Tins investigation was undertaken to bring up to date the records of the Middlesex Hospital, and to attempt to discover whether there was any statistical evidence to support commonly held views on factors affecting prognosis and methods of treatment. 836 cases of carcinoma of the female breast treated during the years 1926-35 were analysed and a smaller group of 265 cases, treated during the years 1936-40, were the subject of a further investigation to discover the effect of more modern methods of treatment.The hopes held at the start of this work have not been completely fulfilled. For such an investigation to give a clear cut statistical answer to all of the controversial questions, careful planning and equal distribution of cases, according to the extent of the disease and type of treatment used, would be necessary. This was not so in this series of cases. As a result the evidence on some points is too inconclusive to give an answer.Careful analysis of results would seem to be essential if methods of treatment are to be guided by fact rather than by impressions. If, in the future, reliable figures on a large scale are to be obtained, co-ordination of policy in treatment and planning ahead in large medical formations will be necessary. Whether this is acceptable or desirable may be open to argument, for it must inevitably mean some acceptance by each individual surgeon of the combined opinion of his colleagues. Only thus, however, can the successful analysis of large series be substituted for the analysis of smaller series for individual surgeons.Despite the difficulties that have arisen owing to the many variations in technique during the years covering this investigation, certain facts that may be of interest and value have emerged. Some previously held views have been supported; others appear to be incorrect.The histological findings in this series have been omitted owing to the impracticability of examining over 1000 sections in the time available. This in itself would make the basis of a separate inquiry. From the statistical point of view therefore it has been assumed that any variations in the histological grouping of the growths have been spread evenly throughout the series.The presentation of results of an investigation such as this is difficult. While the actual detailed figures and tables by which conclusions have been reached are of importance, their inclusion in the text would make reading wearisome. Therefore simplified tables only have been included. * Late Surgical Registrar (demobilized). MAiddlesex Hospital.

No abstract

No abstract

No abstract

No abstract