A 40–Year-Old Woman Considering Contraception

DR DELBANCO: Mrs B is a 40-year-old, married nurse who has 3 children. She and her husband have decided not to have more children and are considering how to prevent pregnancy in the future. Her insurance is through a managed care organization.

Mrs B experienced menarche at age 12 years. She is generally healthy, does not smoke, and has been pregnant 5 times. She has 3 healthy children and had a spontaneous miscarriage, in addition to a voluntary abortion when she was younger. Mrs B has used several methods of contraception. During college she tried an intrauterine device (IUD), but had difficulty and had it removed. Subsequently, she used a diaphragm until her mid-30s. Only 1 of her children was planned, and she feels that the other pregnancies may have resulted from imperfect use of the diaphragm. After the birth of her youngest child, she began to use both a diaphragm and condoms.

Mrs B developed severe emotional lability premenstrually about 2 years ago. Eighteen months ago, she began using oral contraceptives to alleviate her difficulties, resulting in fewer symptoms and less menstrual flow, though she developed intermittent headaches. Sometimes these headaches seemed migraine-like, and she found that ibuprofen used premenstrually generally prevented them. When Mrs B forgets to take the pill on occasion, she and her husband use condoms because she wants to be "150% sure" of avoiding future pregnancy. Her menses on the pill are regular.

Her past medical history includes mild asthma responsive to bronchodilators, intermittent low back pain responsive to muscle relaxants, and occasional chest pain related primarily to exercise. Exercise testing in the past indicated that she does not have arteriosclerotic heart disease. She uses support stockings for varicose veins, which she expects to have treated surgically in the future. She had a benign fibroid adenoma removed from a breast about 8 years ago.

The findings from the physical examination, including vital signs, are normal. The eye grounds are normal. Total cholesterol level was 3.47 mmol/L (134 mg/dL) in December 1997. The results of her mammography in early 1998 and her recent Papanicolaou test were normal. Current medications include Loestrin 21 1.5/30 (norethindrone acetate, 1.5 mg, and ethinyl estradiol, 30 mg; Parke-Davis, Morris Plains, NJ) daily, albuterol inhalers, as needed, and cyclobenzaprine hydrochloride, 10 mg as needed.

Mrs B and her husband are considering changing their method of contraception because of concerns about effectiveness.

MRS B: After my last child was born 4 years ago, we added a condom with the diaphragm just to be extra sure, because we knew that we did not want any more children, and we were nervous about the failure rate of the diaphragm alone. When we added the condom, I found that my husband did not like using the condom and that both of us were using something that we didn't really like. At that point we didn't even think of an alternative. I never wanted to use the pill. I'd never been on the pill. I strongly objected to it. I did not like the thought of altering my hormones. I felt that was very unnatural because you had something in you when you weren't having intercourse, and it was with you all the time. However, I started on the pill when I saw my primary care physician in June of 1996 with symptoms of severe premenstrual syndrome, which I felt had been increasing over the past year or two. I felt that the symptoms were getting out of control. It was affecting my work life and my home life. I found, after a few months, that taking the pill improved my quality of life dramatically. My periods were fantastic for the first time. I wish I had discovered it when I was 13. I was 39 years old, and I was having the best period I had ever had in my life. I had short periods with hardly any bleeding, no cramping, and they were completely predictable. So I enjoyed being on the pill. The added benefit was that our sex life was great. It was spontaneous. We didn't worry. He didn't have to use a condom, I didn't have to use a diaphragm, so it was very liberating.

I think sterilization is a really hard choice. You have to know that you absolutely don't want any more children. When I turned 40 it became easier to think about it. In my 30s, my obstretician/gynecologist had mentioned it a few times. He asked, "Is your husband ready for his vasectomy yet?" My doctor didn't want to see me have a tubal sterilization and felt that it was my husband's responsibility to have a vasectomy. Actually, I think my husband has been ready and I've been the one holding back. It's hard to make a permanent decision.

DR Z: The main issue for Mrs B is the efficacy of the method. It has become increasingly important to her as she has grown older not to have any more children. She has also had a number of gynecologic issues related to her method of contraception that have influenced her choice. For instance, she was using a diaphragm for a long time and was finding that she had some intense back pain that began on about day 7 of her cycle and continued, predictably, for about a week each month. It was troublesome to her because it would wake her up at night, and she found it predictably worse if she used her diaphragm.

Mrs B had another problem that has made her choice complicated. On the pill, she had some premenstrual headaches that worsened, probably migraine headaches. About a year ago, I thought about suggesting that she consider another method of contraception, but she really wanted to continue with it. The benefits of the pill were far outweighing the problems with the headaches.

What are Dr Peterson's thoughts about the use of oral contraceptives in women older than 40 years and up to menopause? Over time, I am hearing more questions from patients about whether it is an acceptable method of contraception to use almost up to menopause.

What is the epidemiology of unwanted pregnancies in older women? What options are available for this patient? What are the risks, benefits, and costs associated with each method? What are the issues with respect to her partner? How do you counsel women and their partners about their choices? What are the risks of unintended pregnancy for a woman of this age? What are the risks and benefits of using oral contraceptives for women of this age? What do you recommend for Mrs B?

DR PETERSON: Many women are in Mrs B's position. The highest number of births in US history occurred during the post–World War II baby boom from 1947 to 1965. Thus, more women than ever are in the later childbearing years. Mrs B is concerned about the possibility of pregnancy. Although she is in the age range where fecundity is, on average, rapidly declining,¹ she still must consider herself at risk for unintended pregnancy until menopause—defined as the permanent cessation of menses, which occurs in the United States at an average age of 48 to 52 years. The ability to achieve pregnancy in the later reproductive years is evident from the number of unintended pregnancies that occur among women after age 35 years.

Many pregnancies among women in their 30s and 40s are unintended. In 1994, an estimated 482400 pregnancies occurred in US women aged 35 to 39 years, and 98300 pregnancies in women aged 40 to 44 years.² Of these, the proportion of unintended pregnancies was 40.8% among women aged 35 to 39 years, and 50.7% in women aged 40 to 44 years.² The rates of unintended pregnancies among women aged 35 to 39 and 40 to 44 years in 1994 were 17.8 and 5.0 per 1000 women, respectively.² More than half of unintended pregnancies among women aged 35 to 39 years and aged 40 years and older ended by induced abortion in 1994 (56.3% and 64.7%, respectively).²

Most US women are currently using some form of contraception. The proportion of US women aged 15 to 44 years using contraception increased from 55.7% in 1982, to 60.3% in 1988, and to 64.2% in 1995.³ In 1995, 73.1% of women aged 35 to 39 years and 71.4% of women aged 40 to 44 years were using contraception.

Mrs B has recently used, or is considering using, those methods of contraception most popular among women aged 35 years and older—surgical sterilization, combined estrogen-progestin oral contraceptives, and condoms. In 1995, among women aged 35 to 39 years, 41% of women using contraception had undergone tubal sterilization, 19% had partners who had undergone vasectomy, 17% were using condoms, 11% were using oral contraceptives, 3% were using diaphragms, 1% were using IUDs, fewer than 1% were using contraceptive implants or injectable contraceptives, and about 7% were using other methods. Among women aged 40 to 44 years, 50% of women using contraception had undergone tubal sterilization, 20% had partners who had undergone vasectomy, 12% were using condoms, 6% were using oral contraceptives, 3% were using diaphragms, 1% were using IUDs, fewer than 1% were using contraceptive implants or injectable contraceptives, and about 7% were using other methods.³

The health benefits of contraceptive use for older women are both contraceptive, ie, related to the health benefits of avoiding unintended pregnancy, and noncontraceptive. For most women in the later childbearing years, like Mrs B, the combined contraceptive and noncontraceptive benefits clearly exceed the health risks for all methods of contraception.⁴ Harlap et al⁴ have estimated that the annual risk of death associated with pregnancy and childbirth for women aged 35 to 39 years using no method of contraception is 11.7 per 100000 women. The comparable risk for women aged 40 to 44 years is 20.6 per 100000 women. Both estimates far exceed the risks for women in these age groups who use any method of contraception, with the noteworthy exception of women older than age 35 years who smoke and use oral contraceptives. For example, the risk of death among oral contraceptive users aged 35 to 39 years is 1 per 100000 women; the comparable risk for women aged 40 to 44 years is 1.9 per 100000 women. Although the annual risk of death associated with using no method of contraception far exceeds that for use of any method among all age groups, the health benefit of contraceptive use is greatest for women in the 35- to 39-year-old and 40- to 44-year-old age groups because the risk of mortality associated with pregnancy is greatest among these age groups.

Contraceptive effectiveness has influenced the choice of contraceptive methods for Mrs B, as it will for most women. Trussell⁵ has estimated the effectiveness of each contraceptive method according to characteristics of the users. Table 1 provides separate failure rate estimates for use among typical couples and for use among couples who use the method perfectly (both consistently and correctly). However, the characteristics of the user influence the effectiveness of some methods more than others. For example, tubal sterilization and IUDs are highly effective regardless of user characteristics, but the effectiveness of oral contraceptives and barrier methods differs strikingly for typical as opposed to "perfect" users. To anticipate the likelihood of successful use, women considering oral contraceptives or barrier methods need to take into account which category of user they will likely be. Estimates of contraceptive failure rates are based largely on studies of younger women. In general, because of the impact of age on fecundity, the failure rates for older women will be lower than those listed in Table 1. Thus, for a woman in Mrs B's age group, many methods are highly effective, when used consistently and correctly.

Trussell et al⁶ have estimated the costs of contraceptive use, considering method used, side effects, and unintended pregnancies. All 15 methods evaluated were more cost-effective than the use of no method. The Copper T 380A IUD, vasectomy, levonorgestrel implant, and injection of depot medroxyprogesterone acetate were the most cost-effective methods over 5 years of use (Table 2).

Over the last several decades, the United States has witnessed a dramatic increase in the number of women and men undergoing sterilization. Both tubal sterilization and vasectomy are convenient for those willing to undergo surgery because a single act results in highly effective and lasting protection. Although findings from the US Collaborative Review of Sterilization have indicated that pregnancy after tubal sterilization is more common than previously believed, the 10-year cumulative probability of pregnancy for women aged 34 to 44 years at the time of surgery was low (from 0.2% to 1.9% depending on the method).⁷ The rates of vasectomy failure are extremely low, regardless of a man's age.⁸

Surgical sterilization has relatively few noncontraceptive benefits, with none documented for vasectomy. Tubal sterilization appears to reduce the risk of pelvic inflammatory disease among women at risk for sexually transmitted infections, and may reduce the risk of ovarian cancer,⁹ although the mechanism for this protective effect against ovarian cancer remains unclear.

Mrs B and her husband are considering vasectomy. The major potential disadvantage of both vasectomy and tubal sterilization is that they must be considered permanent. Although both vasectomy and tubal sterilization are sometimes reversible, depending on patient and surgical characteristics, reversal procedures are costly and of uncertain success. The major known health risks of both vasectomy and tubal sterilization are the one-time risks of the surgical procedure, which are low. The estimated case-fatality rate from tubal sterilization is 1 to 2 per 100000 procedures.¹⁰ Major morbidity from tubal sterilization is uncommon, with most studies reporting rates of 1% to 2% or less.¹¹ Mr and Mrs B should note that vasectomy is even safer.¹² Minor complications after vasectomy are frequent, however, with scrotal swelling, ecchymosis, or pain occurring in up to 50% of men. These symptoms usually resolve spontaneously within a week or two. Hematoma formation generally occurs in less than 1% of vasectomies, and infections generally occur in less than 2%.¹²

No long-term major adverse health consequences of either tubal sterilization or vasectomy have been identified. Good evidence indicates that tubal sterilization is associated with little, if any, change in menstruation within the first 2 years after the procedure. Thereafter, evidence is limited and inconsistent.¹³ On balance, little evidence exists for a posttubal sterilization syndrome. Several studies suggest that sterilized women may be more likely than nonsterilized women to undergo subsequent hysterectomy.¹⁴ Any association is likely attributable to nonbiological factors, such as alterations in the threshold of decision making regarding hysterectomy; little evidence exists for a biological relationship. Although regret after tubal sterilization is common among women sterilized at a young age,¹⁵ regret appears uncommon among women sterilized at 35 years and older. Although tubal sterilization dramatically reduces the risk of all pregnancies (including ectopic pregnancies), ectopic pregnancies can occur remote from sterilization; such pregnancies were observed in the 10th year after all 4 methods of laparoscopic sterilization in the US Collaborative Review of Sterilization.¹⁶

Questions about the relationship between vasectomy and the risk of coronary heart disease appeared in 1974 following a report that monkeys were at increased risk for atherosclerosis after vasectomy. At least 9 subsequent human studies found no such increased risk, and the investigators reporting the initial finding in monkeys were unable to reproduce their results.⁸ Thus, good evidence indicates that vasectomy does not influence the risk of cardiovascular disease. The question for the 1990s regarding vasectomy is whether men like Mrs B's husband are at increased risk for prostate cancer. Several well-publicized studies^{17 - 18} reported an increased risk, but several subsequent studies found no such increase in risk.^{19 - 21} Because the associations in the positive studies were relatively weak, chance or bias could easily have explained the study findings.²² Biological arguments for a potential harmful effect are no more plausible than ones for a protective effect.²² A 1995 editorial from the National Cancer Institute concluded that "vasectomy appears either not to cause prostate cancer or to have only a relatively weak relationship to the disease."²³

Mrs B currently takes oral contraceptives but has expressed concerns about their effectiveness because she occasionally misses a pill. Although correct use of oral contraceptives is nearly 100% effective, an estimated 16% of oral contraceptive users are inconsistent in their pill taking.²⁴ Thus, typical pregnancy rates are approximately 5% per year of use. For some women, fears—mostly unfounded—of adverse health effects result in avoiding or discontinuing use of oral contraceptives. Most US women of reproductive age have used oral contraceptives. Memories of adverse effects in the past—particularly bleeding problems, including spotting, breakthrough bleeding, and amenorrhea—may cause some to avoid oral contraceptives in subsequent years. However, many women with menstrual abnormalities attributable to aging will have markedly improved menstrual bleeding patterns while using oral contraceptives, just as Mrs B reports. Other adverse effects that older women experienced with past oral contraceptive use, such as nausea, headaches, breast tenderness, or weight gain, may or may not occur with later use. For example, most evidence suggests that current low-dose oral contraceptives have little, if any, impact on a woman's weight.²⁵

Combined oral contraceptives have numerous important noncontraceptive benefits. Mrs B experienced the well-documented beneficial effects on the menstrual cycle, including a reduction in dysmenorrhea. This benefit may be particularly important for women experiencing menstrual abnormalities, eg, cycle irregularity associated with aging. Many find that oral contraceptives control menstrual abnormalities until menopause. Oral contraceptive use should be discontinued after menopause, because even low-dose oral contraceptives have 4 to 7 times the dose of estrogen typically used for hormone replacement.

Two important and well-documented benefits of oral contraceptive use are approximately 2-fold reduced risks of endometrial cancer (the most common gynecologic cancer in US women) and ovarian cancer (the most lethal gynecologic cancer in US women).⁴ The protective effect against both increases with longer duration of oral contraceptive use and persists for at least 15 to 20 years after cessation of use.^{4 ,26 - 27} Some evidence suggests that oral contraceptive use decreases perimenopausal bone loss²⁸ and may also reduce the risk of endometriosis.²⁹ Whether oral contraceptives reduce the risk of uterine leiomyomas is unclear. Although higher-dose oral contraceptives reduce the risks of functional ovarian cysts, lower-dose pills are unlikely to provide any substantial protection.³⁰

Mrs B has asked whether she can safely continue taking oral contraceptives until menopause. Questions regarding the safety of oral contraceptive use among older women have focused on the risks of cardiovascular disease, breast cancer, and cervical cancer. Women who have taken oral contraceptives in the past experience no adverse impact on the risk of myocardial infarction, stroke, or venous thromboembolism regardless of duration of use.³¹ Studies of low-dose oral contraceptives, in use today, likewise find no effect of past use, and are generally reassuring regarding current use.^{31 - 34}

Current use increases the risk of venous thromboembolism. Although low-dose oral contraceptives carry less risk than the higher-dose preparations used in the past, they nonetheless increase the risk, but to levels less than that attributable to pregnancy.³¹ The incidence of venous thromboembolism is estimated to be 10 to 21 per 100000 woman-years for oral contraceptive users, and 4 per 100000 woman-years for nonusers; 60 per 100000 pregnant women are estimated to develop venous thromboembolism.³¹ Fortunately, the case-fatality rate for venous thromboembolism is only 1% to 2%. Women who cannot take combined oral contraceptives because of susceptibility to venous thromboembolism may consider progestin-only contraceptives, including progestin-only pills.

In contrast to venous thromboembolism, the baseline incidence of myocardial infarction increases sharply with age. Although studies of low-dose oral contraceptives and the risk of myocardial infarction and stroke among older women are still limited, available evidence suggests little, if any, increase in risk among healthy, nonsmoking women.^{32 - 34} The excess risk of myocardial infarction attributable to oral contraceptive use among such women is estimated to be less than 1 case per 100000 woman-years.³³ The annual excess risk of death from cardiovascular disease (venous thromboembolism, myocardial infarction, and stroke) attributable to oral contraceptive use among healthy, nonsmoking women aged 40 to 44 years has been estimated at 2 to 3 deaths per 100000 women.³⁵ By contrast, women older than 35 years who smoke are at substantially increased risk for myocardial infarction, because oral contraceptive use and smoking act synergistically to increase this risk.³² The annual excess risk of death from cardiovascular disease among women aged 40 to 44 years who smoke is estimated at 6 to 12 deaths per 100000 women.³⁵ Thus, women choosing oral contraceptives after age 35 years should stop smoking. Those who will not stop smoking should choose another method of contraception.

Mrs B also asked if taking oral contraceptives increased her risk of breast cancer. A recent pooled analysis of data on breast cancer and oral contraceptive use has clarified this relationship.³⁶ Pooled data from 54 epidemiologic studies provide information on 53297 women with breast cancer and 100239 controls. Four key study findings are broadly reassuring: (1) No increased risk of breast cancer was associated with past (10 or more years after cessation of use) use of oral contraceptives, regardless of the duration of use. (2) Findings were similar for women with and without a family history of breast cancer. (3) Current or recent (up to 10 years after stopping use) oral contraceptive users were at a slightly increased risk for breast cancer diagnosis. The excess in risk was for tumors less advanced clinically than those diagnosed among women who never used oral contraceptives. Clinically advanced disease was less common among current or recent oral contraceptive users. (4) The increase in risk for current or recent users was the same, regardless of duration of use.

Taken together, these findings indicate that oral contraceptives do not initiate new tumors. They are consistent with earlier detection of existing tumors, late-stage promotion of tumors, or some combination of the two. The use of oral contraceptives for less than 1 year conferred the same excess risk as 15 years of use and the excess in risk was associated with tumors localized to the breast arguing strongly that at least some of the increased risk is related to earlier detection of existing tumors. Finding less clinically advanced disease among oral contraceptive users is consistent with oral contraceptive users having existing tumors detected earlier than nonusers, and with oral contraceptive use having a beneficial effect on the rate of growth of tumors or their tendency to metastasize. The findings are also compatible with the possibility that oral contraceptive use and pregnancy have similar effects on breast cancer risk. Long-term survival data for oral contraceptive users and nonusers related to risk of breast cancer are not yet available.

The relationship between oral contraceptive use and cervical cancer remains controversial. Whether the observed increase in risk seen in several studies with long duration (≥5 years) of oral contraceptive use was caused by oral contraceptive use is unclear.³⁷ The biologic plausibility for a causal relationship is greater for adenocarcinoma, an uncommon type of cervical cancer,³⁸ than for squamous carcinoma.

Mrs B had headaches prior to starting oral contraceptive use. There are 2 major types of headaches, nonvascular, which includes tension headaches, and vascular, which includes common (without neurologic symptoms) migraine and classic (with neurologic symptoms) migraine headaches. Most headaches in women are nonvascular, and, in general, oral contraceptive use does not influence the prevalence and severity of nonvascular headaches. Most women with common migraine headaches are appropriate candidates for a trial of oral contraceptive use. Such use may increase, decrease, or have no effect on the frequency or severity of headaches.³⁹ Although data regarding the risk of stroke among women with migraine headaches who use low-dose oral contraceptives are limited and inconsistent,²⁹ theoretical concerns regarding women with classic migraine headaches lead most to conclude that such women should avoid oral contraceptives altogether.^{29 ,39} Women who develop focal neurologic symptoms or worsening of their headaches and women with the new onset of headaches while using oral contraceptives should discontinue use, at least temporarily, and undergo evaluation.

Relatively few women in the United States use IUDs, but users express a high degree of satisfaction.⁴⁰ Infectious morbidity attributable to the Dalkon Shield—a device removed from the US market in 1974—has inappropriately created a black cloud over IUDs in the United States. Women at low risk for sexually transmitted diseases (STDs) are at low risk for pelvic infection while using currently available IUDs.^{41 - 42} Most IUD-attributable infections are related to insertion of the device and occur within 3 weeks of insertion.⁴² The most commonly used IUD in the United States today—the Copper T 380A—is approved for up to 10 years of use and has long-term effectiveness similar to that of tubal sterilization.⁴¹ The rate of pelvic inflammatory disease associated with use of this device in a World Health Organization study was 0.59 per 1000 woman-years of use.⁴² Some older women at low risk for STDs may, nevertheless, be inappropriate candidates for an IUD. In particular, as women age, they are more likely to have abnormalities of uterine size or shape attributable to leiomyomas, which may be incompatible with effective IUD use. Nonetheless, many older women are excellent candidates for IUD use.

Mrs B did not include either depot medroxyprogesterone acetate or levonorgestrel implants on her list of potential choices. Among older women, both methods have effectiveness comparable to tubal sterilization, although both are associated with menstrual abnormalities that have limited their use. Some women with menstrual abnormalities may choose to use depot medroxyprogesterone acetate, in part because about 50% of women using depot medroxyprogesterone acetate for 1 year report amenorrhea. Many women may find amenorrhea due to depot medroxyprogesterone acetate an attractive alternative to their menstrual abnormalities occurring without hormonal contraception. Both depot medroxyprogesterone acetate and levonorgestrel implants provide long-acting protection. Progestin-only pills are also available but are used in the United States primarily as an alternative to combined oral contraceptives by women who are breast-feeding.

Although Mrs B and her husband have decided firmly against having further children, some older women may be trying to time their next pregnancy. Such women should be aware that the return to fertility is prompt after discontinuation of nearly all contraceptive methods. The single important exception is use of depot medroxyprogesterone acetate—for which the return to fertility can be delayed substantially.⁴³ Thus, many older women who are delaying pregnancy will want to avoid use of depot medroxyprogesterone acetate.

Emergency contraception available in the United States includes combined oral contraceptives, progestin-only pills (minipills), and the Copper T 380A IUD.⁴⁴ Of these, combined oral contraceptives are used most commonly and are the method determined by the Food and Drug Administration to be safe and effective for emergency contraception.⁴⁵ Although the safety of emergency contraceptive use has not been studied directly in older women, data on oral contraceptives argue for the safety of this approach. Pregnant women should not use emergency contraceptive pills; however, studies of oral contraceptives argue against an adverse effect on an established pregnancy.⁴⁵ When treatment with emergency contraceptive pills begins within 72 hours after unprotected intercourse, the risk of pregnancy is reduced by at least 74%.⁴⁶ About 50% of women using emergency contraceptive pills have nausea and about 20% vomit. Use of antinausea medication 30 minutes to 1 hour before each of the 2 doses of emergency contraceptive pills may reduce nausea.⁴⁵ Although emergency contraceptive pill use is cost-effective for preventing unintended pregnancy,⁴⁴ it is not as effective as consistent use of other contraceptives. Thus, it should be considered truly an emergency measure.

Mrs B is in a mutually monogamous relationship and not at risk for STDs. Many women older than age 35 years, however, find themselves at risk for both pregnancy and STDs. Consistent and correct use of male latex condoms is highly effective for preventing both human immunodeficiency virus (HIV) infection,⁴⁷ other STDs, and pregnancy. However, inconsistent use is not a reliable strategy for preventing either pregnancy or infection. Some women at risk for both pregnancy and STDs will find use of a dual-method approach (eg, sterilization plus condoms) acceptable; others may prefer the single method of using latex condoms consistently and correctly. Some evidence suggests that condom use for disease prevention is lower among women using methods other than condoms for pregnancy prevention.⁴⁸

Although no clinical studies are available to evaluate the effectiveness of female condoms for HIV prevention, the consistent and correct use of female condoms is expected to substantially reduce the risk of STDs. Although diaphragm use may reduce the risk of some STDs, diaphragms should not be assumed to protect against HIV infection.⁴⁹

Mrs B, like most other healthy, nonsmoking women her age, has several highly effective and safe contraceptive options. She and her husband are interested in vasectomy, which is an appropriate choice for many couples who are certain that they want no more children and are willing to accept a one-time, low risk of surgical complications for lasting, high-level protection against pregnancy. Mrs B is also enthusiastic about the effect that oral contraceptive use has on her menses. As long as she remains healthy and does not smoke, she is at low risk for using oral contraceptives until menopause. However, if her migraine headaches worsen, she should discontinue oral contraceptive use and undergo evaluation.

AN OBSTETRICIAN/GYNECOLOGIST: What do you think about venous thromboembolism and the third-generation progestins?

DR PETERSON: The issue is controversial, with much of the debate centered around whether bias likely explains the 1.5-fold to 2-fold increased risk of venous thromboembolism reported in some studies of third-generation pills. At a meeting of the World Health Organization in November 1997, experts concluded that preparations containing desogestrel and gestodene (which is not available in the United States) probably carry a small excess risk of venous thromboembolism relative to pills containing levonorgestrel. If that is the case, the excess risk translates into about 1 to 2 cases per 10000 pill users annually.³⁵ Users of any oral contraceptive have a low risk of venous thromboembolism; that risk is greater than the risk for nonusers, but substantially less than the risk during pregnancy.

AN OBSTETRICIAN/GYNECOLOGIST: This particular patient said she didn't like the IUD because she had problems with it when she was in college. Surely there's a difference between putting an IUD in now, compared with 20 years ago. In my experience, if you explain the IUD to patients, particularly if the patient is multiparous and in a mutually monogamous relationship, the IUD is an effective and good form of birth control. A lot of women actually choose it once they understand it.

DR PETERSON: I agree with you. The problems Mrs B had early on could well have been attributable to her being nulliparous at the time, and, as you suggest, the modern IUDs are clearly different from those used decades ago. For multiparous women, surveys suggest a high level of user satisfaction with the currently available IUDs, which are popular in Europe and in many other countries. Europeans also have the levonorgestrel-releasing IUD, which is associated with fewer menstrual abnormalities. The Copper T 380A, by far the most commonly used IUD in the United States, has an expulsion rate of about 5% to 6% in the first year and about 1% to 2% thereafter. The rate of removal of the Copper T 380A for pain or bleeding is about 12% in the first year and about 3% thereafter.⁵⁰ The progesterone-releasing IUD available in the United States is associated with fewer menstrual abnormalities than the Copper T 380A but has the important disadvantage of needing to be replaced annually.

AN OBSTETRICIAN/GYNECOLOGIST: As far as IUDs are concerned, my patients are worried less about pelvic infections than about heavier menstrual flow and cramps.

DR PETERSON: Women should be counseled about possible increases in pain or bleeding prior to IUD insertion. Some women are clearly put off by that potential. Women who experience problems may have relief with prostaglandin synthetase inhibitors. Those women who don't have sufficient relief are not good candidates for continued use.

AN OBSTETRICIAN/GYNECOLOGIST: It would have been interesting to have Mrs B's husband interviewed. What about the acceptability of vasectomy to men? I find with my patient population that the women may think that is a good option, but their partners don't always agree.

DR PETERSON: Clearly it is a couple's decision.

DR PARKER: In the postmenopausal population, estrogen with progesterone has been shown to be useful in reducing coronary disease. Why was there speculation that this combination would be deleterious in the premenopausal woman?

DR PETERSON: The beneficial effects of estrogens for older women include direct effects on coronary artery endothelium, changes in lipoprotein metabolism, and effects on fibrinogen. Concerns for premenopausal women are related to estrogen dose and a dose-dependent effect of estrogens on the risk of thrombosis. The dose of estrogens in oral contraceptives is much higher than that in pills used for hormone replacement. For example, the lowest-dose 20-µg combined oral contraceptive pill is about 4 times the dose in 0.625 mg of conjugated equine estrogens for hormone replacement. The increase in risk of thrombosis with increasing dose of estrogen is well established. Until recently, many believed that the dose of estrogens used for hormone replacement was too low to increase risk. Recent studies^{51 - 52} call that belief into question. Nevertheless, it is important to make the transition to lower doses of estrogen as soon as the transition is warranted.

AN OBSTETRICIAN/GYNECOLOGIST: For someone who chooses pills in her 40s, how do you recommend accomplishing the transition from pills to hormone replacement therapy?

DR PETERSON: The goal is to ensure that contraceptive doses are provided for as long as needed but not beyond, so the key is determining when menopause has occurred. There are different strategies. One common approach is to check the follicle-stimulating hormone (FSH) level on day 7 of the pill-free week. But FSH levels fluctuate, so a single level is not 100% reliable. Some evidence also suggests that FSH levels may be a less sensitive indicator of menopause than either the FSH and luteinizing hormone ratio (which is >1 with menopause) or estradiol levels obtained on day 7 of the pill-free week.⁵³ When to begin testing for menopause is another issue. Because most women experience menopause at age 48 to 52 years, some clinicians will continue women on oral contraceptives to about age 50 years, and then make that first assessment. If there is no indication of menopause, oral contraceptives could be continued with periodic reassessment.

DR DELBANCO: Why have doctors been resistant to emergency contraception? Should physicians routinely offer emergency contraception to their patients?

DR PETERSON: Emergency contraception has been around for a long time but used infrequently. I think there will be more stimulus for its use now with the Food and Drug Administration approval in February 1997 of emergency contraceptive pills, with the publicity, with an emergency contraceptive internet site (http://opr.princeton.edu/ec/), and with the toll-free hotline (1-888-NOT-2-LATE) to find the clinician in your area who will prescribe emergency contraceptive pills. Physicians have to examine the use on their own to make decisions about their prescribing practices, but all should be aware that the Food and Drug Administration has determined that emergency contraceptive pills are safe and effective.45

AN OBSTETRICIAN/GYNECOLOGIST: I recently heard about the possibility of over-the-counter emergency contraceptives, such as 4 oral contraceptive pills being available that way. It seems to me that the health risk would be modest. Has there been any more talk about that? Do you think there would be any health disadvantages to that?

DR PETERSON: I don't know where it is in terms of the likely availability. In Washington state, emergency contraceptive pills are available directly from pharmacists through pharmacist prescriptive practice agreements with physicians. One related issue is whether the use of emergency contraceptive pills is contraindicated among women with absolute contraindications to oral contraceptive pills. Some believe that there should be no contraindications to emergency contraceptive pills because of the short duration of use, but that's only opinion at this point—we have no studies to directly address this issue. In the United Kingdom, where emergency contraceptive pills are available by physician prescription, few adverse events associated with emergency contraceptive pill use have been reported to the Committee on Safety of Medicine.⁵⁴

AN OBSTETRICIAN/GYNECOLOGIST: It seemed that Mrs B could have used any type of contraception that she chose. In my practice I usually ask patients, "What do you want to try first?"

DR PETERSON: There are a variety of good approaches to helping our patients make their best choices about contraception. Studies of the safety and effectiveness of contraceptive methods are extensive, but there is a gap between available science and what our patients understand. We can close that gap.