The use of attainment scaling in the evaluation of a regional mental health program.

The general evaluation of the success of a regional mental health program was attempted through the use of the Goal Attainment Scale. There were 278 goal guides scored for 96 sample patients at periods of 6, 12, and 18 months after admission. The interrater reliability between pairs of raters at follow-up interviews was high (.87). The intraclass correlation for goal guides constructed by three independent goal setters was moderate (r=.71). The mean scores of the three raters, when compared to a consensus goal score, attained an intraclass coefficient of .91. Thus the goal-attainment scoring procedure was found to be reliable. However, no evidence could be discerned to support its validity as a measure of a patient's treatment outcome.