National Survey of Family Growth (NSFG) reveal that over 7% of reproductive age women report having had sexually intercourse in the last 3 months without using a method of contraception.2 We need to improve the use of effective contraceptive strategies for women and men who consider themselves abstinent, but engage in occasional sex.

Globally, FABM, withdrawal, LAM, and other behavioral approaches are classified as “nonmodern” or “traditional methods.”3 Utilization of the
different methods in this category varies greatly across the world. As of 2016, 3.0% of reproductive age contracepting women worldwide report using periodic abstinence, but in some areas, that number can be up to 20%.4 In the United States, estimates of FABM use are complicated by assignment rules. The percent of sexually active women who relied on FABM use was estimated to be 2.2% in the 2013 to 2015 NSFG, but rose to 3.2% if couples who used condoms, withdrawal, or EC during at-risk days were included.5 Withdrawal by itself is used by many more couples; most people have used withdrawal at some time in their lives.6 In 2008 to 2012, 15.8% of women reported consistently relying on withdrawal and intermittent or episodic use was even more common.7 In a 2012 survey of 4,634 U.S. women aged 18 to 39, 13% reported that withdrawal was the most effective method they had used in the last 30 days; 12% relied on it exclusively, but another 21% had used it in combination with some other more effective method.8

Behavioral methods other than LAM are associated with higher failure rates than barrier, hormonal, intrauterine, and permanent contraceptive methods. Women who participate in clinical trials of FABM may be highly motivated to use the method but differ from women in clinical trials of more effective options in their willingness to accept a higher pregnancy risk. Collectively, FABMs are estimated to have a 15% first-year failure rate in typical use.9 With correct and consistent use, FABM techniques have first-year failure rates that range from 2% to 23%.9 Withdrawal has a 4% first-year pregnancy rate with correct and consistent use, but a 20% first-year failure rate in typical use.9 A systematic review concluded that effectiveness studies of each FABM in typical use are few and only of low to moderate quality; variability in the studies precludes any comparisons between individual FABM techniques.10,11 Most studies show high first-year discontinuation rates for FABM.12

FABM, withdrawal, and other behavioral methods require clear communication between sexual partners, agreement about their reproductive plans, and a commitment from both partners to make their method work. The success of FABM assumes the woman has the ability to say no to unprotected intercourse, but the 35% of women who suffer sexual/physical violence and coercion worldwide may well lack that ability.13

Many clinicians consider behavioral methods to be poor options for most women.14,15 In some circles, withdrawal is not even considered a legitimate contraceptive method, even though its contributions to fertility control have been well documented for over a century. For sexually active women, health care providers tend to discourage use of these behavioral methods, especially FABM, because of the relatively high failure rates and, perhaps, their personal lack of familiarity with FABM, their lack of time to provide adequate counseling, or their underestimation of the potential these methods offer for contraception.16 Obstetricians often do not routinely discuss LAM as a contraceptive, especially in developed countries where breastfeeding rates are low to moderate and breastfeeding for a full 6 months is uncommon.17 Depriving a patient access to expert guidance about
these methods forces her to seek out information from less reliable sources, which perpetuates the cycle of suboptimal usage and higher failure rates.

There are definite advantages to behavioral methods. These methods have no or minimal direct costs. No drugs or medical system interactions are required for most techniques. Withdrawal is always available; it can be used with any coital act at any time in the cycle. It is one of the only methods available to men. There are no medical contraindications to any behavioral method although cycle variability in women with irregular menstrual cycles precludes use of most FABM and the onset of menstrual-like bleeding provides notice that LAM should no longer be used.

Fertility Awareness-Based Methods

FABMs utilize various techniques to detect at-risk days in the woman’s cycle when unprotected intercourse is most likely to result in pregnancy. During those identified at-risk days, couples can choose to abstain from intercourse altogether (periodic abstinence) or to use other contraceptive methods of birth control, usually barriers or withdrawal. FABMs help couples gain understanding about their reproductive functioning and potential. This understanding can facilitate achieving pregnancy when it is desired and can help women recognize and anticipate symptoms of premenstrual syndrome or other catamenial problems.18 FABMs also provide an opportunity for anovulatory women to detect their reduced fertility.

The various FABMs differ in the techniques used to calculate at-risk days and what methods to use during those days. There are three approaches used to calculate at-risk days (Box 11.1).

Each technique has traditional tools for women to use, but most also utilize more modern apps. In older parlance, these methods were collectively referred to as “natural family planning.” This unfortunate phrase should be avoided, not only because it implies that all other methods are “unnatural” but also because it is technically incorrect. Libido in women is known to increase around the time of ovulation.19,20,21 To call a method “natural” that asks couples
to abstain at the time in the woman’s cycle when she is most “receptive” to sexual advances seems to be a cruel perversion of terminology.10

Each FABM technique relies on assumptions about the biology of ovulatory cycles and fertilization. Although there have been modern refinements in the estimates of fertility potential by cycle day,22 most FABMs still rely on earlier observations of the timing of ovulation23 and assume the following about ovulation and sperm.

Ovulation occurs at the following median times:

  • 35 to 44 hours after the initial rise of luteinizing hormone (LH)

  • 16 hours after the LH peak (range 8 to 40 hours)

  • 24 hours after the estradiol peak (range 17 to 32 hours)

  • 8 hours after the rise in progesterone (range 12.5 hours before to 10 hours after)

  • 12 to 16 days before the onset of next menses, depending on cycle length

In addition, the following observations about ovulation and sperm are assumed to be true:

  • No more than one ovulation occurs each cycle.

  • Ova can only be fertilized during the 12 to 24 hours following ovulation.

  • Sperm have the capacity to fertilize ova 3 days after they have been deposited in the vagina (however, this is really 5 days—see below).

Taken together, these estimates lead to the conclusion that there are only 6 days in a cycle when fertilization is possible—the 5 days preceding ovulation and the day of ovulation. However, variability in cycle-to-cycle lengths means that the time during a cycle when those 6 days occur may fluctuate, and even more days must be considered to be at risk when prospective assessments of fertility potential are being made. In addition, the fact that sperm can fertilize an egg for up to 5 days (not just 3 days as assumed in early FABM models) insures some built-in failure rate but makes the cycle length-based methods more acceptable by reducing the number of at-risk days.

Cycle Length-Based Methods

“Calendar Method” and “Rhythm Method”

Success with these cycle length methods requires that the woman has relatively minor variability in her cycle-to-cycle lengths. To determine if a woman can rely on these methods to identify the at-risk days, she must first record her menstrual cycles for 6 to 12 months. Classically, women collected this information during the year of engagement that preceded their wedding at a time when it was assumed the couple needed no contraception. Once enough information has been collected to determine the woman’s shortest and the longest cycles, her at-risk days are identified using the following formulas:

  • First at-risk (fertile) day each cycle = number of days in shortest cycle minus 18

  • Last at-risk (fertile) day each cycle = number of days in longest cycle minus 11

Recent studies have shown that 20% of women studied for just four cycles had cycle lengths that varied by at least 14 days, making this method almost unfeasible. Another 26% had cycles that varied by 7 days or more, which would mean that they would be at risk for pregnancy at least 15 days per cycle.24

To use this technique, the woman indicates the at-risk days for the cycle on a calendar (paper or electronic) as soon as she starts her menses (hence the name “calendar method”). Women must be aware of cycles that are outliers (extra-long or short cycles) to see if they need to recalculate their at-risk days. Clinicians should be aware that women often report using the “rhythm method” without gathering the needed baseline information or doing any calculations; they only abstain for a few days when they sense (often incorrectly) that they are fertile.25,26,27

An older meta-analysis reported first-year failure rates for the calendar/rhythm method ranged from 15% to 19%.28 An international study of women who used periodic abstinence based on the calendar method reported a typical use failure rate of 14%.29 The unpredictable nature of a late ovulation, even in women who typically have regular cycles, likely contributes to these failure rates.30 If couples use barrier methods during at-risk days, the failure rates may be lower because of higher compliance.27

Standard Days Method (“CycleBeads”)

More recently, props, such as “CycleBeads,” have been introduced to help women (couples) track at-risk days. The Standard Days Method relies on the observation that 78% of cycles range between 26 and 32 days, meaning that fertility is most likely to occur on days 8 to 21.31 The beads (Figure 11.1) are color coded to alert users to at-risk days and to let them know if the cycle length is outside the appropriate range for the method.

Figure 11.1 CycleBeads. An elastic band is advanced over a new bead each day to inform the user (and her partner) where she is in her cycle. CycleBeads are available at

Color-coded beads are assembled in a circle; an elastic band is advanced over one colored bead a day. The red bead represents the first day of bleeding and is followed by 7 brown (safe day) beads. CycleBeads 8 to 19, which represent the at-risk days, are colored white; in some versions, these beads glow in the dark. Beads 20 to 25 are brown. Bead 26 is brown-black and is used to remind women that if their menses start before this day, their cycles are too short, and this is not an appropriate method for them to use. Beads 27 to 31 are brown again. The 32nd bead is a black cylinder with an arrow arm showing the direction to advance the elastic band; this color reminds the woman that if she has not started her menses by then, she should consider pregnancy testing, or that her cycles are too long for her to rely on these beads.

First-year failure rates from the Standard Days Method clinical trial were 4.8% with correct and consistent use and 12% for the whole study population.32 A meta-analysis of studies from 14 countries where the Standard Days Method was provided as a routine option showed a 14% failure rate,33 consistent with a recent systematic review finding of 12%.29 Interestingly, in one clinical trial, 23% of women had unprotected intercourse during at least one identified fertile day during one or more of the 13-cycle trials.34 Based on these failure rates, WHO has classified Standard Days Method as a “modern method,” a classification that has been openly challenged.35,36

“Apps” Based on Cycle Length Alone

Approximately 80% of 18- to 49-year-olds in the United States own a smart phone. Each of the technology giants, like Apple® and Google®, provides health applications for free to their users. Over 1,100 applications (apps) for tracking the menstrual cycles have come on the market, some requiring payment and others available for free.37 These applications are highly variable in quality. Some apps provide digital platforms for existing cycle length-based methods, such as the calendar method and the Standard Days Method (available at The data collected by some apps have been used to develop algorithms to predict fertility. For example, using the data of over 7,000 women with reliable menstrual and ovulation records, investigators fitted linear functions to the length of the cycle and timing of ovulation and concluded that the middle day of a mean menstrual cycle was the best estimate of the next ovulation.38 A dynamic optimal timing (DOT) model was developed based on data from two cohort studies with over 10,000 women with cycles of 20 to 40 days. These identified between 11 and 13 at-risk days per cycle.39,40 When women first start to use this method, many days are identified as high risk, but as information about the woman’s cycle becomes more robust with time, the algorithm can more accurately recognize fewer at-risk days.41 The first 6-month failure rate was 3.5%; the researchers reported that all those pregnancies resulted from incorrect use.42

In general, smart phone applications based on menstrual cycle data alone have been disappointing. Fewer than 20% of apps correctly predict the atrisk dates.37,43 Because of this high error rate, it is important the clinicians guide their patients to accurate apps. Generally, apps discussed later in this chapter that utilize additional clues about ovulation timing are better options. Some observers have raised questions about the social, cultural, ethical, political, and privacy issues raised by these apps.44

Sign/Symptom-Based Ovulation Detection

Another approach to identify fertile days relies on biologic changes that can be detected around the time of ovulation. Persistently elevated estradiol levels (>200 pg/mL for 50 hours) that trigger the LH surge prior to ovulation induce several changes in cervical mucus, vaginal secretions, and saliva. Other physical changes, such as basal body temperature (BBT) and alterations in cervical mucus texture and composition, generally follow in short order. Together, these effects can help detect impending ovulation; their resolution can suggest the end of the at-risk days. These methods can be used by regularly cycling women as well as those with less predictable menstrual patterns, including women transitioning from breastfeeding or hormonal contraceptive use and women with irregular or infrequent menstrual cycles.45,46

For most of the cycle, the mucus secreted by the cervix is composed of poorly hydrated mucin. The resulting mucus is thick, scant, and viscous, forming a barrier to sperm and microorganisms. High estrogen levels at midcycle induce a number of changes to the composition and hydration of mucus that result in an increase in volume and a highly fluid low-viscosity consistency favorable to sperm penetration.47,48

Billings Ovulation Method

The Billings technique uses cervical mucus changes to monitor potential fertility throughout the woman’s cycle. This method was approved by the Vatican and tested and promoted by the World Health Organization (WHO).10,49 The woman evaluates the characteristics of her cervical mucus sampled at the introitus at least twice daily after her menses. Some experts advise that the woman assess her cervical mucus before and after each episode of urination by wiping front to back with toilet tissue to obtain a specimen. The woman assesses mucus presence or absence, quantity, color, fluidity, glossiness, transparency, and stretchiness (elasticity), charting the most fertile characteristics of the mucus over the entire day on a colorcoded chart each night. She should evaluate the mucus for one to three cycles to familiarize herself with her patterns before attempting to rely on the method. Couples committed to this method need to avoid actions that complicate mucus evaluation including douching and intercourse on consecutive days; the day off permits mucus evaluation without any contamination by the ejaculate.

Early phase cervical mucus is scant and sticky. The volume of cervical mucus increases 2 to 9 days before ultrasound-detected ovulation.50
The Billings method assumes that cervical mucus can be detected on the vulva near the introitus by most women 4 to 7 days ahead of ovulation.51 As ovulation approaches, the mucus becomes more abundant, clearer in color, and elastic. The last day with clear (transparent like a raw egg white), slippery, and elastic (it can be stretched for several centimeters before it breaks) mucus is called the “peak day.” This day, of course, can only be retrospectively diagnosed. In the Billings method, the peak day mucus is assumed to occur at the time of ovulation. However, more recent studies show that ovulation actually occurs with a mean of 0.9 days following that peak day but ranges from 2 days before peak mucus to 3 days after.50

Classically, women using the Billings method were told to follow four rules to utilize their cervical mucus findings as effectively as possible to prevent pregnancy:52,53

  • There must be no intercourse if the woman has any bleeding or spotting.

  • Early in her cycle, when there is no cervical mucus or the secretions are consistently scant, cloudy, and sticky, intercourse can occur but no more frequently than every other day (although multiple acts are permitted on any 1 allowable day).

  • Coitus is to be avoided once mucus becomes abundant and clear.

  • Abstinence must be maintained or a backup method used until 4 days following the peak day mucus (assumed ovulation day). If the earlier cervical mucus changes resolve without signs of ovulation, intercourse may resume after 4 days of dryness.

Under the Billing method, 95% of all women will have 4 to 12 days of observable secretions, so many days of abstinence or other method use may be required. Since sperm can survive for up to 5 days, some experts now advise that women who rely on the Billings method classify all days when secretions can be detected as “at-risk days.”54 Older studies estimate firstyear failure rates with classical Billings method to be 3% and 23% in more generalized use.52,54,55 Cervical or vaginal infections and vaginally administered medications interfere with interpretations of cervical mucus.

A modification of the Billings method, the Creighton model NaPro Education Technology, uses picture and a more precise, standardized way to describe the cervical secretions.56

TwoDay Method

This method greatly simplifies the Billings method by decreasing the number of cervical mucus analyses. The TwoDay Method is a real-time, focused method in which the patient touches her introitus (with finger or tissue) at least once daily or as frequently as each time she urinates to determine by sensation or observation if there is any moisture present. The woman asks, “Am I dry today?” If she is dry, she then asks “Was I dry yesterday?” Unprotected coitus is allowed only if she has been dry for 2 consecutive days. Although multiple acts of coitus may be allowed on any one day, the
next permissible day for coitus would be 3 days later because the moisture from that day’s ejaculate will be detected as wetness the following day. This method is portable, allows for cycle-to-cycle variability (including infrequent menses), and is easy to teach. One multicountry study found that first-year failure rate with correct and consistent use was 3.5%; the total first-year failure rate in the study was 14%.57 Typical use failure rates are not available.

Basal Body Temperature Method

This method is based on the fact that progesterone is thermogenic. In the follicular phase, BBT generally ranges from 97.0°F to 98.0°F. One day before ovulation, the BBT drops to its nadir. With the formation of the corpus luteum following ovulation, progesterone production increases and the BBT rises 0.5°F to 1.5°F. The higher temperature is maintained until the progesterone levels fall in late in the luteal phase, usually within 1 to 2 days before or at the onset of menstruation.47

To detect changes in temperature, a special BBT thermometer is required. Whereas a standard thermometer is accurate to 0.2°F, a BBT thermometer is accurate to 0.1°F. Women are asked to plot, on paper or on an electronic chart, their BBT taken at rest at the same time each day using the digital thermometers applied to the same site (e.g., oral, axillary, etc.) without arising for the preceding 3 to 6 hours. These conditions can be challenging. Illness, sleep disturbances, getting out of bed to urinate, or answer a baby’s call can render the measured temperature invalid. Sleeping in late on weekends is not permissible. Even if the woman can collect accurate measurements, only 80% of cycles may have interpretable patterns.

To interpret the results, ovulation can be detected by using the “coverline method” or the “three-over-six rule.” In the former, a horizontal threshold line is drawn at a level 0.15°F above the highest temperature recorded in the first 10 days of the cycle; ovulation is documented when temperatures taken later in the cycle cross that line. For the second, three consecutive temperatures must be 0.2°F higher than the highest point of the previous six temperature measurements, with at least a 0.4°F difference between the lowest and highest measurements.47 BBT measurements are most effective as a postovulatory method that allows unprotected intercourse to start only after BBTs have been elevated for at least 3 days. Abstinence or other methods must be used at all other times earlier in the cycle—usually 16 days a cycle.18

Hormonal Testing

Most hormone monitors are designed to detect rising estradiol levels, which portends the onset of fertility and LH levels to indicate impending ovulation. Ovulation does not occur before the urinary LH peak day or the serum LH surge day.50,58 Over-the-counter kits to detect the presence of urinary LH are best used once or twice a day starting from idealized cycle day 10 or 11 to detect the initial LH increase.58 Interestingly, in 75% of cases, the onset of the LH surge starts between midnight and 08:00 hours.47,59 Once LH is
detected by these tests, ovulation generally follows in 20 ± 3 hours.47 But since these tests may detect LH well before the peak, ovulation may occur as late as 3 days after a positive test.47

Urinary LH detection kits may be helpful for timing intercourse when couples seek pregnancy, but by themselves may have limited value in preventing pregnancy unless they are used in algorithms that permit unprotected intercourse only in the postovulatory period (4 days after LH surge). This limitation is because the LH surge occurs too close to the time of ovulation and cannot alert couples of impending risk 3 to 5 days before ovulation. LH detection may also be useful as a component in mixed methods.

Other body fluids have also been tested for their ability to predict ovulation. In 88% of cycles, at-risk days correspond to times when ferning is seen in saliva samples, a sign of impending ovulation. This test is used in some family planning products such as Ovate 1 Ovulation Monitor and Maybe Baby Saliva Ovulation Tester. On average, the onset of ferning precedes ovulation by 6 days. Electric resistance of saliva and vaginal secretions also changes about 6 days prior to ovulation.

Mixed Methods

Symptothermal Methods

This category includes a wide range of method combinations. When BBT measurements are paired with cervical mucus monitoring, unprotected intercourse before ovulation that would not be allowed with BBT testing alone is possible every other day. This combination relies on cervical mucus changes to identify infertile days during the follicular phase (dry days). At-risk days start when changes in cervical mucus are noted and end after both techniques agree intercourse is safe: 4 days after peak mucus and 3 days after the onset of temperature rise.

Cycle length methods can also be added to BBT shifts with or without cervical mucus determinations. The consistency of the cervix (hard vs. soft), its position (verted or more central), and external os characteristics (open or closed) can also be monitored as part of mixed FABMs. Mittelschmerz, breast tenderness, vulvar swelling, and other symptomatic indications of impending ovulation can also be added to the list of techniques that are used to identify at-risk days.60

Symptohormonal Methods

One version, the Marquette method (, combines urinary LH and estrone-3-glucuronide levels measured with the Clearblue Easy-Fertility Monitor and cervical mucus methods. In one study, this approach was associated with a first-year failure rate of 2%.61,62 A multicenter U.S. study used those elements along with BBTs, which resulted in a perfect-use 12-month pregnancy rate of 0.6% and a total 12-month pregnancy rate of 6.6%.62

Only gold members can continue reading. Log In or Register to continue

Mar 1, 2020 | Posted by in OBSTETRICS | Comments Off on Behavioral Methods of Contraception
Premium Wordpress Themes by UFO Themes