Abstract
Using Data Mining to Explore Complex Clinical Decisions: a Study of Hospitalization After a Suicide Attempt
Baca-Garcia E~~Perez-Rodriguez M~~et al
This study reanalyzed, using data mining techniques, a published study of variables that predicted psychiatrists’ decisions to hospitalize in 509 cases of adult suicide attempters who were assessed in a Spanish emergency department. 5 analyses done by psychiatric researchers using traditional statistical techniques classified 72%-88% of patients correctly. The model developed by researchers with no psychiatric knowledge & employing data mining techniques used 5 variables (drug consumption during the attempt, relief the attempt was not effective, lack of family support, being a housewife, & family history of suicide attempts) & classified 99% of patients correctly. (37 refs.)