The main purpose of this study was to evaluate the effectiveness of the brief version of the World Health Organization Quality of Life (WHOQOL-BREF) questionnaire as a predictor of mortality amongst the elderly. A total of 689 male residents of veteran homes, all above the age of 65 years, were randomly selected in 2001. The Taiwan version of the WHOQOL-BREF was administered as the baseline, with each of these subjects being interviewed and subsequently followed up for mortality until the end of 2003. Data on self-reported global health, life satisfaction, medical status, physical performance and health behavior was also collected. Following the death of 105 of the 689 subjects during the 2-year follow-up period, the relative risk (RR) of death was subsequently assessed using Cox's proportional hazard regression analyses. After adjusting for other predictors (age group, chronic diseases, emergency visits, hospitalization, physical performance, regular exercise, self-reported global health and life satisfaction), almost all of the WHOQOL-BREF items and domains failed to predict mortality; the one exception was working capacity levels (score 1-2 vs. score 4-5) which, after adjusting for other predictors, did demonstrate the ability to predict mortality (RR = 1.96, p < 0.05).