The early detection of depression and suicidal ideation is essential for the prevention of suicide in the community. This study therefore aimed to develop a brief screen for depression and suicidal ideation that can be easily administered in primary-care settings. The Self-rating Depression Scale (SDS), the Hospital Anxiety and Depression Scale (HADS) and a modified version of the Composite International Diagnostic Interview (CIDI) were administered by 353 residents of a single community aged 64 years and over. A five-item screen was derived from SDS and HADS, using CIDI as the external criterion (study 1). The scale was modified so that it was more appropriate for our use, and was labeled the Depression and Suicide Screen (DSS). Its validity and reliability were examined among a further 382 residents of the same community aged 64 years and over, using the Short-Form Geriatric Depression Scale (GDS-S) as the external criterion (study 2). The DSS was internally consistent (Cronbach's alpha=0.62). Its reliability in detecting depression (defined as >or= 6 in GDS-S) and suicidal ideation (screened out by the inquiry by our trained staff) was 0.768 and 0.721, respectively. For depression, the sensitivity was 0.705; specificity, 0.729; positive predictive value (PPV), 0.446; negative predictive value (NPV), 0.888; and the overall diagnostic power, 0.723. For suicidal ideation, its sensitivity was 0.698; specificity, 0.693; PPV, 0.317; NPV, 0.926; and overall diagnostic power, 0.694. The DSS demonstrated a reasonable level of sensitivity and specificity in identifying both depression and suicidal ideation among the elderly within a community.