Prognosis, the prediction of the outcome of an illness, has long been recognized as a necessary skill of a physician. Hippocrates (translated Chadwick and Mann, 1950) said:

“It seems highly desirable that a physician should pay much attention to prognosis. If he is able to tell his patients when he visits them not only about their past and present symptoms, but also to tell them what is going to happen, as well as to fill in the details they have omitted, he will increase his reputation as a medical practitioner … It is impossible to cure all patients: that would be an achievement surpassing in difficulty even the forecasting of future developments … By realizing and announcing beforehand which patients were going to die he would absolve himself from any blame.”

In addition to the reasons given by our distinguished forebear we now also require to read into the future so that we can make the best use of the limited facilities available for the sick. This is particularly true for elderly psychiatric patients for whom treatment facilities are still very inadequate. It has been suggested (Cook, Dax and Maclay, 1952) that mentally ill elderly people should by preference be admitted to short-stay units attached to general hospitals for the assessment of prognosis and probable response to treatment. They suggested that a 3-4-months stay might be necessary. Experience in a diagnostic outpatient clinic (Norris and Post, 1954) led one of us (F.P.) to believe that it should be possible to classify patients according to the best method of dealing with them in a shorter time even than the six weeks recommended by the B.M.A. Report (1947). The present study was designed to test this belief. Our hopes were only partly fulfilled, but a post hoc analysis of our original data, in conjunction with the known results of the follow-up of cases, has suggested that with more knowledge of the factors influencing prognosis over a short term, much more accurate forecasts could have been achieved. The second half of this paper is devoted to a discussion of these factors. A preliminary report was given at the Third International Congress of Gerontology (Kay et al., 1954).