A to X: the problem of categorisation of drugs in pregnancy — an Australian perspective

Given that over 80% of women use at least one prescribed or over-the-counter medication (typically one to three) at some time during their pregnancy, most medical practitioners who treat women of reproductive age can expect frequent questions about the use of medications during pregnancy and breastfeeding.1 In addition, as the average age of women having babies increases, their likelihood of having medical disorders that complicate pregnancy (such as hypertension) or chronic conditions also increases.

A recent Australian study found that 322 of 819 pregnant women (39.3%) reported a chronic health condition during pregnancy, the most common being asthma, blood-related disorders (eg, thrombosis, haemorrhage and/or anaemia) and diabetes.2 Of concern, 107 out of 181 of those who reported a chronic health condition and use of regular prescribed medication (59.1%) reported non-adherence to medication for a number of different reasons. Most of the participants had “some concerns” about using any medicine during pregnancy and almost one-third believed that natural remedies were safer than other medicines during pregnancy.

Complementary medicines include nutritional supplements, vitamin and mineral preparations, and herbal, aromatherapy and homoeopathy products, and an estimated $1.3 billion was spent by Australians on such products in 2004.3 In an Australian study completed between 2005 and 2007, about one-third of pregnant women reported taking complementary and alternative therapies to treat common complaints during pregnancy (eg, vitamins and minerals to treat colds and leg cramps, yoga and aromatherapy).4 Complementary medicines are included on the Australian Register of Therapeutic Goods as listed (low-risk) or registered (higher-risk) medicines, but most complementary medicine products do not carry a pregnancy risk category, even though some may have significant effects on pregnancy or fetal development.

Since the 1960s and the birth of babies in Australia and Europe with severe birth defects following early pregnancy exposure to thalidomide, there has been a general reluctance on the part of doctors to prescribe, and women to take, medications during pregnancy because of fears about potential teratogenic effects. In 1963, as a direct consequence of the thalidomide tragedy, the Commonwealth Department of Health established the Australian Drug Evaluation Committee (ADEC) as an independent committee to advise on the safety of new drugs being introduced into Australia and to monitor and evaluate potential adverse effects of drugs already in use. In 2010, ADEC was replaced by the Advisory Committee on Prescription Medicines.

An ad hoc working party of ADEC published a unique Australian categorisation of the risk of drugs in pregnancy (Categories A, B1, B2, B3, C, D and X). In the United States, the Food and Drug Administration (FDA) adopted a labelling format for safety of drugs in human pregnancy in 1979, and other countries developed their own categorisations. The ADEC categorisation is similar to the Swedish system and, although it shares the same letters as the FDA categorisation, there are some notable differences (particularly the Australian B1, B2 and B3 categories).

The first Australian Medicines in pregnancy booklet was published in 1989. The last hard copy (fourth) edition, retitled Prescribing medicines in pregnancy, was published by the Therapeutics Goods Administration (TGA) in 1999.5 This resource subsequently went online, and the currently available Prescribing medicines in pregnancy database was revamped in May 2011.6 Unfortunately, the content of the data on drug safety in pregnancy and reliance on the categories did not change.

A recent audit of practice of 80 general practitioners and 50 pharmacists by MotherSafe (a counselling service for women and health care providers who are concerned about exposures during pregnancy and breastfeeding) found that both groups were generally very conservative regarding advice about medications during pregnancy and breastfeeding, relying heavily on the ADEC categorisation and company product information listed in MIMS (the Monthly Index of Medical Specialties).7 The survey found that 80% of pharmacists and GPs used MIMS (ie, the categorisations and product information, where pregnancy and lactation are almost without exception included under special precautions or contraindications) as their primary source of information on medication safety in pregnancy. While 25% of pharmacists used the Australian medicines handbook, only 4% of pharmacists and 2% of GPs used the TGA’s Prescribing medicines in pregnancy booklet.

Almost half of Australian drugs fall into one of the B categories because of the paucity of available human pregnancy data. Undoubtedly the most problematic of the Australian categories is B3 — limited use by pregnant women and women of childbearing age, without an increase in the frequency of harmful effects on the fetus, but with animal data showing increased occurrence of fetal damage. If the definition of the B3 category were explained to an average pregnant woman, she would probably never take a B3 drug because of anxiety about what this could mean for her baby. However, the data regarding many B3 drugs (such as combinations of long-acting β-agonists plus inhaled corticosteroids) are limited but reassuring, and uncontrolled asthma in pregnancy is a significantly greater risk than the potential risks (derived from animal data) posed by these medications.8,9

The biggest problem of the ADEC system is its alphabetical nature, which implies (incorrectly) that there is a gradation of risk, with the B category being “worse” than the A category, and so on. Confusingly, some publications that list the categories alphabetically (eg, MIMS) add a note explaining that “allocation of a B category does not imply greater safety than the C category” — counterintuitive and distinctly unhelpful to say the least.

Unfortunately, the apparent simplicity of the categories means that clinicians tend to use it as a “bible”, rather than as a guide, which can result in misinterpretation of risk. The ready availability of the categories means that practitioners would rarely critically assess the quality or content of the original studies on which the categorisation was based and, therefore, would not consider the complexities involved in balancing the risks and benefits of using or not using a particular drug for a specified indication at a certain stage in pregnancy.

There is also an assumption that drugs in the same category carry a similar risk, which is often erroneous. For example, valproate and paroxetine are both in the D category, but the former is associated with a significantly increased risk of birth defects and neurodevelopmental sequelae following use in the first trimester, whereas there are conflicting data about paroxetine being associated with a slightly increased risk of cardiac and other defects.

Also, the categories do not take the stage of pregnancy into account. For example, tetracyclines cause tooth discoloration in the fetus when taken after 14 weeks’ gestation, so being categorised D in the first trimester is misleading and unnecessarily worrying. In addition, the categories do not adequately differentiate between different pregnancy situations — planned versus unplanned pregnancy and essential versus non-essential medications. Furthermore, the categories rarely take the dose or route of exposure into account — topical and inhaled exposures are generally less concerning than oral or intravenous exposures as they result in less systemic absorption, lower maternal serum concentrations, and thus minimal transplacental passage and a negligible chance of affecting the embryo.

The categories are assigned before drugs are marketed and are often based solely on the results of animal reproductive studies, owing to a general paucity of human pregnancy data. The categories are rarely changed despite new (and often reassuring) evidence because of a general reluctance to advocate the safety of drugs in pregnancy.

Other limitations of the ADEC categorisation include the fact that it does not cover environmental agents, chemicals, infectious agents, illicit drugs or complementary medicines and that, contrary to the understanding of many medical practitioners, the categories are not applicable to breastfeeding.

In general, recommendations given to women about medications in pregnancy are cautious at best and scaremongering and inappropriate at worst. Unfortunately, the advice given by health care providers is compounded by misleading information obtained from the internet and other lay sources. Misleading advice, often based largely on the ADEC categorisations, can result in significant consequences for both the mother and baby. Some women stop taking essential medication because of fears about fetal safety, thus putting themselves and their baby at risk of an untreated illness (which is often a higher risk than the potential risk posed by the medication). Other women have their medication switched, on misunderstood grounds of fetal safety, from those that are beneficial to those with unknown or less efficacy — for example, switching antidepressants from citalopram (C category) to mirtazapine (B3 category).

Equally concerning is that, when pregnant women are told that they should stop their effective, prescribed medications (eg, antidepressants) because of their categorisation, some may self-medicate with complementary medicines (about which there are usually less pregnancy safety and efficacy data) or potentially more harmful substances such as alcohol, cigarettes or illicit drugs. Worse still, some women consider terminating otherwise wanted pregnancies because of perceived safety concerns based on a drug’s categorisation.

In a group of 177 of callers to MotherSafe (from 2005 to 2007) who had been considering termination of a pregnancy because of a range of exposures (including drugs, radiation and vaccines), only 30% had been exposed to major teratogens such as retinoids and antiepileptic drugs. Most had received information from other sources (including health care providers and the internet), which had caused them such anxiety that they were considering terminating their pregnancy.10

For several years, concerns have been voiced about the appropriateness of the current system of labelling of drugs for safety in pregnancy in the US.11 In 2008, the FDA proposed major revisions to the labelling of prescription drugs because of these concerns. All labels would include a general statement about the background risk of birth defects for all pregnancies as well as information about whether there was an active pregnancy exposure registry for that particular agent and, if so, how to enrol in it. The new labelling would: remove the letter categorisations; present information regarding fetal risks in a more narrative style; put discussions about risk in the contexts of background risk of birth defects and drug indication; document how the risk was determined (eg, extrapolated from animal data or obtained from human data); and include information (if available) about drug dosing in pregnancy. New and relevant information would be incorporated as it became available.12

As of late 2011, the FDA’s final rule on pregnancy and lactation labelling was in the final writing and clearance process; this promises an exciting new era in the field of rational use of medicine in this important population group. Given that the FDA review has been ongoing for the past 3 years, and it will be several more years before any objective evaluation of the new US system can be made, it may be a long time before Australian regulatory authorities look at the issue of improved labelling of drugs for safety in pregnancy or implement change.

I hope that this article will stimulate debate about the direction that Australian regulatory authorities should take to develop an improved, more rational approach to labelling of drugs for safety in pregnancy and breastfeeding in the near future.