The US Food and Drug Administration (FDA) approved a drug called Seasonale® on September 5, 2003. Media coverage before approval was already significant, and is expected to increase over the next few months. Seasonale® is a repackaging of oral contraceptive pills in a new schedule: 84 days of active pills and a seven day pill-free interval. Women who take this extended schedule of oral contraceptives will have fewer pill-free intervals. In this case, four times per year, the intended effect being four periods per year. At this year’s meeting of the Society for Menstrual Cycle Research in Pittsburgh, PA, Dr. Jerilynn C. Prior and I participated in a symposium entitled “The menstrual suppression controversy: medical, psychological & sociocultural perspectives”. I presented a review of the published medical data on the use of extended oral contraceptive pills to suppress menstruation, and this article is a summary of what we learned in that review. The context of our analysis is that there has been considerable press coverage of menstrual suppression, and that it has been predominantly positive, with suggestions that menstrual suppression with an extended schedule of oral contraceptives (“Long OC”) is a safe, healthy, reversible choice that empowers women and gives them control over their bodies. We wanted to examine what data were available to support those claims of effectiveness, safety and reversibility.
Limitations of the studies
Our literature search found 10 published papers with data on Long OC use. To evaluate the effects of a treatment, researchers use a comparison group, and the best way to make a comparison is to assign the research participants to the groups by chance (to randomize them). Only 2 of the 10 studies used this randomized design. The most common research focus was on the effectiveness of Long OC at controlling bleeding.
None of the studies were blinded, that is, both investigators and the participating women in the studies knew who was taking Long OC and who was taking Standard OC. This introduces the possibility of bias (conscious, or unconscious) in the data. None of the studies included a group of women with normal, unmedicated menstrual cycles. Finally, most women in the studies were women who had already been using oral contraceptives. Because of this, women who do not tolerate oral contraceptives well are less likely to be in the studies. What this means is that the data on Long OC were mainly with a population of women who had already tried regular OC, and who knew what the researchers expected, and which study group they were in.
Extended use of oral contraceptives and bleeding
Dr. Ann-Christine Cachrimanidou provides the best data in a one year study performed at three clinics in Sweden (Cachrimanidou et al., 1993). She randomized 284 women to take either Long OC on a 63/7 schedule (n=198) or Standard OC (21/7) (n=96). Two-thirds of the women in each group were taking oral contraceptive pills immediately before they began the study, about 15% had never taken oral contraceptives, and the remaining women were past users. Cachrimanidou asked women to record their bleeding patterns on a monthly calendar, indicating days with spotting (requiring no more than one pad or tampon) and bleeding. She further categorized the data according to whether it was a pill-free day, or a day with an active pill. Menstrual bleeding on a pill-free day is called withdrawal bleeding, and is expected, or “scheduled” bleeding. Menstrual bleeding on a day with active pills is called “breakthrough bleeding” or “unscheduled” bleeding. Cachrimanidou found that the amount of scheduled withdrawal bleeding was less in the Long OC users than in the Standard OC users. Long OC users had slightly more days of bleeding per withdrawal bleed, but had only one pill-free interval during the same time that the Standard OC users had three pill-free intervals.
However, that was not the whole story. Women on Long OC reported significantly more days of unscheduled bleeding and spotting, particularly during the first 63 days of active pills. Women who were not already taking oral contraceptives and started Long OC had the most problems with unscheduled bleeding and spotting. On average, women reported unscheduled bleeding or spotting on 14.9 out of a possible 63 days, that is, almost a quarter of the time. Women who switched from a standard schedule of oral contraceptives (“Standard OC”) to the Long OC schedule had spotting on an average of 5.2 days, and 1.9 days of breakthrough bleeding. The amount of unscheduled bleeding for women switching from Standard OC to Long OC was similar to that for women who were newly starting a Standard OC schedule. Unscheduled bleeding is the most common reason that women left the studies early. The amount of unscheduled bleeding drops off over time. However, there were still more breakthrough bleeding days on Long OC than Standard OC at the end of the one-year study.
A second randomized comparison of Long OC (this time on a 42/7 schedule) with Standard OC (21/7) was conducted by Dr. Leslie Miller in Seattle, WA (Miller & Notter, 2001). This too was a one year trial, with 90 women randomized equally to the Long OC and Standard OC arms of the study. Unfortunately, Dr. Miller did not distinguish between scheduled and unscheduled bleeding in her analysis. The number of spotting days was not significantly different between the Long OC and Standard OC groups. However, there were fewer bleeding days in the Long OC group than in the Standard OC group. The Standard OC group had twice as many pill-free days as the Long OC group, but statistically had no fewer days of spotting, and only 60% (rather than the expected 100%) more days of bleeding during the year. Although they are lumped together in this analysis, women experience scheduled and unscheduled bleeding as different events.
These two studies show that Long OC regimens do reduce the number of days of bleeding and spotting relative to a Standard OC schedule. However, the decrease in days of withdrawal bleeding because of fewer pill-free days is accompanied by an increase in days of unscheduled bleeding and spotting. This effect is particularly strong for women who are not current Standard OC users, and during the first few cycles of Long OC. Poor control of bleeding is the major reason that women stopped using Long OC.
Side effects and symptoms
These studies asked women about side effects and symptoms, and, if women left the study early, they were asked their reasons for doing so. The common reasons for discontinuing studies of Long OC include headaches, breast tenderness, mood changes, “premenstrual syndrome”, and weight gain. These are also reasons that women on Standard OC discontinued the studies. Cachrimanidou found that women on Long OC were more likely to leave the study because of problems with bleeding (13.1% versus 2.1%), while more women on Standard OC left because of headaches (9.4% versus 1.5%). While most women reported no change in most symptoms, both Miller and Cachrimanidou found that women on Standard OC had more days with headaches than women on Long OC.
When talking about Long OC and symptoms, there are two additional studies that I should explain, because they are often quoted inappropriately. These studies come from the clinical practice of Dr. Patricia Sulak, an obstetrician-gynecologist who uses oral contraceptives as a therapy for women with difficult menstruation. For many years, she has encouraged women to extend the period of active pills beyond the standard 21 days. She recommends that women add three weeks at a time to their schedule of active pills, and systematically increases or decreases the number of active pills, depending on the reaction of the woman to the increased schedule. The first paper specifically included only women who found relief on oral contraceptives, and whose symptoms intensified during the pill-free window (Sulak, Cressman, Waldrop, Holleman, & Kuehl, 1997). Forty-eight percent of the women had menstrual migraines as their worst complaint, while 78% had painful periods, 76% had menstrual migraines, 36% had heavy flow and 32% are described as having “PMS-type” symptoms. Of these 50 symptomatic women, 37 chose to stay at a Long OC schedule for at least three cycles, and 13 chose to revert to a Standard OC schedule, eight for reasons of breakthrough bleeding or spotting, and six because of headaches. Some women (14/27 continuing on Long OC) report no side effects or problems, while some of her patients make an understandable choice to put up with some spotting and breakthrough bleeding in order to reduce other symptoms, such as menstrual migraines.
In a second paper (Sulak, Kuehl, Ortiz, & Shull, 2002), Sulak reports on 292 patients who were offered the opportunity to use a Long OC schedule. Again, these are patients from her clinical practice, many of whom are referred to her for menstrual problems. She found that 9% of women chose not to try Long OC, 19% tried Long OC and then stopped oral contraceptive use altogether, 13% tried Long OC but switched back to Standard OC, and 59% continued on a Long OC schedule. The average Long OC schedule of women in her practice is 12 weeks of active pills.
While these studies illustrate that some women find relief from symptoms with Long OC, they are often used to suggest that all women would benefit from Long OC. Sulak’s population is quite different from those of the other studies. First, they are women who have chosen to see her with menstrual problems. Second, many of them are using oral contraceptives as a medication and do not desire birth control. Almost all other studies are based in family planning clinics, and draw their participants from women who request oral contraceptives for birth control. Third, they are older women than the women in other studies. Fourth, and perhaps most importantly, they have a powerful motivation to put up with side effects in order to avoid significant symptoms that are more common for them during the pill-free interval.
Long OC and safety
In our review, we were also concerned about what research has been conducted on safety. Standard OCs are known to increase the risk of thrombosis (clotting), which can be very serious. In a large 25-year study of 46,000 women in the UK, Dr. Valerie Beral found that the use of oral contraceptives were associated with higher mortality from cerebrovascular events (such as stroke), and lower mortality from ovarian cancer and colorectal cancer. On balance, women who have used oral contraceptives had no higher mortality than non-users (Beral et al., 1999).
In addition to our interest in the standard risks of oral contraceptives, we also considered possible increased risks from Long OC. Long OC use results in a slightly higher total exposure to the high levels of estrogens in oral contraceptives. It also limits the amount of time that a woman’s body is free from external hormones. In a normal menstrual cycle, there is a cycle between high estrogen exposure and a time of low estrogen exposure. Long OC reduces the time that a woman’s breast tissue and endometrium are given a break from the high levels of estrogen.
Finally, oral contraceptives act by suppressing the complex and intricate hormonal feedback system between the hypothalamus, the pituitary and the ovary. They act at the level of the brain, preventing the signals that normally initiate follicular development and ovulation. Women who try to become pregnant after taking oral contraceptives take several extra months, on average, to become pregnant, when compared with women who stop using a barrier method (e.g., condoms or diaphragm) or an IUD (Bracken, Hellenbrand, & Holford, 1990). Higher dose oral contraceptives are associated with a longer delay to conception. Long OC has a slightly higher total dose of estrogen, but also has the potential to suppress the hypothalamic-pituitary-ovarian axis more strongly than Standard OC.
(Cachrimanidou et al., 1994) studied 13 women on Long OC (63/7) and 7 women on Standard OC to look at clotting factors and cholesterol. All of the women in her study were new users. There were no statistically significant differences in blood lipids (cholesterol, HDL, LDL) between the two groups. For the most part, blood lipids did not change across the year. Total cholesterol tended to rise across the year, and this increase was significant in the Long OC group. There was also an increase in the VLDL component of cholesterol after 3 months in the Standard OC group. There were increases in triglycerides under both OC regimens. Factors related to coagulation increased under both OC regimens, while coagulation inhibitors showed small decreases. This study demonstrates some of the physiological changes that may be associated with increased risk of clotting in OC users. Long OC users showed a non-significant tendency to be more affected than were Standard OC users.
We found no studies of breast or endometrial safety in these papers on Long OC. There are anecdotal reports of perhaps a dozen women among the various studies who discontinued Long OC and become pregnant. However, we did not find any systematic study of the return to fertility or normal ovulatory function following discontinuation of a Long OC schedule.
Long OC and menstrual taboos
Advocates of menstrual suppression with an extended schedule of oral contraceptives have several arguments. Some suggest that women might occasionally extend the number of active pills to delay menstruation on particular occasions. Some suggest that women taking oral contraceptives should have a choice about how frequently they menstruate. They often point out that the original schedule of oral contraceptive pills, 21 days of active pills followed by a seven day pill-free window (a 21/7 schedule, for short), was an arbitrary one, chosen in the vain hope that a normal menstrual pattern would render oral contraception acceptable to the Vatican.
Other advocates of menstrual suppression argue that fewer menstrual bleeds will result in less menstrual discomfort and premenstrual discomforts. Finally, some have suggested that unmedicated menstruation is itself unhealthy, and that extended use of oral contraceptives will improve women’s health.
Menstruation is a taboo in our culture, and women who are menstruating take efforts to conceal the fact. Women differ in how they feel towards menstruation, and those women who find menstruation most distasteful and embarrassing are the women who are most interested in the use of Long OC for menstrual suppression. Women also vary in the amount of discomfort they experience with menstruation. In a study presented at the recent Society for Menstrual Cycle Research meetings, distaste was more important than discomfort in predicting women’s attitudes to menstrual suppression with Long OC (Hoyt & Andrist, 2003).
Adolescents may be particularly vulnerable to Long OC. Having someone discover that you are menstruating is among the top mortifying moments that teenage girls share with magazine column writers (Houppert, 1999). Being able to avoid and control menstrual flow may become a status symbol as well as a convenience, adding peer pressure to the other reasons a young woman might use Long OC. A recent “how to” article for clinicians documents a variety of physical and social reasons to prescribe extended oral contraceptives to adolescents (Sucato & Gold, 2002). While the author states that there are no data on Long OC use by teenagers, she does not go so far as to urge caution, nor to comment that there are contraindications for oral contraceptive use on any schedule. The potential for harm from Long OC use is greater when taken by developing girls. It is common for there to be a vulnerable period during growth and development, and it is not known what effect extending the suppressive effects of oral contraceptives will have on a developing reproductive system. Risks must be weighted against benefits, and the prevention of pregnancy in a sexually active teenager is a greater benefit than the suppression of menstruation for convenience. Moreover, in general, Long OC for menstrual suppression is likely to be sought at a younger age than OC for contraception.
In the coming weeks and months, there will likely be much more media coverage of the idea of menstrual suppression with Long OC. There are research studies that have been conducted on Seasonale®, but the results of these studies have not yet been published. To date, the studies that have been published show that extending the number of active pills between pill free intervals decreases the amount of scheduled withdrawal bleeding, but increases the amount of unscheduled bleeding and spotting, compared with Standard OC schedules. New users have more problems than women switching from Standard OC to Long OC, and the amount of unscheduled bleeding becomes better with time. The primary reason for women to discontinue Long OC is dissatisfaction with the bleeding patterns.
In our review we found that, contrary to what has been commented on in the media, there are a number of important questions about the safety of using Long OC that have not yet been addressed. There are no published data on how an extended OC schedule affects the endometrium or breast tissue, and there are no systematic data on how long it takes women to return to normal ovulatory function or to become pregnant after stopping a Long OC regimen. Particularly for adolescents, it is important to confirm that the suppression of menstruation with Long OC is reversible and does not affect development. The fact that many women have been using Long OC for years suggests that there is no dramatic effect. However, we need systematic trials, not just testimonials and expert advice from clinicians who see a small number of women. Recent results from the Women’s Health Initiative trial of estrogen-plus-progestin hormonal therapy amply demonstrate that randomized placebo controlled trials are required to test even the most firmly held and widespread beliefs about the benefits of preventive hormone therapy.
- Beral, V., Hermon, C., Kay, C., Hannaford, P., Darby, S., & Reeves, G. (1999). Mortality associated with oral contraceptive use: 25 year follow up of cohort of 46,000 women from Royal College of General Practitioners’ oral contraceptive study. British Medical Journal, 318, 96-100.
- Bracken, M. B., Hellenbrand, K. G., & Holford, T. R. (1990). Conception delay after oral contraceptive use: the effect of estrogen dose. Fertility & Sterility., 53(1), 21-7.
- Cachrimanidou, A.- C., Hellberg, D., Nilsson, S., von Schoulz, B., Crona, N., & Siegbahn, A. (1994). Hemostasis profile and lipid metabolism with long interval use of a desogestrel-containing oral contraceptive. Contraception, 50 (August), 153-165.
- Cachrimanidou, A.- C., Hellberg, D., Nilsson, S., Waldenstrom, U., Olsson, S.- E., & Sikstrom, B. (1993). Long interval treatment regimen with a desogestrel-containing oral contraceptive. Contraception, 48 (September), 205-216.
- Houppert, K. (1999). The Curse: confronting the last unmentionable taboo: menstruation. New York: Farrar, Strauss and Giroux.
- Hoyt, A., & Aandrist, L. C. (2003). Women’s attitudes and beliefs about menstrual suppression. Society for Menstrual Cycle Research, Conference Presentation.
- Miller, L., & Notter, K. M. (2001). Menstrual reduction with extended use of combination oral contraceptive pills: randomized controlled trial. Obstetrics & Gynecology., 98 (5 Pt 1), 771-8.
- Sucato, G. S., & Gold, M. A. (2002). Extended cycling of oral contraceptive pills for adolescents. Journal of Pediatric & Adolescent Gynecology., 15 (5), 325-7.
- Sulak, P. J., Cressman, B. E., Waldrop, E., Holleman, S., & Kuehl, T. J. (1997). Extending the duration of active oral contraceptive pills to manage hormone withdrawal symptoms. Obstetrics & Gynecology, 89 (2), 179-83.
- Sulak, P. J., Kuehl, T. J., Ortiz, M., & Shull, B. L. (2002). Acceptance of altering the standard 21-day /7-day oral contraceptive regimen to delay menses and reduce hormone withdrawal symptoms. American Journal of Obstetrics & Gynecology, 186 (6), 1142-9.