BACKGROUND: We previously demonstrated that a computerized psychiatric screening interview (the PRIME-MD) can be used in the Emergency Department (ED) waiting room to identify patients with mental illness. In that trial, however, informing the ED physician of the PRIME-MD results did not increase the frequency of psychiatric diagnosis, consultation or referral. We conducted this study to determine whether telling the patient and physician the PRIME-MD result would result in the majority of PRIME-MD-diagnosed patients being directed toward treatment for their mental illness. METHODS: In this single-site RCT, consenting patients with non-specific somatic chief complaints (e.g., fatigue, back pain, etc.) completed the computerized PRIME-MD in the waiting room and were randomly assigned to one of three groups: patient and physician told PRIME-MD results, patient told PRIME-MD results, and neither told PRIME-MD results.The main outcome measure was the percentage of patients with a PRIME-MD diagnosis who received a psychiatric consultation or referral from the ED. RESULTS: 183 (5% of all ED patients) were approached. 123 eligible patients consented to participate, completed the PRIME-MD and were randomized. 95 patients had outcomes recorded. 51 (54%) had a PRIME-MD diagnosis and 8 (16%) of them were given a psychiatric consultation or referral in the ED. While the frequency of consultation or referral increased as the intervention's intensity increased (tell neither = 11% (1/9), tell patient 15% (3/20), tell patient and physician 18% (4/22)), no group came close to the 50% threshold we sought. For this reason, we stopped the trial after an interim analysis. CONCLUSION: Patients willingly completed the PRIME-MD and 54% had a PRIME-MD diagnosis. Unfortunately, at our institution, informing the patient (and physician) of the PRIME-MD results infrequently led to the patient being directed toward care for their psychiatric condition.