Patient Education and Contraceptive Compliance
Paula J. Adams Hillard
Table Of Contents
Paula J. Adams Hillard, MD
THE PROBLEM ON UNINTENDED PREGNANCIES|
COMPLIANCE WITH SPECIFIC METHODS OF CONTRACEPTION
|THE PROBLEM ON UNINTENDED PREGNANCIES|
Definition and Extent of the Problem
U.S. data from the 1995 National Survey of Family Growth (NSFG) revealed that 28% of all women of reproductive age had ever experienced an unintended birth.1 About 20% of these women indicated that the birth had been unwanted, while 80% stated that it was mistimed. Estimates from earlier NSFG data cycles factoring in pregnancies that ended in abortion placed the percentage of unintended pregnancies (as opposed to births) as greater than 50%.2 An equal proportion of unintended pregnancies end in abortion (44%) compared with birth (43%).3 About one half of unintended pregnancies occurred among the 10% of women using no method of contraception. The remaining half (1.7 million pregnancies in 1988) occurred in women who were using a method of contraception, which demonstrates that all methods of contraception can fail to prevent pregnancy.2
Contraceptive efficacy can be calculated from population-based surveys such as the NSFG or from clinical trials and investigations. In general, failure rates derived from population-based surveys are higher than those from clinical trials, in part because women who participate in clinical trials are likely different from those who do not and may be more motivated or have incentives to use a given method of contraception correctly.4 A precise definition of contraceptive effectiveness is the proportional reduction in the monthly probability of conception, a value which is neither observable nor accurately estimated.4 Thus, contraceptive efficacy is usually assessed by measuring the number of pregnancies that occur during a specified interval of exposure to a given contraceptive method and is reported using either the Pearl index or life table techniques. The Pearl index is widely used and is required by the U.S. Food and Drug Administration (FDA) for approval of a new method of contraception. It is defined as the number of pregnancies per 100 woman-years of exposure. Because failure rates for most methods of contraception decline over time as women gain experience in using the method or those most likely to have a failure get pregnant sooner, the life table analysis calculates a separate failure rate for each month of use. When failure rates using the life table method are reported, 12 months is frequently given as a reference period.
Contraceptive effectiveness depends on both the inherent effectiveness of the method itself and correct or perfect usage of the method. The inherent or theoretical efficacy of a method is difficult if not impossible to ascertain, and even perfect use will not result in zero failures. Failure rates among couples who used a contraceptive method perfectly (both consistently and correctly) have been estimated to range from 0.1% to 10%.5 The failure rates of contraceptive methods during actual use have been shown to be higher than the estimated rates for perfect use.
The failure rate of a method during actual use depends on a number of factors, such as age, experience in using the method, and motivation to prevent pregnancy. These factors result in variations in the percentage of users who use a given method incorrectly or inconsistently. Estimates of failure rates among typical U.S. married women are included in Table 1, along with the lowest expected failure rates.6 Jones and Forrest addressed the issue of underreporting of abortions in the NSFG and arrived at corrected failure rates by method of contraception, age, marital status, and race.7 These corrected failure rates are much higher than the estimated failure rates for perfect use and vary widely depending on the characteristics of the user. For example, the failure rate of white users of oral contraceptive pills aged 35 to 44 years was 2%, whereas nonwhite oral contraceptive pill users younger than 20 years of age experienced an 18% failure rate. In general, failure rates declined with increasing age.
TABLE 1. Percentage of Women Experiencing an Unintended Pregnancy During
the First Year of Typical Use and the First Year of Perfect Use of Contraception
and the Percentage Continuing Use at the End of the First
Year in the United States
a Among typical couples who initiate use of a method (not necessarily for the first time), the percentage who experience an accidental pregnancy during the first year if they do not stop use for any other reason.
b Among couples who initiate use of a method (not necessarily for the first time) and who use it perfectly (both consistently and correctly), the percentage who experience an accidental pregnancy during the first year if they do not stop use for any other reason.
c Among couples attempting to avoid pregnancy, the percentage who continue to use a method for 1 year.
d The percentages becoming pregnant in the two middle columns are based on data from populations in which contraception is not used and from women who cease using contraception to become pregnant. Among such populations, about 89% become pregnant within 1 year. This estimate was lowered slightly (to 85%) to represent the percentages who would become pregnant within 1 year among women now relying on reversible methods of contraception if they abandoned contraception altogether.
e Foams, creams, gels, vaginal suppositories, and vaginal film.
f Cervical mucus (ovulation) method supplemented by calendar in the preovulatory and basal body temperature in the post-ovulatory phases.
g With spermicidal cream or jelly.
h Without spermicides.
i The treatment schedule is one dose within 72 hours after unprotected intercourse and a second dose 12 hours after the first dose. The U.S. Food and Drug Administration has declared the following brands of oral contraceptives to be safe and effective for emergency contraception: Ovral (1 dose is 2 pills), Alesse (1 dose is 5 pills), Nordette or Levlen (1 dose is 4 pills), Lo/Ovral (1 dose is 4 pills), Triphasil or Tri-Levlen (1 dose is 4 pills).
j However, to maintain effective protection against pregnancy, another method of contraception must be used as soon as menstruation resumes, the frequency or duration of breastfeeds is reduced, bottle feedings are introduced, or the baby reaches 6 months of age.
(From Trussell J, Kowal D: The essentials of contraception. In: Hatcher RA, Trussell J, et al, eds. The Essentials of Contraception: Efficacy, Safety, and Removal Considerations in Contraceptive Technology, 17th ed, pp 216---217. New York: Ardent Media, Inc., 1998.)
A number of authors have commented on the imprecision with which the terms contraceptive efficacy, effectiveness, failure rate, and pregnancy rate are used and the extent to which characteristics of the user influence the determination of these measures.4,8,9 The inherent level of protection of the method, differences in fecundity by age, frequency of intercourse, and exposure to the risk of pregnancy influence these measures.8 In addition to these factors, the gap between the lowest expected failure rate of a given contraceptive method and the failure rate in typical use is due to differences in the correct and consistent use of the method (i.e. compliance).
Compliance has been defined as “the extent to which a person's behavior … coincides with medical or health advice.”10 Compliance in contraception refers to “the use of a contraceptive method in an ongoing and consistent manner for the prevention of pregnancy.”11 Thus both continuation and correct use are required. Although some have argued that the term compliance has paternalistic and pejorative connotations, it is widely used in contraception literature. The terms adherence or just simply correct or successful use have been suggested, although they have not yet been widely adopted.
Compliance has been studied in a number of different medical settings, and although there are important differences between complying with a 10-day prescription of antibiotics and the use of oral contraceptives on a daily basis over a number of years to prevent pregnancy, rates of compliance with contraception differ little from rates of compliance with other medical treatments.12 With the antibiotic regimen, compliance is enhanced by a decrease in symptoms; with oral contraceptives, pregnancy is avoided, although this does not provide the same ongoing reinforcement for compliance. The treatment of symptoms is generally an unequivocally positive result, whereas the decision to avoid pregnancy can be associated with ambivalence. The prescription of an antibiotic by a physician involves little choice from the patient, whereas there are a number of patient-chosen options for contraception. The antibiotic regimen is time-limited, whereas the use of oral contraceptives is ongoing. Finally, contraceptive decisions involve complex interactions between sexual partners in a social milieu. For all of these reasons, the study of compliance with contraceptive regimens is complex compared with the study of compliance with other medical regimens.13 However, Cramer has concluded that the study of medical disorders has shown that “no consequence is so severe that all patients can be assumed to comply with the prescribed treatment plan.”12
The issues of contraceptive compliance have received increasing attention and focus in the medical literature. The term, concepts of compliance, and practical problems which contribute to less successful contraception are now familiar to most clinicians. A search of the medical literature using the terms “contraception” and “compliance” revealed no such references prior to 1978; in 1988, there was a marked increase in the number of such citations, and a plateau in citations has occurred since that time (Fig. 1).
In general, research on contraceptive compliance has focused on continuation rather than correct usage.14,15,16,17,18,19 Discontinuation of a method of contraception may be followed by continued sexual activity and the choice of an alternative method of contraception (either reversible or permanent); continued sexual activity and the choice not to use a method of birth control; the decision to be celibate; or the choice of sexual activity that does not entail a risk of pregnancy.
Women who discontinue one method of contraception and who continue to be sexually active while wishing to avoid pregnancy must be encouraged to use another method immediately without any gap in protection. The most effective method for an individual woman is one that she has chosen to use consistently and that suits her individual needs. Pratt and Bachrach noted that 8% of women who discontinued oral contraceptive use did not adopt any method of contraception, continued to be sexually active, and were not seeking pregnancy.20
Nearly one half of unintended pregnancies end in induced abortion.21 It has been estimated that as many as 70% of all abortions could be avoided if all women at risk for unintended pregnancy used the most effective methods of contraception.22 Up to 75% of women undergoing abortion report that they were practicing contraception during the month in which they conceived, representing individuals who experienced a contraceptive failure due to user or method factors.23,24,25 Nearly one half of prior users had discontinued a method of contraception within 3 months prior to conception, representing individuals who were caught in the gap between contraceptive methods.23 Thus, induced abortion is a significant outcome of many unintended pregnancies that result from problems with contraceptive compliance.
Data from the 1982 NSFG were analyzed to address the issue of contraceptive discontinuation.14 About one third of married women discontinued using a method of contraception in the first year of use. Discontinuation rates ranged from 16% for periodic abstinence to 45% for spermicide use. Twenty-six percent stopped using oral contraceptives, 29% discontinued use of condoms, 30% discontinued using a diaphragm, and 21% requested intra-uterine device (IUD) removal. Although some women switched from one method of contraception to another method, about one half continued to be at risk for an unintended pregnancy because they did not use another method of birth control.14
Discontinuation rates vary depending on characteristics of the populations studied. Trussell and associates (see Table 1) list the percentage of women estimated to be continuing use of a given method of contraception at the end of 1 year.26 Two alternative assumptions could be made: that no one becomes pregnant, or that the percentage of women who continue to use the method excludes the women who experienced an accidental pregnancy in the course of the year. Adolescents who begin using a method of contraception are more likely than older women to discontinue its use. The problems of compliance with contraception in adolescents have been widely addressed in the literature.15,16,27,28,29,30 In one study, up to two thirds of oral contraceptive pill users were no longer using the pill at the end of 1 year.31
Factors that have been associated with contraceptive compliance in adolescents include a number of social, developmental, and behavioral variables.32 These variables include the perceived risk of pregnancy, frequency of intercourse, degree of intimacy with a sexual partner, acknowledgment of sexuality, feelings of ambivalence about sexual behavior, religious commitment, emotional and sexual maturity, support by significant others, and experience with contraception.11 Health beliefs about contraceptive methods also relate to both the intention to use contraception and the actual use of contraception.33 Some adolescents who fail to return for follow-up appointments for contraception are not sexually active and thus feel that they do not need ongoing contraception. However, the pattern of adolescent sexual activity tends to be serial monogamy. The individual adolescent who is not currently sexually active may resume sexual activity in a new relationship. She may or may not use contraception effectively with a new partner.
Issues related to contraceptive discontinuation in women beyond the age of adolescence have not been addressed as thoroughly. Why do older women discontinue using a given method of contraception? The answer is complex, but there are a number of psychological factors that relate to satisfaction with the method. Miller has stated that “a woman's contraceptive vigilance … frequently depends upon … the internal balance between her positive and negative feelings toward getting pregnant and between her positive and negative feelings about her current contraceptive method.”34
In addition to psychological traits and attitudes, a variety of other factors affect contraceptive use and compliance. Demographic variables such as age, education, religion, socioeconomic status, marital status, and number of children impact on the effective use of contraception. Older, better-educated women of higher socioeconomic status and income have been shown in some studies to have fewer accidental pregnancies. The motivation to use contraception effectively varies depending on the intent of contraception: to postpone pregnancy, to space births, or to avoid pregnancy.35
Factors related to the specific method of contraception may effect compliance. Methods that are coitally related may be perceived as inconvenient or a barrier to spontaneity, and may thus be used less effectively. Cultural factors also play a part; in some societies, injections are viewed as powerful and effective medicines that may make compliance more acceptable with this mode of administration.
The experience of side-effects is the primary reason that many women give for discontinuing a method of contraception, particularly the most popular reversible method, oral contraception.17,19,20,36,37,38,39 Although some of the studies addressing this issue were performed at a time when oral contraceptive pill formulations contained higher doses of steroid hormones and were thus more likely to be associated with side-effects, the experience of side-effects or even the fear of side-effects remain important factors in contraceptive discontinuation. Continuation rates have been shown to be better with 35-μg estrogen-containing pills than with 50-μg pills.40
It has been stated that the “accurate measurement of compliance is not easy; easy measurements of compliance are not accurate.”10 Methods of measuring compliance include direct methods, which involve biologic assays of serum, urine, or saliva for the level of the prescribed drug or the presence of a marker substance. These methods have been used with oral contraceptives on a very limited basis and are expensive and time consuming for larger populations. Indirect methods of measuring compliance include physician estimates, self-reports or diaries, and measurements of outcome; in the case of contraceptives, the measurement of contraceptive failure is pregnancy.30 Many studies of contraceptive compliance relied on patient self-reporting, although issues surrounding sexuality, privacy, or the personal choice of pregnancy termination may make it more difficult for an individual to admit to less than perfect use of a contraceptive method or to admit to an unintended pregnancy. Innovative ways of recording oral contraceptive pill use electronically have been studied41 and may have future practical uses, such as triggering a reminder alarm. Measurements of outcome (i.e. pregnancy rates) are an inaccurate reflection of compliance, because not all noncompliance results in pregnancy. Pregnancy does, however, represent one measure—the “bottom line”—of contraceptive compliance.
|COMPLIANCE WITH SPECIFIC METHODS OF CONTRACEPTION|
Currently, the most widely used method of reversible contraception in the United States is the oral contraceptive pill, which is used by a little more than one fourth of American couples who are using a method of contraception.1 Worldwide, more than 60 million women are users of oral contraceptives.42
A number of studies have cited the experience of side-effects as a primary reason for oral contraceptive discontinuation or method switching.15,17,19,20,36,37,43 In the past, physicians' concerns about the side-effects of oral contraceptives centered on the risks of medically serious events such as myocardial infarction, stroke, or hypertension. Side-effects that could be bothersome or worrisome to patients, such as nausea or breakthrough bleeding, were given less attention than the more medically serious risks.
The terms “major” and “minor” side-effects illustrates the conceptual framework which was operative in the early years of oral contraceptive use. Oral contraceptive pill formulations have changed over 30 years of use from higher-dose estrogen and progestin-containing pills to lower-dose pills; today's oral contraceptive pills are not the same pills as were used in years past. In addition, data regarding the risks of serious medical complications such as myocardial infarction have been analyzed to demonstrate that smoking is the primary risk factor for cardiovascular events.44,45,46
Although the risks of myocardial infarction in pill users are thought to be primarily due to the effects of smoking, the issue of venous thromboembolism in oral contraceptive users has recently been raised. In 1995, the United Kingdom's Committee on the Safety of Medicines recommended that women should use oral contraceptives containing desogestrel or gestodene only if prepared to accept an increased risk of thromboembolism. This action was based on observational studies that indicated a twofold to threefold increase in the risk of thromboembolism with pills containing these compounds (i.e. third-generation pills) compared with products containing levonorgestrel.47 The fact that these studies point in the same direction is cause for concern, but a number of authors have commented on the likelihood of unrecognized biases in these studies.48 Subsequent studies and reanalyses of the data suggest that a somewhat increased risk may occur with these pills compared with other oral contraceptive pill formulations, although the relative risk is still substantially less than the risk of venous thromboembolism associated with pregnancy, and the absolute magnitude of the risk is small.49,50 However, the initial and subsequent reports and associated publicity led to a major “pill scare” epidemic in the United Kingdom and Europe, where government bodies reacted with a variety of warnings regarding the risks of the pill.51 Many women stopped taking their oral contraceptive pills because of this scare. An increase in the number of abortions was documented in Norway, where sales of oral contraceptives dropped by 17% after the October 1995 report, and a significant 36% increase in abortions occurred during the first quarter of 1996.52 This is a clear illustration of the possible adverse effects of media and a scare about possible medical effects of oral contraceptives.
Other than this brief focus on medically serious potential side-effects of the pill, the emphasis on side-effects and complications of oral contraceptive use has primarily shifted to a focus on the “minor” side-effects. Do these minor side-effects matter? Drug companies have begun to shift to oral contraceptive formulations with lower doses of both estrogen and progestins with the premise that “lower is better.” The World Health Organization has concluded that the goals of contraceptive research is to achieve the lowest dose possible that will maximize contraceptive efficacy and menstrual regularity, minimize the incidence and severity of side-effects, and maintain associated noncontraceptive health benefits.53 It has been widely suggested and seems intuitively obvious that lower hormonal doses of estrogen should result in fewer estrogen-mediated side-effects such as nausea, breast tenderness, and fluid retention. Although comparative data are lacking, there are a number of reports documenting a low incidence of these side-effects among users of the newest 20-μg pills.54,55,56
Clearly, side-effects affect patient satisfaction with the method of contraception and continued use.38 Satisfaction with the clinician who prescribes the pill and the absence of side-effects have been associated with pill continuation and compliance.38 The woman who is dissatisfied with oral contraceptives for whatever reason may decide to discontinue using the method without consulting her physician. One study concluded that explicit medical advice played a less critical role in pill discontinuation than did a woman's own judgment.20
The following types of nuisance side-effects contribute to problems with ongoing use of oral contraceptives: irregular bleeding or lack of withdrawal bleeding, nausea, weight gain, breast tenderness, headache, acne, mood changes such as premenstrual syndrome (PMS) or depression, and skin changes such as chloasma.6 Some authors question the cause-and-effect relation between some of these side-effects and oral contraceptives, because symptoms such as headaches and depression have a high background rate in the population, and baseline rates of these symptoms are often not reported.57,58 Headaches in particular have not been found to occur more frequently among oral contraceptive pill users than among controls.59,60,61
Many of these side-effects were more common with high-dose oral contraceptives than with today's low-dose pill formulations. Data regarding the frequency of these nuisance side-effects are difficult to interpret, and studies reporting their frequency should be reviewed carefully to assess the specific definition of problems such as irregular bleeding. Side-effects have been shown to be the primary factor predicting early pill discontinuation. Multiple side-effects substantially increase the likelihood of discontinuation; in one study, a single side-effect increased the risk by 5%, two side-effects increased the risk by 220%, and three by 320%.39
Methodologic problems abound in studies reporting rates of breakthrough bleeding with various oral contraceptive pills formulations.62 Smoking increases the risk of breakthrough bleeding.63 Double-blind crossover studies comparing the incidence of nuisance side-effects among the low-dose pill formulations are not available. The incidence of breakthrough bleeding or spotting appears to be highest in the first few months of use and appears to be similar or perhaps slightly increased for triphasic formulations compared with monophasic formulations.6 The progestin component of the pill may influence the rate of breakthrough bleeding. Gestodene-containing pills may result in more bleeding,64 whereas levonorgestrel-containing pills may result in less bleeding.65
The potential consequences of abnormal bleeding to an individual patient may be considerable, although there are variations among cultures in the interpretation of the meaning of abnormal bleeding. Women who have irregular bleeding or spotting may become confused about the timing of their “real” menstrual period and thus be unsure about when to start a new package of pills. The experience of irregular bleeding may lead to dissatisfaction and pill discontinuation, even within the first few months of use, particularly if the patient has not been counseled that irregular bleeding may resolve over time with consistent use. Women may lose confidence in a method of contraception that creates nuisance effects such as unpredictable bleeding. A woman may lose confidence in the physician who does not appreciate the impact of irregular bleeding or who fails to warn her of its possibility.66 Irregular bleeding can create anxiety among women who fear that oral contraceptives cause cancer, a belief held by 31% of women surveyed in an American College of Obstetricians and Gynecologists (ACOG) Gallup poll.67 Increased anxiety may translate into additional telephone calls or office visits for reassurance, with a resultant cost in time and health care dollars. In addition, it has been estimated that more than 1 million unintended pregnancies occur each year in the United States as a result of discontinuation or misuse of the pill.68
Lack of withdrawal bleeding is another cycle control problem which may be even more anxiety provoking for women than breakthrough bleeding, primarily because of concern about a possible pregnancy. Studies have shown an incidence of lack of withdrawal bleeding of 2% to 6%.55,62,69 Beliefs about lack of withdrawal bleeding have been noted to include the concern that it “always means pregnancy” among 10%, the concern that it was “harmful” in an additional 33%, and the decision not to restart oral contraceptives in 39%.70
Nausea is a relatively infrequent side-effect with low-dose pills, because it is related to the estrogen component of oral contraceptives. However, one study of triphasic pills found that up to one fifth of women experienced nausea at some time during 6 months of use, with 0% to 2% experiencing vomiting.71 Studies have shown a decrease in the frequency of nausea from the first cycle to subsequent cycles.59 One of the rationales given for decreasing the estrogen dose in pill formulations is to decrease the incidence of nausea, an estrogen-mediated effect. Newer pills containing 20 μg of estrogen or formulations that gradually increase the estrogen dose are being marketed. Women of lower body weight, including adolescents, may be more likely to experience nausea. In addition, adolescents may be more likely to become anxious or to discontinue oral contraceptives if they experience side-effects, including nausea.19
In this weight-sensitive culture, many women are concerned with or have heard about the possibility of weight gain associated with oral contraceptives. Studies suggest that although individual women may gain weight while taking oral contraceptives, in the population as a whole, as many women lost weight as gained weight.59 One double-blind, placebo-controlled study showed that 20% to 40% of all women (including those on placebo) had changes in weight over four menstrual cycles.59 On a cycle-by-cycle analysis, as many or more women taking placebos as taking the oral contraceptives gained at least 2.25 kg (5 lb). Some individual women may be sensitive to the estrogen effect of water retention. Breast tenderness or slight weight gain when starting oral contraceptives may result from this fluid retention, and it may be one side-effect that contributes to dissatisfaction with oral contraceptives. The fact that it occurs infrequently should be conveyed to the patient prior to starting her on oral contraceptives to allay her concerns.
Adolescents are particularly concerned about weight gain. One study noted that 86% of suburban adolescents were worried about the possibility of weight gain related to oral contraceptives.15 In addition, even the perception of weight gain, which was not confirmed by actual weight measurements, was associated with lower rates of compliance and pill continuation.
Headaches are another side-effect of oral contraceptives perceived as occurring commonly. Headaches occur commonly regardless of oral contraceptive use; the Walnut Creek prospective study of side-effects showed no significant differences in headache rates between never users, past users, and current users of oral contraceptives.61 When a patient who is taking oral contraceptives complains of headache, it is important to distinguish the type of headache, which may by muscle contraction-related, vascular, or migraine. The first two of these are unrelated to the use of oral contraceptives. Migraine headaches may be related to oral contraceptive use, but there is very little data to document whether migraine headaches are more likely to increase in frequency, decrease, or remain the same with oral contraceptive use.60
Public perceptions of adverse effects of oral contraceptives are fueled by reports in the print and visual media. Bad news tends to be reported with more frequency and effort, and more attention is focused on reports of possible negative effects or oral contraceptives; thus, these risks become exaggerated in the minds of the general public. The Gallup poll conducted for ACOG found that nearly three fourths of women surveyed believed that taking the pill entailed substantial risks.67 Among those women who believe that there are substantial risks with oral contraceptive use, fear of cancer was reported by 31%, blood clots by 23%, weight gain by 12%, high blood pressure by 12%, and bleeding by 7%. Many adolescents consider birth control to mean primarily oral contraceptives, and one study reported that 40% of adolescents feared that contraception was dangerous.72
Unfounded concerns regarding the risks of oral contraceptives may be fed by negative media coverage and may result in declining confidence in oral contraceptives. This declining confidence, when coupled with womens' concerns and dissatisfaction due to the experience or fear of side-effects, may result in lack of pill compliance or discontinuation. The result of pill discontinuation is an increase in unintended pregnancies and pregnancy terminations. Thus, “minor” side-effects can have major consequences.
In addition, there are complex interrelations among side-effects, pill continuation, and correct pill use, although until recently there have been very few studies addressing this issue.73 Missed pills have been associated with, but do not invariably result in, the side-effect of breakthrough bleeding.74 The experience of breakthrough bleeding, whether because of missed pills or not, may lead to dissatisfaction and pill discontinuation. However, the experience of missed pills without breakthrough bleeding or pill failure (i.e. pregnancy) may inadvertently reinforce the idea that missed pills are not a problem and thus lead to more missed pills. The experience of nausea or vomiting may result in missed or incompletely absorbed pills, which may result in breakthrough bleeding.
Although data are available regarding oral contraceptive continuation, very little is known about the actual usage patterns of the method.75 Recent studies have addressed this issue in some detail, and questions about oral contraceptive pill compliance and missed pills were added to the 1995 NSFG cycle survey.1,41 In addition to questions regarding how oral contraceptive usage can best be measured, other questions have been raised which have not been addressed through careful study. Why do women miss pills? What are the demographic characteristics of women who are more likely to miss taking pills? What are the characteristics of women who are perfect pill users? Are some women at greater risk for contraceptive failure because of biologic factors related to steroid hormone metabolism? Which women effectively use a backup method of contraception?
It is apparent from an examination of the gap between the perfect use pregnancy rates of oral contraceptives (0.1%–1%)6 and the failure rates in typical users (1.9%–18.1%)76 that incorrect or imperfect use is a major contributor to the failure rate. When oral contraceptives are initially prescribed for a woman, she can make the decision not to have her prescription filled or not to begin taking the pill. The extent to which this type of noncompliance is a problem with oral contraceptives is not well established. It has been suggested that 20% to 30% of women who are prescribed hormone therapy for postmenopausal replacement do not have their prescriptions filled.77 One U.S. study has examined the microbehaviors of oral contraceptive pill use and noted that 4% of the women who were described as “pill acceptors” did not take the pills at all during the study interval.
Forgetting to take pills is probably the most common type of error in oral contraceptive pill use, although clinicians have also observed the many ways that pills can be taken incorrectly. One study from Glasgow reported that only 28% of patients were taking the pill perfectly, and 27% reported having missed at least one pill in the previous 3 months.70 A report from South Africa noted that about one third of adolescents studied had missed at least one pill in the previous 3 months, a percentage similar to that noted in Glasgow.78 Another small study from the United Kingdom reported that 46% had missed a pill during the previous 3 months.79 One study of women in Bangladesh included home visits and pill counts and found that the women had over- or under-taken an average of eight pills per month, with 87% missing at least one or more pills per cycle.80
Data from the United States are sparse, but the previously noted study of microbehaviors found that only 42% of the women took their pill every day.81 The same study found that 16% of women reported having a pill or pills remaining at the end of the pill packet or cycle. One clinical trial examining side-effects and the use of a new pill formulation noted that about 15% to 25% of women reported missing at least one pill per cycle.82 These subjects were college students who were personally motivated to prevent pregnancy and who had the extra encouragement and motivation provided by the study coordinators and nurses to comply with pill-taking behaviors. Different populations may be less motivated to take their oral contraceptive pills perfectly. One interesting study involved a comparison between self-reported missed pills and data from an electronic device measuring compliance.41 In 3 months of pill use, the self-report and the electronic report frequently did not agree, and as expected, the proportion of women who missed at least three pills according to the electronic data was triple that of the self-report. Fifty to sixty percent of women reported no missed pills, whereas the electronic data indicated that no pills were missed only 20% to 30% of the time, with the percentage of women missing no pills declining from the first to the third cycle.41 In addition, the electronic data indicated that 30% missed three or more pills in the first cycle; by the third cycle, this figure had increased to 50%. Not surprisingly, self-reported data from the NSFG indicated a lower percentage of women missing pills; 13% of all women reported missing two or more pills over 3 months, and 16% reported missing only one pill over that interval.1 However, of note, about 15% of pill users also use another contraceptive method, which may provide some back-up contraception.83 Characteristics of women who are more likely to use the pill inconsistently include Hispanic and non-Hispanic black women, those who have recently begun use, and those who have had a previous unintended pregnancy.83
In another report of women using electronic monitoring devices, 52% of women never missed a pill over the 3 months of the prospective study, and an additional 21% used a back-up method; the remaining 27% were likely at increased risk for pill failure.84
What do we really know about the consequences of missing a pill periodically? Patients may assume that missing a pill midcycle would be more likely to result in ovulation. However, it appears that the likelihood of follicular development, and thus theoretically the likelihood of ovulation and pregnancy, may be enhanced with prolongation of the pill-free interval (i.e. placebo interval) of 7 days in the typical pill packet.74 This could occur if a woman missed taking a pill or pills at the end of the cycle or did not restart her new packet of pills correctly, as might occur if she neglected to refill her prescription.85,86 Studies addressing this question have included only small numbers, but the link between ovulation and pregnancy as outcomes has not been shown.87 Killick and colleagues studied the effect of increasing the duration of the pill-free interval a varying number of days using hormonal assays and ultrasound scanning.88 Increases in the duration of the pill-free interval resulted in more follicular development and estradiol production, although ovulation was not noted to occur. Importantly, there was a great deal of variation among individuals in the group. It is theoretically possible that some women are particularly sensitive to the lack of suppressive doses of contraceptive steroids that occurs during the pill-free interval and thus might be at increased risk of pregnancy with missed pills; however, this has not been conclusively demonstrated. Goldzieher, in a review of the pharmacology of contraceptive steroids, noted large variations in serum levels from patient to patient within relatively homogeneous populations as well as differences between populations in different countries.89
Problems with the transition from one pill packet to the next have been noted by Potter and Williams-Deane.75 In some countries and with some pill packet configurations, patients are instructed to begin taking the first packet of pills on day five of their cycle. This instruction has led to confusion regarding the duration of a waiting period between pill packets, with some patients believing that they should always start the next pill packet on day 5 after the onset of withdrawal bleeding. Because there is some variation in the onset of withdrawal bleeding, some women may thus be starting the next cycle as late as 8–10 days after the previous packet. It has been argued that the 28-day pill packets are less likely to be associated with transition problems than are 21-day packets.78,90
One major component of compliance with oral contraceptive use is an understanding of the correct instructions for starting the pills and for what to do if pills are missed. The instructions for starting the first packet of pills vary by the type of pill (triphasic versus monophasic) and manufacturer. In the past, most pill packets used in the United States were designed with a Sunday start, whereas most pill packets in Europe are designed so that the patient takes her first pill on the first day of menses. More recently, some manufacturers have designed pill packs that can be used for either a Sunday start or an any day start. The advantages of the Sunday-start packet are related to the familiarity of some U.S. users and providers with this option, as well as the fact that clinicians may find it easier if all of their patients have a consistent pattern of use. One disadvantage of the Sunday start is that a pill user will complete her packet of pills on a weekend, at a time when it may be more difficult to contact the clinic or physician's office to get a refill.90 The first-day start has the advantage of reliably suppressing follicular development from the onset of pill taking, obviating the use of a back-up method of contraception. Although the use of a back-up method of contraception for a variable interval of time with the first cycle of use has been widely advocated, there is little data to indicate the extent of compliance with this advice.
Currently, many patients receive information about the initiation of pill use and what to do if pills are missed primarily from the patient package insert that must be included with each prescription for oral contraceptives. The FDA held a meeting of its Advisory Committee on Fertility and Maternal Health Drugs in 1991 that addressed the issue of labeling instructions regarding the use of oral contraceptives. The patient package insert has been found to contain insufficient and conflicting instructions that are difficult for the average patient to understand and may contribute to errors of compliance.91 The FDA Advisory Committee unanimously recommended to oral contraceptive manufacturers that they develop standardized and simplified instructions for oral contraceptive use. These instructions now include specific information regarding missed pills and starting days. It is hoped that the provision of sufficient, consistent, and easy-to-understand information results in improved compliance and thus fewer unintended pregnancies worldwide. In one study, women who did not read or understand the pill packet instructions were more likely to miss two or more pills per cycle.73
Cost and inconvenience of getting pill refills can also have an impact on compliance, and efforts to minimize this barrier may be helpful in improving compliance.92 Recent studies and reports have highlighted the likely relation between lack of HMO and private insurance coverage for oral contraceptives and unintended pregnancies.93,94 Efforts are currently underway in U.S. Congress to mandate contraceptive coverage by companies that cover other prescription medications.
STRATEGIES FOR ENCOURAGING ORAL CONTRACEPTIVE COMPLIANCE.
Strategies for encouraging oral contraceptive compliance include being accessible to patients who have concerns about or who fear side-effects.73,95 Those fears must be acknowledged and alleviated through the provision of accurate information about oral contraceptives. The potential benefits of pregnancy prevention and lowered risk of medical problems such as pelvic inflammatory disease, anemia, and endometrial and ovarian cancer should be noted and the favorable risk/benefit ratio discussed. Knowledge of noncontraceptive benefits of the pill has been associated with lower rates of missed pills.73 Patients should be cautioned about the effects of missed pills. They can be informed that missing a pill may be associated with breakthrough bleeding or spotting, although this is not invariably the case. The risks of escape follicular development associated with missed pills, particularly at the beginning or end of the cycle that would extend the pill-free interval, should be noted.74,88
The fact that nuisance side-effects may frequently result in dissatisfaction should be acknowledged to the patient at the onset of pill use but with the proviso that alternative formulations are an option if the side-effect is persistent. The patient should be cautioned not to stop taking the pills without first discussing the perceived problem with her physician.
A thorough program of patient education allows the patient to be aware of the possibility of side-effects. Frequent follow-up visits may also be helpful, particularly with adolescents. Poor compliance and pill dissatisfaction with resultant discontinuation can thus be minimized. The possibility of side-effects must be addressed, because no currently available oral contraceptive carries a guarantee of the of the absence of side-effects. Low-dose pill formulations that are associated with a low incidence of nuisance side-effects such as breakthrough bleeding, lack of withdrawal bleeding, nausea, headaches, and breast tenderness should be prescribed when possible.
Barrier Methods of Contraception
The failure rates with various barrier methods of contraception including spermicides relate, at least in part, to lack of use with each act of intercourse or to improper use. Recent arguments have been made that the use of two concurrent barrier methods of contraception, when used consistently and perfectly, results in higher efficacy and pregnancy prevention than the use of either method alone, as the assumptions for independence in consistent use are probably not valid.96 Dissatisfaction with barrier methods that results in discontinuation tends to be relatively high; in the national survey data, 30% of married women discontinued use of the diaphragm at the end of a year, 29% discontinued reliance on condoms, and 45% discontinued spermicide use.14
In a study that looked at the components of compliance with the use of condoms, Oakley and Bogue defined microbehaviors of condom use, which included putting the condom on prior to penetration, holding the condom in place during withdrawal, withdrawing while the penis is still erect, and using a condom during every act of intercourse.97 Depending on the age group, less than one-third to only slightly more than one half of users used a condom during every act of intercourse.98
Consumer awareness of condoms has increased markedly since the beginning of the human immunodeficiency virus (HIV) epidemic, and sales and use have also grown considerably.99,100 Among adolescents, the use of condoms has increased. Data from the NSFG indicate that 37% of adolescents report using condoms as their method of contraception, and condom use was reported by 54% of all women at the time of first intercourse.1 The availability of condoms in school-based clinics has been shown in two studies to result in increased use of condoms without increased rates of sexual activity.101,102 Recent surveys have begun to focus on the reality that women/couples are using dual methods, most commonly a hormonal method and condoms. Data from the 1995 NSFG found that 8% of all women using contraception were using the pill and condom.103 There is evidence that offering users of oral contraceptives information about a choice of barrier options (spermicidal film versus condoms) resulted in increased barrier use without decreasing condom use.104 There is, however, a complex interaction surrounding the use of two methods, and in one sample, about one fourth of dual users reported plans to reduce condom use in the future.105 Consumer perceptions of condoms have changed as a result of increased publicity and public comfort in talking about the method.
Recent focus on HIV prevention has led to the increased use of condoms. Given the potential risks and sequelae of HIV and other sexually transmitted diseases (STDs), it is no longer appropriate for clinicians to discuss only contraception without also addressing STD risk and the use of barrier methods.106 Latex condoms, when used consistently and correctly, are highly effective in reducing the risk of infection with HIV and other STDs.107 In one study in which heterosexual couples who were HIV serodiscordant were observed for seroconversion, none of those who used condoms consistently became HIV positive, whereas 4.8 persons per 100 person-years developed HIV when condoms were used inconsistently.108
Compliance with barrier methods of contraception (i.e. correct and consistent use) is determined by a complex interaction of characteristics of the specific method, characteristics of the user, and the specific situation.109 Different barrier methods may interfere with sexual spontaneity and enjoyment to varying degrees. Negotiation between the sexual partners regarding barrier method use is affected by the stage of the sexual relationship, adolescent development, comfort with sexuality and sexual communication, other characteristics of the relationship such as mutuality or collaboration, and whether there are elements of unequal relationships, coercion, force, or abuse.110 Previous contraceptive use and familiarity with the method are also factors.109
Although most barrier methods can be obtained without a prescription from a provider, clinicians still have an extremely important role in promoting effective and consistent use. The quality of the evidence that latex condoms are effective in reducing the risk of HIV transmission is good, whereas the evidence for other barrier methods is less firmly established. There is still an urgent need for the development of better barrier methods which meet consumer needs and are effective in both pregnancy prevention and reducing the risk of STD acquisition.
Contraceptive Methods That Are Less Compliance Dependent
INTRA-UTERINE DEVICE (IUD).
Certain methods of contraception are less dependent on the user for effectiveness than others. In recent years, there has been increasing focus and attention on these methods, as compliance has been recognized to be an issue in contraceptive effectiveness. The IUD, injectable methods, and implantable methods do not require ongoing motivation to the same extent as coitally related methods or oral contraceptives.
The IUD is used widely in China and Europe, although its use in the United States is low and has been declining.111,112 There have been recent attempts to revive the popularity of the IUD,113,114 although NSFG data from 1995 indicate use by less than 1% of U.S. women.1
The IUD is a model for a contraceptive method that requires little ongoing user compliance. The IUD does require correct use or periodic checking for correct placement to ensure that the device has not been expelled unnoticed.5 IUD users in clinical trials are typically examined more frequently by a clinician than are typical users; thus, expulsions would be noted more consistently, resulting in somewhat lower failure rates than among typical users. However, the gap between the lowest reported failure rates and failure rates among typical users is smaller than with other methods such as oral contraceptives.
DEPOT MEDROXYPROGESERONE ACETATE.
One long-acting injectable progestin, available as depot medroxyprogesterone acetate (DMPA; Depo Provera), was approved for contraceptive use in the United States in 1992 and has been shown to be highly effective.115 The most commonly used regimen involves an injection every 3 months, and is touted as “the birth control you only have to think about four times a year.” Compliance with an every-3-months regimen may have its own difficulties, and early studies of efficacy reported relatively high rates of subjects who were lost to follow-up.5 The percentage of individuals who were lost to follow-up represents one measure of compliance and patient satisfaction. In one recent U.S. study, only 57% of women returned for a second shot, and 63% of those who received a second injection returned for their third, with a 1-year continuation rate of only 23%.116
The use of DMPA is associated with side-effects which include an almost invariable change in menstrual cycles, mild weight gain, and mood changes.91 As with oral contraceptives, these factors may influence acceptability or continuation. Prior to its approval for general use, DMPA had been used in the United States for individuals with special needs in relation to contraception, such as drug abusers or women with developmental delays; one of the clear benefits of the method is the relatively infrequent compliance requirements.91
Contraceptive implants, including levonorgestrel-containing subdermal devices (Norplant) which became available for use in the United States in 1991, are highly effective at preventing pregnancy.117 If the user is able to tolerate side-effects such as irregular bleeding, the continued use of the subdermal implant containing levonorgestrel requires no ongoing compliance for 5 years after the initial insertion. Continuation rates with this method are typically 70% to 80% or higher.31,118,119 User error is not a factor in effectiveness. The fact that the removal of the devices requires a minor surgical procedure helps to encourage continued use, although the coercive potential has been a focus of public and scholarly discussion.120 Discontinuation is thus an active event rather than a passive one of nonuse. Side-effects such as bleeding irregularities are typically the most common reason given for discontinuation.121,122 Adolescents and adults have been found to have comparable continuation rates and tolerance of side-effects.123,124 Studies comparing Norplant and oral contraceptive use in adolescents have found significantly higher continuation rates and thus lower pregnancy rates among Norplant users.31,125,126
Other injectable or implantable devices may become available over the next decade.127 A single-capsule, biodegradable, subdermal device that releases levonorgestrel over 12 to 18 months is undergoing trials that will reveal its efficacy and acceptability.128 A two-rod levonorgestrel implant has undergone extensive testing and has been shown to be effective with comparable rates of side-effects to the six-rod implant, but it has not been marketed due to concerns regarding liability and the litigious legal climate in the United States.129 Monthly injectable methods have been tested, and cycle control has been shown to vary among different formulations.130
It has long been recognized that contraceptive emergencies can occur. A discussion of patient education and compliance would be incomplete without a discussion of this option, which recognizes the problems of compliance with other contraceptive methods and begs for greater public awareness. Hatcher and colleagues, in the classic reference, Contraceptive Technology, have stated,
It has been calculated that the use of emergency contraception could potentially result in the prevention of 2.3 million pregnancies per year and 1 million abortions in the United States.131
A variety of methods of postcoital contraception, sometimes called “morning-after pills,” “emergency contraceptive pills,” or more recently “emergency contraception,” have been studied. The most commonly prescribed method in the United States is the Yuzpe method, which involves the use of high-dose combined oral contraceptive pills.132 Other regimens have been described, including the use of RU-486 and other antiprogestins, which have been reported to be associated with fewer side-effects than the Yuzpe regimen.
Emergency contraception has been called the “best-kept contraceptive secret”133 and it has also been stated that it is a method about which “women know little, and men know less.”134 Recent efforts from ACOG and the American Medical Association (AMA) have sought to publicize the method's availability to the public and increase knowledge and prescription rates among clinicians.135,136 In a study reported in 1994, few primary care physicians or reproductive health care providers had ever discussed emergency contraception with patients or had literature available.137 Expanding the role of nurses, nurse practitioners, and nurse-midwives may be one method of increasing awareness and prescription of this method to all women seeking contraceptive options.138,139 Options for making the method more freely available though over-the-counter or direct-from-pharmacist availability or by routine prescription for all patients have been debated and studied; these would likely expand the opportunities for primary pregnancy prevention.140,141,142,143,144,145,146
Some populations have unique needs in terms of contraception and special problems related to compliance with contraceptive use. Adolescents may represent the largest group of special-needs patients. Women at risk for fetal anomalies are another such group, as are women with medical problems that would make pregnancy a high-risk situation.
Adolescents are less likely than older women to use contraceptive methods consistently and correctly. Relatively few adolescents report the use of any method of contraception at first intercourse, and many delay for at least 6 months after coitarche.11 Factors that have been cited as affecting compliance in adolescents include the discrepancy between emotional and sexual development; the reluctance to acknowledge sexual activity to peers or parents, thus precluding reinforcement by these groups; the typical adolescent sense of invulnerability and belief that pregnancy is unlikely (“It can't happen to me”); the nature of sexual activity, which may be infrequent, unplanned, and unpredictable; the fact that many disadvantaged teens lack goals for the future which provide motivation for postponing sexual activity; a lack of understanding of menstrual physiology and fertility; and previous experience with contraceptives.11
The most popular method of contraception among adolescent young women is oral contraception.21 Condoms are the next most popular method. Misinformation regarding oral contraceptives is widespread among young women and their mothers. The mothers of today's adolescents may themselves have taken high-dose oral contraceptives and have concerns about the risks of both serious and nuisance side-effects which do not apply to the low-dose pills that are currently available. In addition, they were often exposed to the negative media influence during the 1970s.43 Providing accurate, current information to adolescents and their mothers (if the daughter has actively involved her mother in the decision-making process) is helpful in dispelling myths which may contribute to fear, dissatisfaction, and discontinuation.
Adolescents have a number of concerns and fears about the nuisance side-effects that may occur with oral contraceptive use. Almost invariably, they have heard about the possibility of weight gain, and may be concerned about interaction of the pill with cigarettes, possible fertility problems, or even birth defects that they fear may result from oral contraceptive use.15 The experience of side-effects or even the perception of weight gain are associated with lower rates of pill continuation.15 Because one typical pattern of sexual activity during adolescence is serial monogamy, adolescents may adopt an on-again, off-again use of oral contraceptives which may leave them vulnerable to gaps in contraceptive coverage if they discontinue oral contraceptive use when they break up with a boyfriend but fail to reestablish use when they are once again sexually active in a new relationship. This sporadic use may also contribute to a longer total duration of nuisance side-effects, which tend to decrease in frequency over time but which may recur with each new restart of oral contraceptive use.95
Adolescents require ongoing reinforcement and encouragement to use contraceptive methods effectively. Frequent follow-up visits are important, as is an emphasis on the availability of advice by telephone should questions arise. Adolescents should be specifically counseled not to discontinue their method of contraception without discussing the problem with their clinician.
Women at High Risk for Fetal Anomalies
Although the list of known and well-established teratogens is relatively small, certain drugs such as lithium carbonate or isotretinoin (Accutane) are used among women of childbearing age. Lithium carbonate carries a risk of congenital malformations. Isotretinoin, which is used to treat severe cystic acne, is associated with a rate of fetal anomalies approaching 25% for pregnancies that proceed beyond 20 weeks.147 The manufacturer of Accutane, together with the FDA, implemented a multicomponent pregnancy prevention program in 1988. Follow-up study of this program found that the pregnancy rate among users of Accutane was substantially lower than that in the general population.148 The choice or recommendation of a method of contraception for these women, and for others who are taking medications which may carry some risk of fetal anomalies, must take into consideration the failure rate of the method in typical users and the likelihood of compliance in the individual patient.
In addition, Grimes has argued that the strategy for contraceptive counseling for women with a high risk of fetal anomalies must include knowledge of the woman's attitudes toward pregnancy termination.149 If a woman is willing to consider pregnancy termination if she conceives while taking a medication associated with a high risk of fetal anomalies, a method of contraception which carries a higher failure rate in typical users and which may be more subject to compliance failures may be acceptable. Alternatively, if the woman is unwilling to consider pregnancy termination, only a method which is highly effective should be offered. The problems of poor compliance should be considered, and a method that is less compliance dependent (e.g. an injectable or implantable method) or nonreversible sterilization may be most appropriate, as suggested by the Teratology Society.150
Women With Medical Problems
Women with medical problems such as hypertension or diabetes mellitus represent another large group of women for whom pregnancy would entail potential risks to their own health or the health of their fetus. The risks and benefits of effective contraceptive methods must be compared with the risks and benefits of pregnancy, whether it is terminated or allowed to proceed to term. Pregnancy entails a risk of morbidity and mortality that is dependent on age, the outcome of pregnancy, and any underlying medical conditions. Older women have a greater risk of death associated with pregnancies which end in birth than do younger women, at least partially because they are at increased risk of having underlying medical problems.
Harlap and colleagues compared the risks of pregnancy-related death among women using no method of contraception and the risk of death among those using various methods of contraception, considering the failure rates of these methods.21 They also compared pregnancy-related mortality among women using no method of contraception and the mortality associated with various methods of contraception, factoring in the effectiveness of each method in preventing pregnancy as well as the potential risks of mortality from the method itself.21 All methods of contraception are safer than using no method and having a birth.
For an individual woman, especially one with chronic medical problems, it is impossible to accurately predict the magnitude of risk of mortality which pregnancy would entail, although the risks associated with pregnancy for various populations give some information which may be useful in decision-making. Clearly the more severe the underlying disease, the greater the potential risks associated with pregnancy. The benefits of using a method of contraception that is highly effective may outweigh the potential risks associated with its use if the consequences of an unintended pregnancy would be severe. For example, a woman with diabetes mellitus who is poorly compliant with her insulin regimen as well as with the use of barrier contraceptives may benefit from a highly effective method of contraception which is less compliance-dependent such as the implantable subdermal levonorgestrel system. For this woman, uncontrolled diabetes is associated with fetal and maternal risks; the potential for an unintended pregnancy is high if barrier methods are not used perfectly, and the benefits of the very effective subdermal system appear to outweigh the potential risks.
Thus, the contraindications for each method of contraception should be viewed relative to the risks of pregnancy, and a list of “relative contraindications” to the method should be obtained,151 (although the current phrase is “disadvantages and cautions”).6 The patient and her willingness to comply with a given method of contraception must be considered on an individual basis, but compliance with the method as well as its effectiveness are particularly important considerations for women who would face risks if they were to have an unintended pregnancy. Goldzieher has reviewed the options of hormonal contraception for women with various underlying medical problems, taking into account the current literature and an assessment of the risks versus the benefits of these methods.152
In 1994 and 1995, the World Health Organization convened two scientific expert group meetings to review the data from clinical and epidemiologic research on medical criteria for contraception prescribing and recommend medical eligibility criteria for the various contraceptive methods.153 Four categories were described for classification, ranging from category 1, a condition for which there is no restriction for the use of the contraceptive method, to category 4, a condition which represents an unacceptable health risk if the contraceptive method is used. This represents an attempt to encourage the use of an evidence-based approach developed with the medical judgments and opinions of experts in the field on contraception.
A major premise among clinicians and researchers in the area of contraceptive compliance is that adequate knowledge is not the sole requirement for correct use but it is essential.154,155 Compliance is clearly an essential component of contraceptive efficacy. Efforts aimed at increasing a patient's knowledge and understanding of a prescribed therapy are a necessary first step toward compliance.156 The provision of clear and simple instructions, given both orally and in written form, results in greater information retention in the general medical setting. Forgetting to take pills is probably the most common type of error in oral contraceptive pill use, although clinicians have also observed the many ways that pills can be taken incorrectly.156 Attention to readability of patient education materials is critical in the provision of written materials, because large segments of the U.S. population (approaching 20%) have been noted to be functionally illiterate (i.e. not able to read materials written at a fourth to fifth grade level).157
The clinician-patient interaction is an important component of reproductive counseling and has been described as the main determinant of the accuracy and completeness of patient data, diagnostic accuracy, efficiency in the encounter, compliance, patient understanding of problems, and patient and physician satisfaction.158 However, little attention has been given to the teaching or practice of this critical skill. Involvement of the patient herself in decision-making is also essential.159
History-taking should include questions that help to establish a patient's background knowledge and health beliefs regarding methods of contraception. Frequently, myths and misinformation will surface which may erode confidence in the method if not dispelled. These specific misperceptions can be addressed by the clinician in an effort to increase the likelihood of compliance. Clinicians can make efforts to become better sources of information to their patients; the use of ancillary health personnel within an office or clinic setting may be of benefit.156 Data suggest that counseling can improve contraceptive use.66,160,161,162
Physicians must be sensitive to the fears and concerns about various methods of contraception which the patient may express, because these fears are not always rational or easily dispelled by information alone and may result in problems with compliance. Data from the ACOG Gallup poll highlights the extent to which misinformation is prevalent among the general public.67
Clinicians should be realistic with patients about the possibility of nuisance side-effects, which may have an impact on method satisfaction and continuation.36 An explanation of types of side-effects that may be seen with a given method of contraception, the likely course of the side-effects (i.e. that they may decrease over time), and reassurance about their meaning are helpful. The physician may state, “If you have problems with your method of birth control, do not stop using it without discussing it with me, unless you are planning a pregnancy.” The problem of a gap in effective contraception that may be associated with method switching can be addressed.
The patient must be informed of the efficacy of each method of contraception, including both estimates of the lowest expected failure rate and the failure rate in typical users. She should be informed that her own efforts in good compliance should result in a failure rate that is lower than the rate in typical users. She should be encouraged to thoughtfully assess her own likelihood of compliance, given past experiences with contraception and considering her partner's concerns and wishes. Active involvement and support from a partner may positively affect compliance. Most patients are unaware of the potential noncontraceptive benefits of contraceptive methods.163 The benefits of oral contraceptives are seldom a focus of the media, whereas the potential risks are emphasized.
Some patients realize that they have had considerable difficulty with compliance in the past. The clinician can encourage these women to choose a method of contraception that is less compliance dependent. The results of noncompliance (i.e. the risks of unintended pregnancy) should be pointed out. The clinician should form an opinion of the individual patient's likelihood of compliance with a method of contraception and consider how realistic he or she judges the woman's self-assessment to be.156 The physician can review past compliance with antibiotic prescriptions, return appointments, and past unintended pregnancies. Although the best compliance will occur with a method that the individual woman has freely chosen, feels comfortable and is satisfied with, and is easy to use and produces few side-effects, compliance is a complicated issue.164 Ambivalent feelings about pregnancy may further complicate compliance.
There is evidence from the compliance literature on general medical care that monitoring compliance through follow-up appointments is of benefit in improving compliance. Haynes has described three principles which underlie programs that have been found to be effective in improving compliance: reducing barriers to compliance, increasing attention to and supervision of noncompliant patients, and reinforcing and rewarding compliance at every patient contact.165 Asking the patient directly and specifically if she consistently uses her method of contraception may not invariably identify the individual who has poor compliance, but the individual who does admit to poor compliance is clearly at risk. Recognition and praise for effective compliance are suggested mechanisms for rewarding compliance with medical therapies that are not commonly employed in thinking about contraception. Perhaps if physicians acknowledge the difficulties involved in effective use of contraception, a more effective therapeutic alliance between patient and physician could be formed. Studies have suggested that patients who are satisfied with their physician are more likely to comply.156 Effective communication is an essential component of patient satisfaction.
The clinician must be aware of special populations in whom issues of compliance are particularly important: adolescents, women with medical problems, or women at high risk for fetal anomalies. Special attention to the provision of highly effective methods of contraception and a focus on the hazards of noncompliance may begin to address the problems of these groups.
Improving contraceptive efficacy may require new developments in pharmacology and technology. Methods of contraception which are easier to use and require less ongoing compliance, possibly utilizing alternative delivery systems, will be helpful. A better understanding of the correlates of individual variations in hormonal metabolism may help pinpoint the individual at high risk of method failure if compliance is not perfect. Studies of the psychology of contraception and the behaviors involved may provide further information regarding the nature of noncompliance. Technological developments such as inexpensive predictors of ovulation166 may provide further information that will aid in effective contraception and compliance.
Recognition of the importance of compliance in effective contraception is newly dawning. An increasing body of research is available to address concerns about behavioral and biologic differences among women that result in contraceptive noncompliance and unintended pregnancies. Efforts aimed at increasing understanding of compliance and noncompliance with contraceptive methods will undoubtedly help to narrow the gap between the lowest reported failure rates and the failure rates of contraceptive measures during typical use.
1. Abma J, Chandra A, Mosher W, Peterson L, Piccinino L: Fertility, family planning, and women's health: New data from the 1995 National Survey of Family Growth. National Survey of Family Growth, National Center for Health Statistics. Vital Health Stat 23: 1– 125, 1997
31. Polaneczky M, Slap G, Forke C, Rappaport A, Sondheimer S: The use of levonorgestrel implants (Norplant) for contraception in adolescent mothers. N Engl J Med 331: 1201– 1206, 1994
40. Salway S, Fauveau V, Chakrabarty J: Introducing the low-dose pill to Bangladesh; issues of continuation and failure [published erratum in Contraception 50(4):397, 1994]. Contraception 49: 171– 183, 1994
46. Lewis MA, Spitzer WO, Heinemann LA, MacRae KD, Bruppacher R, Thorogood M: Third generation oral contraceptives and risk of myocardial infarction: An international case-control study. Transnational Research Group on Oral Contraceptives and the Health of Young Women. BMJ 312: 88– 90, 1996
49. Spitzer WO, Lewis MA, Heinemann LA, Thorogood M, MacRae KD: Third generation oral contraceptives and risk of venous thromboembolic disorders: An international case-control study. Transnational Research Group on Oral Contraceptives. BMJ 312: 83– 88, 1996
54. Dusterberg B, Ellman H, Muller U, Rowe E, Muhe B: A three-year clinical investigation into efficacy, cycle control and tolerability of a new low-dose monophasic oral contraceptive containing gestodene. Gynecol Endocrinol 10: 33– 39, 1996
55. Endrikat J, Jaques MA, Mayerhofer M, Pelissier C, Muller U, Dusterberg B: A twelve-month comparative clinical investigation of two low-dose oral contraceptives containing 20 micrograms ethinylestradiol/75 micrograms gestodene and 20 micrograms ethinylestradiol/150 micrograms desogestrel, with respect to efficacy, cycle control and tolerance. Contraception 52: 229– 235, 1995
56. Archer DF, Maheux R, DelConte A, O'Brien FB: A new low-dose monophasic combination oral contraceptive (Alesse) with levonorgestrel 100 micrograms and ethinyl estradiol 20 micrograms. North American Levonorgestrel Study Group (NALSG). Contraception 55: 139– 144, 1997
65. DelConte A, Loffer F, Grubb HS: Cycle control with oral contraceptives containing 20 micrograms of ethinyl estradiol. Contraception 59: 187– 193, 1999
69. Schilling LH, Bolding OT, Chenault CB et al: Evaluation of the clinical performance of three triphasic oral contraceptives: A multicenter, randomized comparative trial. Am J Obstet Gynecol 160: 1264– 1268, 1989
102. Guttmacher S, Lieberman L, Ward D, Freudenberg N, Radosh A, Des JD: Condom availability in New York City public high schools: Relationships to condom use and sexual behavior. Am J Public Health 87: 1427– 1433, 1997
103. Abma J, Chandra A, Mosher W, Peterson L, Piccinino L: Fertility, family planning, and women's health: New data from the 1995 National Survey of Family Growth. National Survey of Family Growth, National Center for Health Statistics. Vital Health Stat 23: 1– 125, 1997
122. Harel Z, Biro FM, Kollar LM, Rauh JL: Adolescents' reasons for and experience after discontinuation of the long-acting contraceptives Depo-Provera and Norplant. J Adolesc Health 19 (2): 118– 233, 1996
129. Sivin I, Viegas O, Campodonico I et al: Clinical performance of a new two-rod levonorgestrel contraceptive implant: A three-year randomized study with Norplant implants as controls. Contraception 55: 73– 80, 1997
130. Sang GW, Shao QX, Ge RS et al: A multicentred phase III comparative clinical trial of Mesigyna, Cyclofem and Injectable No. 1 given monthly by intramuscular injection to Chinese women. I. Contraceptive efficacy and side effects. Contraception 51: 167– 183, 1995
132. Trussell J, Ellertson C, Stewart F: The effectiveness of the Yuzpe regimen of emergency contraception [published erratum in Fam Plann Perspect 29(2):60, 1997]. Fam Plann Perspect 28: 58– 64, 1996
148. Mitchell AA, Van Bennekom CM, Louik C: A pregnancy-prevention program in women of childbearing age receiving isotretinoin. N Engl J Med 333: 101– 106, 1995
150. The Public Affairs Committee, The Council of the Teratology Society: Recommendations for isotretinoin use in women of childbearing potential. Teratology 44:1, 1991
160. Nelson AL: Counseling issues and management of side effects for women using depot medroxyprogesterone acetate contraception. J Reprod Med 41: 391– 400, 1996
161. Levine AS, Holmes MM, Haseldon C, Butler W, Tsai C: Subdermal contraceptive implant (Norplant) continuation rates among adolescents and adults in a family planning clinic. J Pediatr Adolesc Gynecol 9: 67– 70, 1996
162. Davie JE, Walling MR, Mansour DJA, Bromham D, Kishen M, Fowler P: Impact of patient counseling on acceptance of the levonorgestrel implant contraceptive in the United Kingdom. Clin Ther 18: 150– 159, 1996