Timing is crucial: Or is it? Avoid being too dogmatic

By Robert A. Hatcher, MD, MPH

Chairman and Senior Author

Contraceptive Technology

Professor of Gynecology and Obstetrics

Emory University School of Medicine

Atlanta

Couples considering contraceptives and their health care providers face myriad questions about the timing of contraceptive use. Some-times our clients come to us with mistaken ideas. As often as not, we providers are the source of misinformation. In either case, timing errors, misconceptions, and oversimplifications can cause trouble. Trouble in family planning often can be spelled unintended pregnancy. The following are legitimate questions. The answers to these questions, however, have caused problems. Sometimes the proper answer is, "I don’t know."

Postpartum women

1. When are women in your program asked to return for their postpartum visit?

Answers to this question vary from two to six weeks. More clinicians are moving toward the two- to four-week postpartum checkup.

2. If a woman is breast-feeding her baby, is immediately postpartum, and wants to receive her first shot of Depo-Provera before leaving the hospital, is this appropriate? Is it ever appropriate? Should it be the standard of care?

It is appropriate for some women. Postpartum return rates are notoriously low in some settings. If you’re certain a woman won’t be returning, providing her first depo-medroxyprogesterone (DMPA) injection just before leaving the hospital poses no serious problem. This is done fairly routinely at Grady Memorial Hospital in Atlanta.

Case one: A 19-year-old woman is denied DMPA in the hospital after the delivery of her second child. She does not return on the appointed date for her six-week postpartum exam because she learns that she will not see the same nurse midwife who delivered her baby. She is given an appointment several weeks later to see the nurse midwife with whom she had developed a good relationship. She was pregnant when she returned for her postpartum visit, so it turned out to be her first antepartum visit instead. After delivering her third child at age 20, she again planned to nurse. The policy still prohibited giving Depo-Provera to a lactating woman. The nurse midwife found a way to get around this policy and gave the woman her initial injection before she left the hospital.

Case two: A 24-year-old woman returns at three weeks postpartum wanting Depo-Provera but already has had intercourse 15 times. Her pregnancy test is negative. She is given condoms and told to return in a week. By this time she has had intercourse six more times. She used a condom each time except for the last, which was the night before this five-week postpartum visit. Her pregnancy test again was negative. She was given emergency contraceptive pills, her first injection of Depo-Provera, and told to use condoms for another week. She probably would have been better served by receiving her first DMPA injection postpartum before leaving the hospital.

3. On average, when does ovulation occur in a postpartum nonlactating woman?

Forty-five days postpartum, although few first ovulations are followed by a normal luteal phase.1

4. Can ovulation occur in a breast-feeding mother who has not resumed menstruation?

Yes! The probability that ovulation will precede menses is:

• 33 to 45% during the first three months;

• 64 to 75% during months four through 12;

• 87 to 100% after 12 months.2,3

Ovulation returns even more quickly in women who are not nursing.

5. For how long do American women breast-feed their babies?

In the United States, only one of two infants is breast-fed. The mean (average) time to complete weaning is only 23 weeks (just over five months). The median time to complete weaning is only 13 weeks (three months).4

6. When do American couples resume sexual intercourse?

Most American couples resume sexual intercourse within several weeks of delivery. Among lactating women, 66% are sexually active in the first month postpartum, and 88% are sexually active in the second month postpartum.5

7. When should women who are not lactating be encouraged to start taking combined oral contraceptives?

Clearly there is no consensus on this important question. Traditionally, most clinicians in the United States have recommended some delay in initiating pills. Each organization has a slightly different approach. One of the most conservative patterns has been to have a woman return for a six-week postpartum visit, prescribe or provide pills at this time, and tell her to start pills after her next period. Her next period may be weeks later, and since ovulation may precede her first period, she is at risk of unintended pregnancy unless she uses another contraceptive before starting pills.

If clinicians wait until they see a woman in the postpartum period before prescribing or providing pills, the postpartum visit cannot be delayed until six weeks. For a number of reasons, it makes good sense to schedule women for a postpartum visit at two to four weeks rather than at six weeks. But even if women are seen at two to four weeks, the problem may not be solved, because there may be a significant delay until that first period arrives. It is a calculated risk to delay starting pills until after a woman’s first postpartum menstrual period.

Clearly, the least-confusing approach is most likely to prevent an unintended pregnancy. That approach is to provide pills after delivery and recommend that the postpartum mother who is not breast-feeding start taking them the day she returns home. The advantages of starting the pills quickly are that when a woman is seen right after delivery, clinicians can be certain she isn’t pregnant, hasn’t ovulated, and is highly motivated to avoid pregnancy. Instructions on how and when to take the first pill should be simple and unambiguous.

Keeping instructions unambiguous is extremely important. Providing contraceptive supplies or actually initiating contraception (inserting an IUD or Norplant implants, giving a woman her first Depo-Provera injection, or performing tubal sterilization, for example) before a woman leaves the hospital is becoming an important goal of contraceptive programs.

The concern with starting pills immediately after delivery has been that women are at an increased risk for thrombophlebitis for the first week to 10 days postpartum. How are we to deal with this information and provide pills in a manner easy for women to understand?

Data are lacking

Certainly one of our goals in providing contraceptives is to keep women out of harm’s way — to protect their health and well-being — by analyzing the data on published risks of various interventions. Unfortunately, to my knowledge, there are no epidemiologic data demonstrating that postpartum women using combined pills are at increased risk for thrombophlebitis compared to postpartum women not using pills.

So the clinician has little information to go on in assessing the benefits and risks of starting pills immediately after delivery. Always, the specific details of an individual woman’s history must be taken into consideration. Permitting a woman to decide for herself what contraceptive to use and when to start using it often is the best course of action rather than an ironclad policy dictated by the clinician or agency developing its protocol.

Since ovulation usually does not occur before four weeks postpartum after a delivery following a gestation of 28 or more weeks, it has been our policy for a number of years at Grady to give women several cycles of birth control pills while in the hospital. We ask them to wait until the Sunday after their babies are three weeks old to take the first pill.

In this hospital-based program, women are given two cycles of pills before they leave the hospital, so they can begin pills just before they return for their postpartum visit, which we now schedule at four weeks following delivery. This means that women are taking the first pill after the period of increased risk for thrombophlebitis following pregnancy.

Each woman starts her pills each month on Sunday, as do our other patients (although we realize that this also is an arbitrary practice). The program also prevents the legal implications of a postpartum thromboembolic event being blamed on pills (whether or not pills were responsible).

It makes sense to prescribe or actually provide pills while a woman is in the hospital and tell her exactly when to begin taking them. Definitely avoid instructing women to wait until after their first menstrual bleeding to start taking pills. As to the exact number of days postpartum to start pills, it seems that no matter when you choose to start pills, there will be advantages and disadvantages.

8. When should breast-feeding women be encouraged to start taking combined oral contraceptives?

No controversy: Start pills when baby is completely weaned from breast-feeding.

Minimal controversy: Start pills as weaning begins.

Some controversy: Start pills when woman begins to supplement breastmilk with other foods.

Extensive controversy: Start pills in woman who will be fully or almost fully breast-feeding (minimal or no supplementation of breastmilk with other foods) once she has established a good flow of breastmilk. (The World Health Organization in Geneva, Switzerland, strongly discourages this.)

Combined oral contraceptives

1. After how long should a woman on combined oral contraceptive pills be encouraged to take a break from them?

This practice is discouraged whether a woman has been on pills a year, 10 years, or 20 years. A woman who has been taking pills for 10 years has been off of them for one of each four weeks the whole time. So in 10 years she has been off pills for 21¼2 years.

2. Is it important for women to take combined oral contraceptives at the same time each day?

It is quite important from a compliance standpoint. However, it is not very important in terms of effectiveness and certainly makes minimal difference in terms of spotting if it is just a matter of one to several hours.

Emergency contraception

1. When should emergency contraceptive pills (ECPs) be taken?

Soon after unprotected sex. The usual interval is 72 hours.

2. When can they be taken?

Failure rates for the first, second, and third 24 hours are the same. ECPs may work for four to five days.

3. Can they be taken later than 72 hours after unprotected intercourse?

It’s best not to be too dogmatic about the 72-hour cutoff. A woman can be several hours late.

4. What should a woman who has missed two pills do?

A woman may be a candidate for ECPs if she has missed two pills.

Progestin-only pills (mini-pills)

1. What is the best time of day to take progestin-only pills (POPs)?

A major effect of mini-pills is to cause a thickening of the cervical mucus. This effect diminishes toward the end of the 24 hours before the next pill is taken. John Guillebaud, MA, FRCSE, FRCOG, medical director of the Margaret Pyke Family Planning Centre in London, calls the mini-pill the "tea-time pill" so it is taken each day in the late afternoon and causes a thick cervical mucus by the time a woman is likely to have sex. In our country, it is often recommended that mini-pills be taken with the evening meal.

2. How important is it to take progestin-only pills at the same time each day?

Taking mini-pills on time is much more important than it is for combined pills.

References

1. Gray RH, Campbell OM, Zacur H, et al. Postpartum return of ovarian activity in nonbreast-feeding women monitored by urinary assays. J Clin Endocrinol Metab 1987; 64:645-650.

2. Campbell OMR, Gray RH. Characteristics and determinants of postpartum ovarian function in women in the United States. Am J Obstet Gynecol 1993; 169:55-60.

3. Lewis PR, Brown JB, Renfree MB, et al. The resumption of ovulation and menstruation in a well-nourished population of women breast-feeding for an extended period of time. Fertil Steril 1991; 55:529-536.

4. Visness CM, Kennedy KI, Gross BA, et al. Fertility of fully breast-feeding women in the early postpartum period. Obstet Gynecol 1997; 89:164-167.

5. Ford K, Labbok M. Contraceptive usage during lactation in the United States: An update. Am J Public Health 1987; 77:79-81.