Survival rates for cancer patients may be improved by treatment in specialised cancer centres, according to Cochrane researchers. In a review of recent studies, they found that women diagnosed with gynaecological cancer lived longer when treated in specialist compared to non-specialist units. In the past, cancer patients were often treated by non-specialist surgeons and hospitals. This is changing and in developed countries, most cancer care is now organised into networks of specialised centres, with on-site experts and specialised nursing staff. This centralised approach, although costly, may help improve outcomes for patients. The review focused on data from five studies, altogether involving more than 62,000 women who were treated from the late 1990s onwards. Over 48,000 of the participants were involved in one study carried out in 2009, whilst the smallest study involved just 250. Due to clinical differences between the studies, the researchers combined the data in different ways. Incorporating three studies and over 50,000 women with gynaecological cancer, one analysis showed that women treated at teaching centres or regional cancer centres lived longer compared to those treated at community or general hospitals. An alternative combination of three of the studies, involving over 9,000 women with ovarian cancer, showed that those treated at institutions with on-site gynaecologists lived longer compared to those treated at community or general hospitals. \"We found consistent evidence for specialist treatment prolonging survival rates in women with gynaecological cancer,\" said lead researcher, Professor Yin Ling Woo of the University of Malaya Cancer Research Institute, University of Malaya, in Kuala Lumpur, Malaysia. \"The effect seemed to be strongest for ovarian cancer, although most evidence came from developed countries.\" The researchers estimate that survival rates could be improved by around 10%, although uncertainties in the data mean the actual figure could range between 1% and 18%. Better designed studies are needed to confirm the results. All five studies included in the review used electronic records to identify patients after they had received treatment, meaning it was difficult to be sure that the centralised and de-centralised treatment groups were similar. \"Ideally, women should be allocated to specialist and non-specialist treatment groups in advance to ensure that there is no bias in the data,\" said Professor Woo. \"These higher quality trials are needed to assess whether the extra cost is worthwhile, especially as many countries have limited resources for specialist care.\"