Program versus time: Length of stay patterns in alcoholism

Abstract

Various studies suggest that, in predicting length of stay (LOS), hospital policy, demographic or pathognomonic factors may be more important than type of treatment given. This investigation compared cumulative discharge curves for 6 sub-groups of male alcoholics ( N = 627): Voluntary versus Committed; Married versus Unmarried; and Age 45 and Under versus Over 45. Actual LOS curves were contrasted with a theoretical curve, representing discharge as a simple function of elapsed time. Sample LOS statistics were also compared with data from an NIMH, nationwide survey. Results showed no significant cumulative LOS differences among the alcoholic sub-groups. During the first week of the program, release rates for all sub-groups greatly exceeded theoretical, then paralleled the theoretical curve until the 10th week of treatment. After the first week of treatment, patients were discharged at a rate similar to that expected if leaving were a simple effect of time elapsed - no program effect on LOS was evident. Comparing the state hospital with the nationwide sample, significantly more state hospital patients left treatment during the first week; but local and national cumulative percentages for 30 days, 60 days and over 60 days were identical.