Amenorrhea associated with contraception—an international study on acceptability

1. Introduction 

When Pincus and colleagues developed the combined oral contraceptive pill, they purposely introduced a regimen which would confer a monthly withdrawal bleed because “these artificial menstrual cycles give assurance to the contraceptive user of ‘normal’ genital function” [1]. In contrast, the 3-monthly injectable method of contraception, depot medroxyprogesterone acetate (Depo Provera®), which became available a decade or more later, inhibits cyclical ovarian activity, doing away with menstrual periods in most users. For many years, amenorrhea was widely regarded as the price women had to pay for the clear advantages of the method in terms of efficacy and duration of action. When the levonorgestrel-releasing intrauterine device (Mirena®) came onto the market in the 1990s, the amenorrhea commonly associated with its use was heralded by the manufacturer, providers and users as a positive benefit of the method. There is no medical advantage to menstruation per se. On the contrary, the morbidity associated with menstruation is impressive. Menstrual dysfunction is one of the most common reasons for which a woman consults her general practitioner and in some countries up to 20% of women will undergo hysterectomy for excessive menstrual bleeding [2].

In the developed world, the gradual acceptance of amenorrhea associated with contraception has recently attracted interest. Surveys suggest that increasing numbers of women welcome methods which induce amenorrhea and indeed many manipulate their pill-taking to achieve it [3], [4]. The debate has been taken a step further recently by the suggestion that whether to menstruate or not should be a matter of choice [5], [6], [7]. However, perceptions of menstruation vary according to culture and religion and women’s attitudes to changes in bleeding patterns associated with contraceptive use vary widely. Research undertaken by the World Health Organization (WHO) published in 1981 suggested that most women (even in the UK, but particularly in the developing world) preferred to have a monthly bleed and were unwilling to use a method of contraception which induced amenorrhea [8]. These findings have strongly influenced not-for-profit agencies involved in contraceptive development, which have tended to shy away from exploring such methods. Since much of the research on attitudes to menstruation was carried out decades ago, we have undertaken a survey among women attending family-planning clinics, and among their providers. The study was designed to explore attitudes towards menstruation and willingness to use a method of contraception which induces amenorrhea.

2. Materials and methods 

The survey was undertaken in two centers in the People’s Republic of China (Shanghai and Hong Kong), two centers in Africa (Sagamu in Nigeria and Cape Town in South Africa), and in one center in Scotland (Edinburgh).

Two questionnaires were developed, one for women attending family-planning clinics (clients) and another for health professionals responsible for providing contraception (providers). Pilot versions of the relevant questionnaire were administered to 20 clients and five health-care providers at each center. Both questionnaires were modified according to the consensus of comments received and a revised version of the questionnaire for providers (which was changed quite extensively) was piloted before being finalized. The questionnaires were translated into local languages where appropriate and independent back-translation was performed to validate interpretation. The client questionnaire contained questions concerning gynecological history; methods of contraception ever used and reasons for stopping; clients/partners views towards menstruation; attitudes towards amenorrhea and, in particular, towards a method of contraception which may result in amenorrhea. The questionnaire for health-care providers probed factors which influenced their choice of contraception for a particular client and their perception of the importance to clients of menstruation during contraceptive use.

3. Results 

The questionnaire was administered to 200 clients in each of the five centers. In Edinburgh, Shanghai and Nigeria, no one refused to participate. In Hong Kong, 4 women and in Cape Town, 12 women declined to take part, mainly due to lack of time. At least 50 providers in each of the five centers were sent a questionnaire. All questionnaires were returned in Cape Town, Shanghai and Nigeria. In Hong Kong, 50 of 76 questionnaires were returned. In order to obtain 50 returned questionnaires from Scotland, a further 50 were distributed to the medical staff of the Sandyford Institute, Family Planning and Reproductive Health Center, in Glasgow. A total of 71 questionnaires were returned from 100 distributed to health-care providers in Edinburgh/Glasgow.

The demographic characteristics of the clients are illustrated in Table 1. As expected, there were significant differences between the various centers reflecting the different cultures. Clients in Edinburgh, Cape Town and Shanghai tended to be younger than those in the other two centers (p < 0.001) and more clients in Hong Kong, Shanghai and Nigeria were married. In Edinburgh most clients did not have children, whereas in Nigeria the majority of clients had four children or more. In contrast to the other two centers, the majority of clients in Edinburgh, Hong Kong and Shanghai stated that they had no religion.

Table 1. Demographic characteristics of clients from all centersa

		ED (n = 200)	CT (n = 201)	HK (n = 200)	SH (n = 200)	NG (n = 200)	Significance of difference between centers
	Age (years) (%)
	20–29	53	56	24	49	13	p < 0.001
	30–39	33	33	44	36	62
	>40	11	7	31	16	26
	Marital status (%)
	Married/Cohabit	49	33	84	81	99	p < 0.001
	Single	48	62	15	20
	Ethnic group (%)
	B	–	34	–	–	100
	W	97	33	–	–	–	p < 0.001
	Ch	–	–	92	99	–
	C	–	34	–	–	–
	Religion (%)
	Christian	35	84	9	1	78	p < 0.001
	Muslim	0.5	9	0.5	1	23
	Buddhist	–	0.5	7	2	–
	No religion	64	6	74	96	–
	No. of children (%)
	0	76	46	31	42	2	p < 0.001
	1	11	26	28	55	5
	2–3	13	25	41	4	36
	≥4	1	4	0.5	–	59

a ED = Edinburgh; CT = Cape Town; HK = Hong Kong; SH = Shanghai; NG = Nigeria; B = black; W = white; Ch = Chinese; C = colored.

The vast majority of women had regular menstrual cycles of average length (26–35 days) and menstrual bleeds of average duration (4–7 days). In no center did more than 4% of women have menstrual periods lasting more than 7 days. At least half of the women in every center reported their menstrual flow as being “normal.” Two percent of women in Hong Kong, 10% in Edinburgh and 29% in Cape Town had no periods—most of them were using Depo Provera as their method of contraception. Of those women who stated that they experienced menstrual problems, the most common complaint in all centers was dysmenorrhoea (32–40%).

Table 2 illustrates the methods of contraception currently used by clients. The patterns of use were similar to those observed in previous studies from these centers [9], [10]. In Edinburgh, the oral contraceptive pill was the most popular choice, while in Africa, injections were the most commonly used method (Depo Provera in South Africa and norethisterone enanthate in Nigeria). Nearly 40% of clients in Hong Kong and Shanghai used condoms.

Table 2. Current contraception (%)—clientsa

		ED	CT	HK	SH	NG
	COC	40	35	28	6	17
	POP	7	5	–	–	0.5
	Condoms	21	2	38	39	23
	IUCD	9	–	12	17	27
	Inject/implant	10	53	9	2	30
	None—at risk	0.5	4	2	13	–
	None—no need	4	2	4	6	–
	Planning pregnancy	3	–	2	0.5	–

a ED = Edinburgh; CT = Cape Town; HK = Hong Kong; SH = Shanghai; NG = Nigeria; COC = combined oral contraceptive pill; POP = progesterone-only pill; IUCD = intrauterine contraceptive device.

Except for black women in the two African centers, more than half the women said they did not like having periods (Table 3). In South Africa, white or colored women were significantly more likely to say they disliked periods compared with black women (p < 0.001). The most common reason for not liking periods was because they were “inconvenient” (65–85%). In Edinburgh and Hong Kong, 33% and 13% of women, respectively, disliked periods because of associated menstrual problems. In contrast, 81% of women in Nigeria said they liked having periods, the most common reason cited was “to get rid of bad blood.” They also liked having a period to reassure them that they were not pregnant. In all other centers, the most common reason for liking periods was because they were perceived as “natural.”

Table 3. Attitudes and experience of clients towards contraception and amenorrheaa

		ED	CT-B	CT-W	CT-C	HK	SH	NG	Significance of difference between centers
	Do you like having periods? (%)
	Yes	26	75	35	42	50	33	81	p < 0.001
	No	74	25	65	58	51	63	19
	Don’t know	–	–	–	–	–	5	–
	If you could choose, how often would you like to have a period? (%)
	Monthly	33	49	30	42	42	43	71	p < 0.001
	3-monthly	20	27	26	15	39	30	12
	Never	37	9	29	36	6	15	13
	Have you used a contraceptive method which stopped periods? (%)
	Yes	23	60	32	37	–	1	20	p < 0.01
	No	77	40	68	63	100	99	80
	Provided your periods and your fertility returned to normal immediately after you stopped using it, would you consider a method of contraception which stopped your periods? (%)
	Yes	65	52	64	61	37	48	73
	No	25	41	26	33	32	35	24
	Undecided	11	7	11	6	32	18	4

a ED = Edinburgh; CT-B = Cape Town (black); CT-W = Cape Town (white); CT-C = Cape Town (colored); HK = Hong Kong; SH = Shanghai; NG = Nigeria.

Offered the choice of how often the clients would like to have a period, there were significant differences between centers (Table 3). More than one-third of all women in Edinburgh and colored women in Cape Town preferred never to menstruate, compared with fewer than 10% of women in Hong Kong and of black women in Cape Town. The proportion of women preferring a monthly bleed ranged from 30% among white women in Cape Town and 33% in Edinburgh, to 71% in Nigeria.

With the exception of Cape Town, the majority of clients in all centers had no experience with use of a method of contraception which stopped periods (Table 3). Within Cape Town, there were, however, differences between ethnic groups (p = 0.002). The majority (60%) of black clients had used a method of contraception which had stopped periods (Depo Provera) compared with fewer than one third of white women and 37% of colored women.

More than half the clients in Edinburgh, and all three ethnic groups in Cape Town, said that they would consider using a contraceptive method which resulted in temporary cessation of menstruation. In Shanghai and Hong Kong, more than one-third of women would be willing to use such a method. In Edinburgh, Cape Town and Shanghai, women who did not like having periods were more likely than those who did to consider a method of contraception which stopped periods (p < 0.001), but this was not the case in Hong Kong or Nigeria. Even amongst women in Nigeria, and black women in South Africa, most of whom had expressed a preference for having periods (Table 3), more than half of the women were favorable to the idea of using a method of contraception which would temporarily end menstruation.

The acceptability of a contraceptive which induced amenorrhea was unrelated to age, parity, education or religion except in Shanghai, where younger women (52%, p = 0.015) and women who wanted children (p = 0.017) tended to be more positive than those who did not. There was also no correlation between willingness to use this method and how the client perceived her own menstrual period (i.e., normal, heavy or light).

There were significant differences between centers in the demographic characteristics of the health-care providers (Table 4). In all centers, over 70% of the providers were women and most were married. However, providers in Edinburgh/Glasgow, Cape Town and Nigeria were older than those in China (p = 0.0013) and the majority in Scotland and Hong Kong did not have children. In Scotland and in Hong Kong, providers were much more likely than their clients to describe their religion as Christian.

Table 4. Demographic characteristics of providers from all centers

		ED (n = 71)	CT (n = 50)	HK (n = 50)	SH (n = 50)	NG (n = 69)	Significance of difference between centers
	Age (years) (%)						p < 0.01
	20–29	17	13	20	26	10
	30–39	36	29	53	42	36
	>40	47	58	27	32	54
	Marital status (%)						p < 0.001
	Married/cohabit	75	64	70	84	93
	Single	24	28	30	16	7
	Ethnic group						p < 0.001
	B	–	31	–	–	100
	W	92	35	–	–	–
	Ch	–	–	100	100	–
	C	–	35	–	–	–
	Religion (%)						p < 0.001
	Christian	71	80	34	–	86
	Muslim	3	8	–	–	14
	Hindu	3	2	–	–	–
	No religion	23	–	62	100	–
	No. of children (%)						p < 0.001
	0	41	24	46	30	7
	1	14	28	24	68	5
	2–3	43	40	30	2	58
	≥4	3	8	–	–	24

ED = Edinburgh/Glasgow; CT = Cape Town; HK = Hong Kong; SH = Shanghai; NG = Nigeria; b = Black; w = White; Ch = Chinese; c = Colored.

The attitudes of the health-care providers towards contraception and amenorrhea are illustrated in Table 5. Most providers in Nigeria, Shanghai and Hong Kong thought that it was important for women to menstruate while using contraception. In all the centers at least 75% thought their clients considered that menstruating whilst using contraception was important. Despite this, in all centers except Shanghai, more than half the providers would recommend a method which stopped menstruation.

Table 5. Providers’ attitudes towards contraception and amenorrhea

		ED	CT-B	CT-W	CT-C	HK	SH	NG	Significance of difference between centers
	Do you think it is important for women to have menstrual bleeding whilst using contraception? (%)								p < 0.001
	Yes	18	47	13	18	65	82	74
	No	82	53	88	82	35	18	26
	How important does the client herself think it is to menstruate whilst using contraception? (%)
	Very important	18	60	25	24	26	28	79
	Quite important	74	33	56	65	72	62	17
	Not important	6	7	19	6	2	10	4
	Would you recommend a method of contraception which stops bleeding? (%)
	Yes	92	71	100	77	66	36	52	p<0.001
	No	8	29		24	34	64	48

ED = Edinburgh/Glasgow; CT-B = Cape Town (black); CT-W = Cape Town (white); CT-C = Cape Town (colored); HK = Hong Kong; SH = Shanghai; NG = Nigeria.

4. Discussion 

Regular monthly periods are a relatively recent phenomenon [11]. Until modern contraceptives became available, most women spent much of their lives pregnant or breastfeeding and, therefore, amenorrheic. Menses only returned briefly at weaning, which was rapidly followed by another pregnancy. The epidemic of menstrual cycles coincided with the demographic transition from developing to developed country status with low fertility rates. Faced with the inevitable inconvenience of repeated menstrual periods, women perhaps unsurprisingly adopted a positive attitude. Thus, periods were seen as “normal,” a means of “getting rid of bad blood” and a sign of fertility. Amenorrhea could indicate ovarian failure and a sign of the end of reproductive potential at the menopause.

These ingrained cultural attitudes may explain the results of previous surveys. In a questionnaire survey of over 5000 women from 10 countries (Egypt, India, Indonesia, Jamaica, Korea, Mexico, Pakistan, Philippines, UK and Yugoslavia) undertaken in the late 1970s, the majority of women in all cultural groups were unwilling to accept a contraceptive which induced amenorrhea [8]. The size of the majority varied from 50% in Korea and 53% in the UK to 85% in India and 91% in Pakistan. Women in this survey were said to feel that amenorrhea was “unnatural” and that menstruation was “an outlet for bad blood.” In a qualitative study undertaken in the early 1990s using focus discussion groups involving 576 women from seven countries (Cambodia, India, Mexico, Pakistan, Peru, South Africa and USA), a majority of women in every site strongly expressed their overall dissatisfaction with available methods of contraception [12]. The authors reported that, when directly asked, “women were largely dismayed by the prospect of a contraceptive method causing amenorrhea.” Women were concerned about the negative health effects, fear of pregnancy and the difficulties that amenorrhea caused with keeping contraception secret from husbands.

In contrast, in our study, more than 50% of women in Edinburgh, Cape Town and Sagamu (Nigeria), and more than one-third of women in Hong Kong and Shanghai, would consider using a contraceptive method which induced amenorrhea. The proportion was high even among population subgroups of women (or cultures) in which the preference for having periods was widespread.

Why should our results be so different from those reported by the two earlier studies? Neither study described the source of the respondents but there is no indication that they were recruited from a health-care setting. Both sought only the views of married women with children. The women who took part in the questionnaire study [8] were markedly older than women participating in our study—in all but two cultural groups, at least one third were aged over 35 and in nine fewer than 15% were aged under 24. In the focus discussion groups [12], the mean age of participants was 32, and although the upper age limit was said to be 35, the reported range was much greater. In our own study, all the participants were attending family-planning clinics, many (76% in Edinburgh) had no children and more than 49% in Shanghai, Edinburgh and Cape Town were aged under 30. The differences could be due, therefore, to the type of women participating in the studies. However, in our survey, the women most likely to consider using a contraceptive which induced amenorrhea were the women from Nigeria (73%). Yet, these were the women who were most likely to like having periods (81%); most likely to choose to have a monthly withdrawal bleed (71%) and tended to view periods as a way of “getting rid of bad blood.” They were also most likely to be married with children. So, in every respect, this group of women was more similar to those who participated in the two earlier studies. The study by Snow and colleagues [12] included 98 women from Cape Town and the WHO study [8] included a center in the UK. While the former study gave no absolute numbers, in the latter 53% of women from the UK were unwilling to accept a method of contraception which induced amenorrhea (compared with only 25% in Edinburgh in 2000). Perhaps a more likely explanation for the difference between the findings of our study and those of WHO and Snow is the effect of the passage of time. In a questionnaire survey of 178 pill users in Sydney, Australia, although 83% of women believed that a monthly bleed was necessary, 27% would choose to bleed every 3 months, 4% every 6 or 12 months, and 15% never [3]. Participants for this study were recruited from family-planning clinics or general practice, two thirds of the women were between 20 and 29 years, and almost half had never been married. Thus, these Australian women were more like our own respondents. Furthermore, this survey was done in the late 1980s when it is possible that women were becoming more accustomed to the concept that amenorrhea might have some benefits and that they do not have to bleed each month. The results may also reflect an increasing willingness of women to accept therapeutic interventions if the benefits are seen to be real.

In a more recent study undertaken in the Netherlands [4] published in 1999, the majority of women aged 15–34 preferred a bleeding frequency of less than once a month, and 26% of women aged 15–19 and 31% of those aged 25–34 would have preferred never to bleed. Given the opportunity to design their own pill regimen, one third of women aged 15–19 opted for a bleed once every 3 months and between 22% (aged 15–19) and 26% (aged 45–49) would design a regimen that induced amenorrhea.

Although neither the WHO survey [8] nor the Snow study [12] specifically asked women whether or not they liked having periods, WHO did ask women about physical discomfort and mood change and whether they would be willing to accept less bleeding. Although more than half of the women in every cultural group admitted to physical discomfort, and many to mood change in association with menses, in every center more than 60% (and in seven groups over 80%) of women were unwilling to accept less bleeding. In contrast, in 2000 in our study, only among black women (in both Cape Town and in Nigeria) did more than 50% of women say that they liked having periods. In Edinburgh, Cape Town and Shanghai, women who did not like having periods were more willing to try a contraceptive method which stopped periods.

There are data which testify to women’s willingness to use the oral contraceptive pill (OC) to induce amenorrhea. More than 25 years ago, it was demonstrated in a Scottish study that the majority of women given the opportunity to use an OC continuously for 3 months found the regimen acceptable [13]. Ninety-one percent of the 107 women who completed the study (out of 195 women who started) refused to revert to the monthly regimen. Most of the women participating in similar trials of extended OC cycles in Australia [14], Sweden [15] and the USA [16] welcomed infrequent menses. Recognizing an increasing tendency for women to manipulate bleeds, a phase III trial of a combined oral contraceptive formulation taken continuously for 84 days followed by 7 days of placebo (Seasonale, Barr Laboratories) is underway in the USA. A similar study of continuous use of the combined contraceptive transdermal patch (ORTHO EVRA/EVRA, R.W. Johnson Pharmaceutical Research Institute, Raritan, NJ, USA) is also ongoing.

Anecdotal evidence from developed countries suggests that it is becoming increasingly common for women to manipulate their cycles to avoid a withdrawal bleed. In the study of Rutter and coworkers [3], 43% of women had used the pill to alter the timing of withdrawal bleeding for special occasions, holidays and weekends. Even women who are not using the OC look for ways to postpone menstruation. A review of prescribing patterns among general practitioners in Oxfordshire (UK) which showed clear peaks of prescribing every year for 4 years over the summer holidays, led the authors to describe norethisterone as a lifestyle drug [17].

Many factors influence a couple’s choice of contraception. Providers, both through which methods they make available and accessible, and by their own attitudes and preferences, have a major influence. Few studies have looked at the influence of providers. Indeed, Snow [12] expressed regret at not probing further about the effect that providers have on determining women’s preferences. In the Scottish study of tricycling the OC, while 82% of women who used the regimen welcomed the reduction in the number of periods, half of the doctors who worked in the participating clinic preferred to stick to the monthly regimen of prescribing [13]. Some 16 years later in the Australian study, 20 female doctors working at a Sydney teaching hospital answered the same questionnaire as the OC users [3]. The doctors were young (mean age: 24) but 40% were married. Sixty percent of them believed that a monthly bleed was necessary on the pill but more than half opted for less frequent bleeding episodes when given the choice of how often they personally would prefer to bleed and 30% chose never to bleed. In our own study, in every center except in white Cape Town, at least 90% of providers thought that clients felt that regular menses were either quite important or very important, while a much lower percentage of the clients themselves seemed to choose a monthly withdrawal bleed. In Nigeria, only 52% of providers, and in Shanghai only 36%, would recommend a contraceptive which induced amenorrhea, yet 73% of women in Nigeria and 48% of women in Shanghai said they would be prepared to try one. Thus, there was a tendency for providers’ recommendations to correlate far more closely with their own views about menstruation than with the views of the women they treat. It is possible that the phenomenon of Westernization which is underway in major Chinese cities like Shanghai enhances the ability of clients to disagree with their providers and to adopt their own views of contraception as they become better informed. In black Africa, even in Cape Town, perhaps women have yet to become emancipated in this respect.

The acceptability of a method is often inferred from continuation rates. However, since most women using contraception are highly motivated to avoid pregnancy, they are prepared to tolerate side effects and the chosen method may be the “least worst alternative.” Continuation rates may only be a reflection of the characteristics of the “best” method available rather than of those of a method which women would be truly happy to use were there more choice. Our study measures the hypothetical acceptability (or desirability) of one feature of a method of contraception—amenorrhea. Responses to a questionnaire survey may not be a true reflection of the number of women who would be prepared to try a method of contraception which induced amenorrhea if the method was actually available nor how many would continue to use it having tried it. However, our survey does suggest that attitudes may be changing. Scientists and funding agencies involved in developing new methods of contraception [18] and improving existing ones should be reassured that amenorrhea could be a popular option.