Why women miss pills: research identifies reasons and suggests how to improve consistency of use.

When used correctly and consistently, oral contraceptives (OCs) are among the most effective reversible methods of contraception. But reported pregnancy rates during the first year of OC use are as high as 32 percent. (1) Because a major contributing factor to these OC "failures" is thought to be missed pills, researchers are trying to determine how women's daily routines, interpretations of pill taking, or knowledge about OCs affects their pill use. Such information is needed so that family planning programs can help clients take OCs more consistently.

One place where this issue has been explored is China. Family planning there is nearly universal among people of reproductive age, and OCs are free and widely available. Yet the pregnancy rate during the first year of OC use has been about 11 percent. (2) much higher than that in many other countries. Investigators from the University of Michigan School of Nursing, Ann Arbor, Michigan, USA, and China's Hunan Family Planning Committee, Hunan Family Planning Institute, and Beijing University have sought to determine why this is so. (3)

Five urban and five rural women who were married and using No. 1, a Chinese brand of combined OCs containing 35 [micro]g ethinyl estradiol and 600 pg norethindrone, were included in the study. All women were instructed to take one pill a day for 22 consecutive days. No placebo pills were available, and women were to resume pill taking on the fifth day of menses. For three cycles, women were given a special pill package with a computer inside that recorded the time and date that each pill was dispensed.

During in-depth personal interviews each woman was asked to develop a detailed calendar showing unusual events that occurred during each of her cycles, including sickness, absence from home, overnight visitors, or other disruptions to home or work routines. Each woman was also shown the computer-recorded data on her pill use and asked to explain missed pills or extended pill-free intervals.

Computer-recorded data showed, notably, that no woman who remained--the study for all three cycles took all of her pills on the correct days. According to the World Health Organization, women may be at increased risk for pregnancy if they miss as few as two active pills in a row (depending on when in the cycle they miss them) or if they extend the pill-free interval beyond seven days. (4) Four women missed at least two consecutive pills during the study; three of the same women also had an overly long pill-free interval. Although no pregnancies were reported, investigators considered three of the 10 women to have been at increased risk of pregnancy.

Analysis showed three main reasons for missed pills: changes in the routine of daily life, absence of husbands, and presence of bleeding. Interviews showed that the women sometimes confused spotting with menses and often did not take pills if they detected any bleeding at all. Data also showed that more rural than urban users took their pills consistently. The researchers hypothesized this occurred because rural users had more routine daily schedules.

On the basis of their data, the researchers suggested several yet-to-be evaluated strategies to improve consistency of OC use in China:

* Educational materials could be created to stress the importance of consistent OC use even when a husband is temporarily not at home.

* Pill-taking instructions could be changed so that women are told to take one active pill daily for 22 days and to resume taking active pills after six pill-free days, regardless of menstruation. This would create a more routine 28-day cycle and remove any link between pill taking and menses, so women would no longer have to interpret the meaning of bleeding.

* Mention of menses could be removed from pill-taking instructions to further dispel the myth that pills cannot be taken on bleeding days.

Bangladesh provides another good opportunity to study pill-taking behaviors. Nearly half of all contraceptive users in Bangladesh take the pill? yet studies show that many Bangladeshi women do not follow correct pill-taking procedures. (6) A recent study conducted by the University of New England in Australia, Ipas in North Carolina, USA, and the University of Dhaka in Bangladesh aimed to determine predictors of inconsistent OC use in rural Bangladesh. (7)

The study included 801 of some 1,400 OC users served by government family planning workers (FPWs) and surveyed between 1995 and 1996 about adherence to OC pill-taking regimens. Women in the study had been using 28-day pill packets containing 21 active pills and seven iron or placebo pills for at least six months.

Self-reports of past pill-taking behavior were recorded for each woman. A woman's pill taking was defined as inconsistent if she remembered missing one or more active pills during the last six months of OC use. Several factors--including religion, place of residence, access to television or radio, duration of OC use, side effects, knowledge about contraindications, and visits by a FPW during the last six months--were also analyzed as potential predictors of inconsistent use.

Half of the women reported missing at least one active pill during the last six months, but an even higher proportion may have used their OCs inconsistently. Research conducted by FHI in 1996 comparing self-reported data on pill-taking behavior with computer-recorded data (such as that used in the study from China) has shown that in self-reports many women underestimate the number of pills that they miss. (8)

Data also showed that four factors significantly increased a woman's risk of inconsistent OC use: lack of knowledge about contraindications to OC use, no visit by a FPW in the last six months, Islam as a religion, and no access to television or radio. Lack of knowledge about contraindications was the most significant predictor of inconsistent use. This finding suggests that, in general, "less-informed women may have a tendency to use the pills inconsistently, and that increased access to more comprehensive information could help to alleviate this trend," the researchers stated.

To increase consistent use of OCs in rural Bangladesh, the authors made the following recommendations, which--while not evaluated in Bangladesh or elsewhere--may be applicable to rural settings in other countries:

* Regular in-service training about issues related to OC use should be offered to providers. All potential OC users should be counseled on the contraindications and possible side effects of OC use, as well as on how to use OCs correctly.

* Regular contact is needed between service providers and clients in rural areas. The family planning program of Bangladesh recently switched from a home-delivery system to a fixed-site, clinic-based delivery system, which needs to be promoted to improve women's awareness, and use, of the clinics.

* For social or religious reasons, some Muslim women have limited mobility within their communities, which may decrease their contact with service providers. Since many Muslim women do not leave the home without a male companion, involving men in women's reproductive health decisions--by counseling men as well as women--could facilitate adherence to OC regimens.

* Behavior change communication materials need to be revised, and mass media programs could be adjusted, to include information on user behavior, such as instructions on how to take pills correctly and on what to do if pills are missed. Also, instructions should reinforce the importance of taking pills every day.

REFERENCES

(1.) Jejeebhoy S. Measuring contraceptive use-failure and continuation: an overview of new approaches. In Bogue DJ, Arriaga EE, Anderton DL, eds. Readings in Population Research Methodology. New York, NY: United Nations Fund for Population Activities, 1993; Fu H, Darroch JE, Haas T, et al. Contraceptive failure rates: new estimates from the 1995 National Survey of Family Growth. Faro Plann Perspect 1999;31(2):56-63.

(2.) Wang SX, Wing SG, Hang M. A study on the effects of common contraception measures in China. Popul Res 1991;29:1 [In Chinese].

(3.) Oakley D, Yu M-Y, Zhang Y-M, et al. Combining qualitative with quantitative approaches to study contraceptive pill use. J Women's Health 1999;8(2):249-57.

(4.) World Health Organization. Selected Practice Recommendations for Contraceptive Use. Geneva, Switzerland: World Health Organization, 2002.

(5.) National Institute of Population Research and Training, Mitra and Associates, and ORC Macro International Inc. Bangladesh Demographic and Health Survey 1999-2000. Dhaka, Bangladesh, and Calverton, MD: National Institute of Population Research and Training, Mitra and Associates, and ORC Macro International Inc., 2001.

(6.) Mitra SN, Lerman C, Islam S. Bangladesh Contraceptive Prevalence Survey, 1991 Key Findings. Dhaka, Bangladesh: Mitra and Associates, 1992; Larson A, Islam S, Mitra SN. Pill Use in Bangladesh: Compliance, Continuation, and Unintentional Pregnandes. Report of the 1990 Pill Use Study. Dhaka, Bangladesh: Mitra and Associates, 1991.

(7.) Khan MA, Trottier DA, Islam MA. Inconsistent use of oral contraceptives in rural Bangladesh. Contraception 2002;65(6):429-33.

(8.) Potter L, Oakley D, de Leon-Wong E, et al. Measuring compliance among oral contraceptive users. Faro Plann Pertpect 1996;28(4): 154-58.