Fifty-one deaths occurred among 310 patients with systemic lupus erythematosus (SLE) observed for 1,234 patient-years. Twenty-one of 97 entry variables at first clinic visit were associated with an increased risk of mortality. When corrected for multiple comparisons, the only risk factor that retained statistical significance was systolic blood pressure. Each millimeter unit increase in systolic blood pressure corresponded to a 2% increase in mortality risk. Stepwise covariance and recursive partitioning analyses tended to identify nonspecific prognostic variables, rather than the classic serological and diagnostic features of SLE. These data suggest that hypertension is a more significant risk factor for mortality in SLE than the more specific measures of disease severity.