Method	%
Combined hormonal contraceptives	18
Progestogen-only pill	6
Implant/injectable	5
Intrauterine device/system	7
Barrier methods	24
Vasectomy	10
Female sterilization	7
Natural family planning	2
Withdrawal	4
No method	25^a

^a Subcategories: ‘not having sex’, 14; ‘pregnant or trying to conceive’, 3; ‘menopausal/infertile/otherwise sterile’, 6; ‘doesn’t like/doesn’t use contraception’, 1.

Source: Office for National Statistics 2007/8 Omnibus Survey Report No. 37. Contraception and Sexual Health. Office for National Statistics, London. Available at: www.ons.gov.uk and www.statistics.gov.uk.

Evidence-based recommendations on which individuals can safely use each contraceptive method were first published by the World Health Organization (WHO) in 1996, and are updated at 4-yearly intervals [WHO Medical Eligibilty Criteria (WHOMEC)]. This guidance has been further modified specifically for use in the UK by the Clinical Effectiveness Unit (CEU) of the Faculty of Sexual and Reproductive Healthcare (FSRH) of the Royal College of Obstetricians and Gynaecologists (RCOG), and is available in hard copy and online as the UK Medical Eligibility Criteria (UKMEC). Contraceptive methods are categorized according to their balance of risk and benefit in the presence of certain medical and lifestyle conditions (Table 29.2). Health professionals prescribing contraceptive methods are also directed to the National Institute for Health and Clinical Excellence guidelines on LARC published in 2005.

Table 29.2 UK Medical Eligibility Criteria for Contraceptive Use (RCOG Press, London, 2009)

Classification categories

1 A condition for which there is no restriction for the use of the contraceptive method

2 A condition where the advantages of using the method generally outweigh the theoretical or proven risks

3 A condition where the theoretical or proven risks usually outweigh the advantages of using the method

4 A condition which represents an unacceptable health risk if the contraceptive method is used

Contraception

Combined hormonal contraceptives

The CHC was approved for use in Britain in 1961. In recent years, different modes of delivery have been developed, with injectable, transdermal and vaginal methods now available in some countries. Although unlikely to be substantially different from the oral preparation in terms of efficacy and safety, they do provide a wider choice of administration which may help to offset the effects of erratic pill taking.

Combined hormonal contraceptives (CHCs) contain oestrogen, usually ethinyl oestradiol, and a synthetic progesterone (progestogen). Most modern pills contain 20–35 µg ethinyl oestradiol with either a second-generation progestogen [norethisterone, levonorgestrel (LNG) or ethynodiol diacetate] or a so-called third-generation progestogen (gestodene, desogestrel or norgestimate). The newest progestogen, drospirenone, has antiandrogenic and antimineralocorticoid properties. The 20 µg pills are as effective as the 30 µg pills but, not surprisingly, are associated with poorer cycle control (Akerlund et al 1993).

The traditional CHC regimen involves taking a tablet for 21 days with a 7-day break, thus inducing a withdrawal bleed in order to mimic the body’s ‘natural’ pattern. Variations on this theme include the everyday formulations containing dummy pills so that users do not have to remember to stop and start (common in the USA), 84-day preparations reducing the number of expected withdrawal bleeds from 13 to four per annum, and a continuous version aimed at banishing bleeding altogether. Injectable CHCs are administered intramuscularly every 28 days, with 70% of women experiencing a withdrawal bleed 18–22 days after injection. They are in common use in Latin America. Despite the plethora of transdermal hormone replacement options, there is only one contraceptive patch available. The patch is applied weekly for 3 weeks prior to a patch-free week, and may help to reduce nausea in susceptible individuals (Audet et al 2001). The latest addition to the CHC family is a vaginal ring, 54 mm in diameter and 4 mm in cross-section. It is made of a soft, ethylene–vinyl-acetate copolymer, lasts for 21 days with a 7-day break, and appears to confer more favourable cycle control than other preparations. These longer-acting CHCs may improve compliance.

Combined pills are available in monophasic, biphasic and triphasic preparations. Phasic pills were developed in order to reduce the total dose of progestogen and, by simulating the fluctuating pattern of steroid concentrations in the normal ovarian cycle, in an attempt to produce better cycle control. There is no good evidence that cycle control is superior to that achieved by monophasic pills.

Mode of action

CHCs work primarily by inhibiting ovulation. Exogenous oestrogen inhibits the secretion of follicle-stimulating hormone, while progestogens inhibit the development of the luteinizing hormone (LH) surge. The CHC also alters cervical mucus, rendering it hostile to the passage of sperm, and also causes endometrial atrophy.

CHCs are highly effective (Table 29.3). The failure rates associated with all forms of contraception depend on the inherent efficacy of the method, but also on the potential for incorrect or inconsistent use. Since ovulation is inhibited in most women who use the combined pill, failure rates of the method itself are low. However, as inhibition of ovulation depends on reliable pill taking, the overall failure rate of the COC, which includes user failure, is higher.

Table 29.3 Pregnancy rates for birth control methods (for 1 year of use)

Method	Typical use rate of pregnancy (%)	Lowest expected rate of pregnancy (%)
Sterilization
Male sterilization	0.15	0.1
Female sterilization	0.5	0.5
Hormonal methods
Implant (Implanon)	0.05	0.05
Depo-provera injection	3	0.3
Combined pill (oestrogen/progestogen)	8	0.3
Mini pill (progestogen only)	8	0.3
Evra patch	8	0.3
NuvaRing	8	0.3
IUD/IUS
Copper T	0.8	0.6
LNG-IUS	0.1	0.1
Barrier methods
Male condom^a	15	23
Diaphragm^b	16	6
Female condom	21	5
Spermicide (gel, pessary)	29	15
Sponge
Parous women	32	20
Nulliparous women	16	9
Emergency contraception
Levonelle	15
Copper IUD	<5
Natural methods
Withdrawal	27	4
Natural family planning (calendar, temperature, mucus)	25	3
No method	85	85

IUD, intrauterine device; IUS, intrauterine system; LNG, levonorgestrel.

This table provides estimates of the percentage of women likely to become pregnant while using a particular contraceptive method for 1 year. These estimates are based on a variety of studies.

‘Typical use’ rate means that the method was not always used correctly, was not used with every act of sexual intercourse or was used correctly but failed.

‘Lowest expected’ rates mean that the method failed despite being used correctly with every act of sexual intercourse.

^a Used without spermicide.

^b Used with spermicide.

Adapted from Trussel J 2007 Contraceptive efficacy. In: Hatcher RA, Trussel J, Nelson A, Kates W, Stewart F, Kowal D (eds) Contraceptive Technology, 19th edn. Ardent Media, New York.

Contraindications

In an analysis of 25 years of follow-up of 46,000 women who took part in the Royal College of General Practitioners’ (RCGP) Oral Contraceptive Study comparing 517,519 years of pill use with 335,998 years of never-use, the risk of death from all causes was similar in ever-users and never-users of oral contraception (Beral et al 1999). Among current and recent users (within 10 years), the relative risk (RR) of death from ovarian cancer was significantly decreased (RR 0.2), while the risks of dying from cervical cancer (RR 2.5) and cerebrovascular disease (RR 1.9) were increased.

Absolute contraindications to CHC use are listed in Box 29.1. They include either a past history of, or existing, cardiovascular disease, migraine and most liver diseases. Women often describe headaches as migraine. CHCs are contraindicated in migraine which is or may be associated with transient cerebral ischaemia. A recent study designed to investigate the risk between migraine and stroke in young women demonstrated a significant increase in the risk of ischaemic, but not haemorrhagic, stroke in women with a personal history of migraine both with and without aura (classic and simple). Concomitant use of the CHC further increased this risk (Chang et al 1999). It is important therefore to take a clear and detailed history before refusing to prescribe a CHC because the woman has an occasional migraine.

Relative contraindications to CHC use include the following:

• factors which increase the risk of cardiovascular disease or venous thrombosis, such as obesity, smoking, hypertension and family history;

• sex-steroid-dependent conditions, including cancer;

• factors which adversely affect liver function; and

• factors which predispose to arterial wall disease.

Comprehensive discussions and lists of contraindications are available in most textbooks of contraception, and WHOMEC and UKMEC.

Side-effects of combined oral contraception

Debate continues over the increased risk of cardiovascular events in women taking CHCs. Until reliable data are available for non-oral routes of administration, it is sensible to assume that similar risks apply to all methods. Cardiovascular disease is very rare in women of reproductive age. Venous thromboembolism (VTE) is the most common cardiovascular complication of CHC use. Oestrogen and progestogen are both metabolized by the liver and alter the metabolism of most substances, including carbohydrates and lipids. Oestrogens also alter coagulation factors. A reduction in antithrombin III and alteration in platelet function increase the risk of VTE by up to seven-fold. A clear history of thromboembolism is a contraindication to CHCs. A possible history or a very strong family history are indications for investigating haemostasis, particularly circulating concentrations of antithrombin III and the factor V Leiden thrombogenic mutation. Population screening for inherited thrombophilias is not currently considered to be cost-effective.

Studies in the mid-1990s appeared to show a differential risk of VTE between the second- and third-generation progestogens, the latter being associated with double the risk of the former (Bloemenkamp et al 1995, Jick et al 1995, WHO 1995, Spitzer et al 1996). These findings prompted the UK Committee on Safety of Medicines to advise women at risk of thrombosis to avoid third-generation progestogens. Although well designed, criticisms of these studies include prescriber bias (whereby women with cardiovascular risk factors and new users would be more likely to be prescribed a newer brand) and attrition bias (existing users susceptible to VTE would be less likely to be using older brands). Latterly, the 2005 EURAS study, funded and designed as a postmarketing cohort study to demonstrate the safety of the combined ethinylestradiol/drospirenone pill, found an increased incidence of VTE across the categories (including pregnant women and non-pregnant, non-CHC-using women) with no difference between progestogen used (Dinger et al 2007). Whilst the large numbers in the study (58,674 women and 142,475 women-years) lend strength to its findings, data from the pharmaceutical industry must always be viewed with some reservation. Of additional interest was the reported three-fold increase in risk for women with a body mass index (BMI) above 30 compared with women with a BMI of 20–25. Table 29.4 shows the comparison of risk estimates of VTE from the two sources.

Table 29.4 Absolute risk of venous thromboembolism (VTE)

Population	VTE per 100,000 women-years
	Committee on Safety of Medicines 1995	EURAS 2005
Non-pregnant, non-user	5	44
Second-generation progestogen COC	15	80^a
Third-generation progestogen COC	25	99
Drospirenone	–	91
Pregnant non-user	60	291

COC, combined oral contraceptive.

^a Levonorgestrel alone.

Current CEU advice to prescribers acknowledges the continuing debate over progestogens, and suggests that individual risk of VTE should be a governing principle in choice of contraceptive method.

Arterial disease, including myocardial infarction (MI) and cerebrovascular accident, results from the (mainly) progestogen-related alteration to lipid profiles together with the oestrogen-associated changes in blood coagulation. A WHO expert group reviewed the data on MI, stroke and hormonal contraception; the conclusions are as follows (World Health Organization Scientific Group 1998).

Acute myocardial infarction

• Acute MI is uncommon in women of reproductive age.

• Women who have hypertension and who take the COC have an increased RR of MI of at least three times that of women who take the COC and are normotensive.

• Smoking increases the risk of MI by 10 times when compared with COC users who do not smoke.

• There is insufficient evidence to allow any conclusion on whether the risk of MI is influenced by the type or dose of progestogen.

There was no increased risk of mortality due to ischaemic heart disease, either during or after COC use, in the report of the RCGP study (Beral et al 1999). However, two recent meta-analyses reported odds ratios of 1.84 and 2.48 for MI with low-dose COC use compared with non-current use [Baillargeon et al (2006) and Khader et al (2003), respectively], casting some doubt on WHO’s conclusion that the risk of MI is not increased in CHC users who do not smoke or have hypertension or diabetes.

Stroke

• Ischaemic and haemorrhagic stroke are both uncommon among women of reproductive age.

• The risk of ischaemic stroke is increased approximately 1.5-fold in pill users who do not smoke and are not hypertensive. In contrast, the risk of haemorrhagic stroke is not increased in these women until they reach 35 years of age, after which the increasing natural risk of haemorrhagic stroke is magnified by CHC use.

• Hypertension increases the risk of both ischaemic (by three times) and haemorrhagic (10-fold) stroke compared with never-users.

• Smoking increases the risk of ischaemic and haemorrhagic stroke (∞2–3) compared with pill users who do not smoke.

• There is insufficient evidence to determine whether the risk of either type of stroke is influenced by the type or dose of progestogen.

Thus the data are reassuring for haemorrhagic stroke, but the risk of ischaemic stroke is increased by COC use; once again, this is supported by the findings of the RCGP study (Beral et al 1999), in which the RR of death from stroke was significantly increased to 1.9 [confidence interval (CI) 1.2–3.1, P = 0.009]. Ten years after discontinuation of the pill, the risk of death from stroke is no longer elevated. For a useful review of hormonal contraception and cardiovascular risk, see Curtis and Marchbanks (2005).

Cancer

Overviews of the risks and benefits of the COC are dominated by breast cancer. Published data are difficult to interpret because pill formulations and patterns of reproduction (particularly age at first pregnancy) have changed with time. In 1996, the Collaborative Group on Hormonal Factors in Breast Cancer reported a meta-analysis of 54 studies involving over 53,000 women with breast cancer and 100,000 control subjects. The group concluded that use of the COC was associated with a small increase in breast cancer, and that the increased risk persisted for 10 years after discontinuation of the pill. The RCGP study (Beral et al 1999) was also reassuring in this respect, since the risk of dying from breast cancer was not significantly increased among current or recent (within 10 years) COC users. The relationship between the pill and breast cancer is difficult to explain because the risk appears to increase soon after exposure, does not increase with duration of exposure and returns to normal 10 years after discontinuation. It has been suggested that starting to use the pill may accelerate the appearance of breast cancer in susceptible women (i.e. late-stage promotion of existing dysplasia). It is also possible that women using the pill have their tumours diagnosed earlier, although it is difficult to explain why a tendency to earlier diagnosis would persist for years after discontinuation. In the large meta-analysis, the risk of breast cancer was also increased among current users of both the progestogen-only pill (RR 1.17) and Depo-provera (RR 1.07), although the number of women using progestin-only methods was small. These findings strengthen the argument for increased detection rather than late-stage promotion of breast cancer. Nonetheless, a biological effect of hormonal contraception has still not been ruled out.

Incident data from a large cohort study from the RCGP (Hannaford et al 2007) has shown significantly lower rates of cancers of the large bowel, rectum, uterine body, ovaries and tumours of unknown origin in ever-users of the COC compared with never-users. There was no material difference between groups for breast cancer, and only small, non-statistically-significant increases in cancers of the lung, cervix and central nervous system. Taken together, there was an absolute rate reduction of any cancer in ever-users of 45 per 100,000 women-years.

Conversely, women taking the COC for more than 8 years did have a significantly increased risk of cervical cancer.

An increase in the risk of squamous carcinoma of the cervix (RR 1.3–1.8) has been recognized for some time. Recent extensive analysis of existing data detected an RR of 1.9 for 5 or more years of COC use (90% CI 1.69–2.31), declining after cessation of use and, like breast cancer, no different from that of never-users at 10 years after discontinuation (International Collaboration of Epidemiological Studies on Cervical Cancer 2007). The relationship is complicated by a number of confounding factors such as patterns of sexual behaviour and the likelihood of having cervical smears. Whether the national introduction in 2008 of vaccination for girls against human papilloma virus types 16 and 18 will have an effect on these figures remains to be seen. An increase in the risk of the much less common cervical adenocarcinoma (RR 2.1, increasing to 4.4 after 12 years of use) has also been demonstrated (Ursin et al 1994).

Other side-effects

Approximately 2% of women become clinically hypertensive after starting the pill. The incidence increases with age and duration of use, obesity and family history. It does not, however, appear to be increased in women with a history of pregnancy-induced hypertension.

An increased risk of gallstones is only significant during the early years of pill use.

A few women develop chloasma on the CHC; both oestrogen and progestogen contribute to this. The chloasma may be slow to fade after the hormones are stopped.

Minor side-effects are most common during the first 3 months of use, and often lead to discontinuation of the method. More common problems include breakthrough bleeding (BTB) or spotting, nausea, breast tenderness, acne and loss of libido. Many of these resolve with time. If side-effects persist after 4 months, it is often worth changing the brand of pill, opting for a lower dose of oestrogen or a different type of progestogen, or changing the mode of delivery. Pills containing antiandrogens are particularly useful for acne.

While the CHC, in general, improves menstrual bleeding patterns, one common reason for presentation to a gynaecologist is BTB. BTB is common during the first three cycles of use. Persistence beyond 3 months may be a result of poor compliance or a coexisting gynaecological disorder such as cervical ectropion (probably more common among COC users), or cervical or uterine polyps. If pelvic examination is normal, it is worth trying a formulation containing a higher dose of oestrogen or a different type of progestogen. After 3 months’ use of a pill containing 50 µg ethinyl oestradiol, the bleeding will often settle and the woman can then resume a lower-dose pill. Alternatively, the vaginal ring may confer improved bleeding patterns. If bleeding persists, try stopping hormonal contraception altogether; if it does not resolve, it should be investigated as intermenstrual bleeding, i.e. by endometrial biopsy and hysteroscopy.

Benefits of combined oral contraception

The CHC confers a number of health benefits. Withdrawal bleeds are usually more regular, lighter and less painful, and the CHC is often the treatment of choice for women with irregular heavy periods (resulting from anovulatory dysfunctional uterine bleeding), premenstrual syndrome, dysmenorrhoea and endometriosis, as ovarian suppression can be achieved without the need for additional contraception (in contrast to danazol). A Cochrane review also concluded that the CHC can improve acne (Arowojolu et al 2009). These benefits can improve compliance significantly (Courtland Robinson et al 1992).

CHC use is associated with a 70% reduction in the risk of endometrial cancer, which may be maintained for up to 15 years after discontinuation (Weiderpass et al 1999). There is a similar duration-dependent reduction in the risk of ovarian cancer (RR 0.64, 95% CI 0.57–0.73) in ever-users compared with never-users (Riman et al 2004), which probably acts through the inhibition of ovulation and lasts for at least 10 years after discontinuation.

Significant numbers of unplanned pregnancies result from women stopping the pill for a ‘break’. There is no evidence that breaks in pill taking reduce the long-term risks. It is not necessary to stop the CHC before planning a pregnancy. There is no evidence of any adverse effect on the fetus of CHC use prior to conception, neither is exposure during pregnancy associated with increased risk of fetal malformation.

Fertility is restored after a delay of 1–3 months, although some women take longer to resume normal cycles. So-called ‘post-pill amenorrhoea’ is almost always associated with cycle irregularity before starting the pill or with coincidental factors associated with secondary amenorrhoea, such as weight loss or stress.

Progestogen-only contraceptives

Progestogen-only contraceptives (POCs) are much less commonly used than CHCs. However, POCs are available in a wider variety of systems, including pills, implants, long-acting injectables and hormone-releasing IUDs.

The mechanism of action of POCs depends on the dose of steroid administered. High doses, such as depot medroxyprogesterone acetate (DMPA), inhibit ovulation. Low doses only inhibit ovulation inconsistently, and the effect varies between individuals. All POCs affect both the quantity and physical characteristics of cervical mucus, reducing sperm penetrability and transport. All have an effect on the endometrium, probably compromising implantation. The recent addition of a 75 µg desogestrel progestogen-only pill (Cerazette^®) combines inhibition of ovulation in 97% of users with the oral route of administration.

The absence of oestrogen in this group of hormonal methods is associated with an absence of cardiovascular risks including VTE. In a case–control study undertaken by WHO (World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception 1998), there was no significant increase in the risk of MI, stroke or VTE associated with use of oral or injectable progestogen-only methods.

The progestogen-only pill (POP or mini-pill) is a useful alternative for women who like the convenience of the COC but for whom oestrogen is contraindicated. Not only does the POP not contain oestrogen, but the dose of progestogen is significantly lower than in equivalent combined preparations. The POP is still a good alternative for women with medical contraindications to the COC, such as migraine and hypertension. It is often advocated for women with diabetes who are at increased risk of cardiovascular disease and who may find that insulin requirements fluctuate with COC cycles. However, the COC is not absolutely contraindicated for diabetic women without long-term complications, whose diabetes is well controlled and for whom pregnancy would be a disaster. The most common indication for POP use in the UK used to be for breastfeeding women, since the POP (unlike the CHC) does not interfere with either the quantity or constituents of breast milk. However, due to its mode of action and likely increased efficacy, the newest POP has wider potential. It is no longer considered necessary to advise women routinely to change from the COC to the POP when they reach the age of 35 years, since the combined pill, in the absence of risk factors for cardiovascular disease, is very safe.

Mode of action

While Cerazette inhibits ovulation in over 90% of cycles, approximately 50% of classical POP users continue to ovulate and menstruate regularly. In these women, the POP works by altering cervical mucus and probably by interfering with implantation. If follicle development is inhibited completely, amenorrhoea results (10–20% of women).

Side-effects

The mode of action of the POP explains its side-effects. Erratic bleeding is the most common cause for discontinuation of the pill; approximately 20% of women will stop using it for this reason.

Follicular growth without ovulation is associated with an increased incidence of functional ovarian cysts. Up to 20% of women using the POP will have a cyst identifiable by ultrasound. Most are symptomless and nearly all resolve spontaneously. A woman found to have a symptomless ovarian cyst should be reviewed after her next menstrual period.

POP use is associated with a slightly increased risk of ectopic pregnancy. Some 2% of pregnancies among POP failures will be ectopic, perhaps as a result of an effect of the progestogen on tubal motility. Since Cerazette inhibits ovulation in almost all cycles, the theoretical risk of ectopic pregnancy is less.

Due to its mechanism of action (50% of women continue to ovulate) and its relatively short half-life (approximately 19 h), the classical POP has a higher failure rate than the COC (see Table 29.3), with users of the established varieties having a 3-h window within which to remember to take their pill. This failure rate is dependent on age and is almost as low as the COC in women aged over 35 years. However, the ovulation-inhibiting Cerazette can be taken in much the same way as the COC with a 12-h window, making it a suitable and reliable choice for younger women as well. There is no direct evidence for advising women who weigh more than 70 kg to take two pills each day.