GPs don't have skills for echocardiography
The US pooled study (News, October 25) claims that HRT reduces mortality in women under 60. Not a single one of the 30 randomised controlled studies found a significant effect of HRT on total mortality.
The claimed 39 per cent reduction in mortality in the 17 studies of younger women was achieved by weighting. Absurdly, a single small study was weighted at 41 per cent. Some 32 of 62 oestrogen-treated ovarian cancer survivors died during this study.
Several studies have found HRT increases the risk of fatal ovarian cancer. The largest study listed, PEPI 1995, recorded a 75 per cent increase in mortality among 701 treated women, but was weighted at only 2.3 per cent.
This widely publicised meta analysis is yet another example of dangerous statistical manipulation. Oestrogens and progesterones are immunosuppressive, carcinogenic, psychoactive and vasoactive. Fortunately worrying side-effects ensure many women discontinue before becomingly fatally ill, but breast cancer still causes death 20 years after diagnosis.
Dr Ellen Grant