BACKGROUND, The authors examined the reliability and validity of the Schedule of Attitudes toward Hastened Death (SAPID), a self-report measure of desire for death previously validated in a population of individuals with the acquired immunodeficiency syndrome (AIDS), among terminally ill patients with cancer. METHODS. The authors interviewed 92 terminally ill cancer patients, all with a life expectancy of < 6 months, after admission to a palliative care hospital. Patients were administered the SAPID, a clinician-rated measure of desire for death (the Desire for Death Rating Scale [DDRS]), and several measures of physical and psychosocial well-being. RESULTS. The average number of SAPID items endorsed was 4.76 (standard deviation, 4.3); 15 patients (16.3%) endorsed greater than or equal to 10 items, indicating a high desire for death. Internal consistency was strong (coefficient alpha = 0.88, median item-total correlation = 0.49), as were indices of convergent validity. Total SAPID scores were correlated significantly (correlation coefficient [r] = 0.67) with the DDRS, and somewhat less so with measures of depression (r = 0.49) and hopelessness (r = 0.55). Lower, but substantial, correlations were observed between the SAPID and measures of spiritual well-being (r = -0.42), quality of life (r = -0.36), physical symptoms (r = 0.38), and symptom distress (r = 0.38). No significant correlation was observed between SAPID scores and social support (r = -0.06) or pain intensity (r = 0.16); however, pain-related functional interference and overall physical functioning were correlated significantly with SAPID scores (r = 0.31 and r = -0.23, respectively). CONCLUSIONS. The SAPID appears to be a reliable and valid measure of desire for death among terminally ill cancer patients. Coupled with previous research in patients with AIDS, these results support the utility of the SAHD for research addressing interest in hastened death in patients with a life-threatening medical illness. (C) 2000 American Cancer Society.