Objective Survival of patients with systemic lupus erythematosus (SLE) has improved significantly, but new morbidities have emerged, leading to altered patterns of outcome in this disease. We examined changes in mortality and other outcomes over time in a large SLE cohort. Methods A group of 1241 patients from the University of Toronto Lupus Clinic followed prospectively were divided into 4 entry cohorts — 1: 1970–1978, 2: 1979–1987, 3: 1988–1996, 4: 1997–2005. These cohorts were followed through four 9-year calendar periods defined over the same intervals. Both cohort and calendar effects were assessed for the following outcomes: mortality (standardized mortality ratio; SMR), disease activity over time (adjusted mean SLEDAI; AMS), cumulative damage (Systemic Lupus International Collaborating Clinics Damage Index; SDI), coronary artery disease (CAD), and osteonecrosis (ON). Cox regression models were used to further investigate mortality and the influence on it of the disease-related factors. Results Over the 36-year period of the study, 211 deaths occurred. The overall SMR in the first and last decades were 12.60 (95% CI: 9.13, 17.39) and 3.46 (95% CI: 2.71, 4.40) respectively. When SMR were stratified by the entry cohort and calendar period, there is evidence of a calendar-period effect but no cohort effect. The AMS decreased over the decades, while SDI, CAD, and ON increased. There were significant detrimental effects for male sex, CAD, AMS, SDI, and use of immunosuppressive drugs and significant protective effects for use of antimalarials and the effect of calendar period on mortality. Conclusion Our study demonstrates improved survival in patients with SLE over a 36-year period. Disease-related variables included in the model are important factors for mortality in this SLE cohort, but could not completely explain the trend of improved survival over calendar period observed.