There is general agreement among those who treat mental illness that psychiatric hospitals should be (a) small, and (b) highly staffed.3 Two previous sources of data provide numerical support for the contention that smaller size and higher staffpatient ratio are related to higher rates of favorable outcome.
The first previous set of data was computed by one of us (LU) from material presented by Bockoven1 on "Moral Treatment in American Psychiatry." For data concerning operation of Worcester State Hospital over 12 consecutive decades, rankorder correlations were 0.76 (P<0.01) between per cent of admissions released and physicians per patient and –0.89 (P < 0.001) between per cent of admissions released and size of patient population. Jenkins,2 using selected psychotic admissions at 12 hospitals, obtained a 0.64 correlation (P<0.05) between early release and per diem expense (reflecting overall staffing) and a