How can it be that recent studies are so at odds with previous ones? The early research into HRT is beset by a number of problems, including:
* a lack of control groups, making it difficult to see the effects of HRT compared with non-users
* selection of healthy women. The women in HRT trials are usually of a higher socioeconomic status. They also tend to be more physically active, to have a more stable blood pressure and are leaner than the average 'real-world' user. This results in the so-called 'healthy-user' effect, where such users are less vulnerable to the adverse effects of HRT
* short-term trials. Women were followed for arbitrary amounts of time, providing no data on the long-term risks and death rates
* no placebo for comparison. A placebo can cure 30 to 90 per cent of problems. If HRT can't beat a placebo, it's probably not much good
* poor reporting of adverse effects. Often, women who suffer adverse side-effects are taken out of a trial early. Other adverse effects deemed to be not directly related to the primary outcome being looked at tend to be dismissed as unimportant
* different definitions of HRT. In early trials, 'HRT' was any kind of hormone supplement given at any dose and in a variety of forms. We now know the route of entry into the body and the dosages can significantly affect the outcome.