OBJECTIVE: To determine the risk and timing of a broad range of infective outcomes and mortality after splenectomy.Design, setting and participants: Analysis of a non-identifiable linked hospital discharge administrative dataset for splenectomy cases between July 1998 and December 2006 in Victoria, Australia. MAIN OUTCOME MEASURES: Age, sex, indication for splenectomy, infectious events and death. Patients splenectomised for trauma were compared with patients splenectomised for other indications. Infectious risk was established using Cox proportional hazards models. RESULTS: A total of 2574 patients underwent splenectomy (with 8648 person-years follow-up). Paediatric cases were excluded, leaving 2472 adult cases for analysis. The most common reasons for splenectomy were trauma (635 [25.7%]) and therapeutic haematological indications (583 [23.6%]). After splenectomy, 644 adult patients (26.0%) had a severe infection, with a rate of 8.0 per 100 person-years (95% CI, 7.2-8.4). The risk of severe infection was highest among patients aged >/= 50 years (1.9 per 100 person-years; 95% CI, 1.6-2.7) and those splenectomised for malignancy (14.2 per 100 person-years; 95% CI, 11.8-17.1). Gram-negative infections represented the most frequent causative organism group accounting for 698 (51%) of bacterial pathogens. Staphylococcus aureus was the second most common causative organism. CONCLUSION: The incidence of severe infection and all-cause mortality differed according to age and underlying reason for splenectomy, and was highest among the elderly and those with malignancy, and was lowest among trauma patients. This highlights the need for targeted prevention programs.