You are here

TGA Internet site archive

The content on this page and other TGA archive pages is provided to assist research and may contain references to activities or policies that have no current application. See the full archive disclaimer.

ADEC statement on use of hormone replacement therapy

Related information

Later information

14 August 2003

ADEC has considered several recent publications on the use of hormone replacement therapy (HRT).

These study results do not warrant an urgent change to any woman's replacement therapy. Women taking any form of hormone replacement therapy should, however, again discuss their circumstances with their doctor.

The Million Women Study [Lancet 2003; 362: 419-427]

Results of a very large and well conducted cohort study undertaken in the United Kingdom of women taking hormone replacement therapies have been published.1 The study looked at the experience of more than 1,000,000 women aged 50-64 years and provided information on risk of occurrence of breast cancer.

Overall, this study confirms and enlarges on the results of the Women's Health Initiative (WHI) study that were reported in July 2002.2 Those results dealt only with combination hormone replacement therapy containing oestrogen and progestogen. The United Kingdom study found a risk of breast cancer with combination hormone replacement therapy that was similar to that of the WHI study.

The Million Women Study in the United Kingdom showed that therapy with oestrogen alone, combination oestrogen with progestogen and with the drug tibolone (Livial) all increase the risk of breast cancer in post menopausal women.

The important findings from this study are:

The greatest increase is with combination therapy.

  • The risk increases with duration of use.
  • The increased risk is first observable by 1 year with combination therapy.
  • Ceasing HRT reverses the risk over time.
  • Risk is not limited to use of oral HRT.
  • There was an excess of breast cancer deaths in women who received HRT.

Important new information from the UK Study is that different hormone replacement products and different routes of use (tablets, patches, implants) all increase the risk Similarly there is an increased risk with both continuous and sequential oestrogen/progestogen treatment.

The United Kingdom study found that, if a woman stops taking hormone replacement therapy, the increased risk reverses quickly and is virtually completely lost after five years off therapy.

It was also observed in the UK study that the progestogen component increased the occurrence of breast cancer over oestrogen therapy alone to the extent that this must be weighed against the benefits of endometrial cancer protection. Although combined treatment reduced endometrial cancer, it increased breast cancer. The ADEC intends to consider this issue further and provide additional advice in due course.

Estimated extra cases of breast cancer from HRT use versus no HRT use per 1000 women taking HRT
  After 5 years After 10 years
Oestrogen only HRT 1.2 extra/1000 5/1000
Oestrogen + progestogen HRT 6 extra/1000 19/1000

Update on results from WHI

Other recently reported results from the WHI study showed that combination hormone replacement therapy did not reduce the risks of developing dementia3 or heart disease4 as had been previously thought. In fact, the studies showed that combination hormone replacement therapy is associated with small but important increases in the risk of dementia, stroke, thrombosis, breast cancer, and heart attacks. Moreover, it has been reported that the combination HRT users in the WHI study, who had no or minimal vasomotor menopausal symptoms, had no quality of life benefit compared to the control women, as measured by 12 aspects including energy, sleep and mood.5

In another published paper4 based upon the WHI data, the difference in risk of coronary heart disease (including acute heart attacks) found over the mean 5.2 years follow up of the WHI study was 6 per 10,000 women (39 cases per year in 10,000 women taking combination hormone replacement therapy compared with 33 cases per year per 10,000 women taking placebo). The increase in risk was observable within the first year of therapy.

Other data

On 8 August 2003, Scarabin and others6 reported the results of a case control study conducted during 1999-2002 in France. The investigators recruited 155 consecutive cases with a first documented episode of venous thromboembolism [VTE] of unknown cause (92 with pulmonary embolisms and 63 with deep venous thrombosis), and 381 controls (women admitted to hospital for other reasons). The study suggests that the higher risk of VTE as shown in the WHI study has been further supported and that oral oestrogen replacement given alone or in combination was associated with an elevated risk of VTE.


Hormone replacement therapy is an effective short-term treatment option for controlling the symptoms of the menopause. For each woman considering use of HRT, it is necessary that the benefits be weighed against the several risks that have been observed, including that of coronary heart disease within one year and breast cancer after one year, of therapy. Hormone replacement therapy should not be used for the long-term prevention of disease.

For women currently taking long-term hormone replacement therapy for the treatment of osteoporosis, the risks now documented must be considered when reviewing individual circumstances as well as the consideration of the benefits and risks of alternative therapies.

For younger women with premature menopause or hypogonadism, the benefits of hormone replacement therapy would be expected to be greater and the risks probably smaller than those reported recently in the WHI study and Million Women Study.


  1. Million Women Study Collaborators. Lancet 2003 362: 419-427
  2. Writing Group for the Women's Health Initiative JAMA 2002 288: 321-333
  3. Shumaker, S A, Legault, C et al. JAMA 2003 289:2651-2662
  4. Manson, J E, Hsia, P H J et al. NEJM 2003 349: 523-534
  5. Hays, J, Ockene, J K et al. NEJM 2003 348: 1839-1854
  6. Scarabin, P-Y, Oger, E et al. Lancet 2003 362: 428-432

Top of page