Data on screening procedures were obtained from the 2000 Juvenile Residential Facilities Census. Data were obtained on the following independent variables: whether all, some, or none of the juveniles were screened; the stage at which screening occured; and by whom the screening was conducted. The dependent variable was whether or not the facility reported a serious suicide attempt, measured by whether the facility transported a resident to a hospital emergency department because of a suicide attempt during a 30-day period preceding the reference day. Multivariate modeling was conducted for 3,690 facilities. Controlling for facility characteristics, the analysis found that screening the entire facility population within the first 24 hours of admission was significantly linked to lower odds of serious suicide attempts. Facilities that screened just some of their residents within 2 to 7 days after admission were determined to have significantly higher odds of serious suicide attempts. These study results held after dividing the sample into configurations based on facility types, size, and population characteristics. 4 tables and 23 references