Abstract

This article reports on the development, implementation, and evaluation of a decision support system (DSS), a hybrid of a statistical model (regression equation) and expert rules, to support this decision. The study focused on the decisions made by social workers, psychologists, and psychiatrists serving as military mental health officers to recommend discharge from compulsory duty in the Israeli army because of mental or emotional difficulties. The process involved three consecutive steps: modeling decisions by statistical analysis of past and current decisions and by eliciting expert rules, developing a hybrid decision support system, and evaluating the DSS through empirical validation and a user survey. The validity of this hybrid DSS was established. The DSS was able to predict 86.5 percent of new decisions mode by mental health officers and 70 percent of past decisions. User acceptance, on the other hand, was low. This article discusses these findings and their implications for practice.