OBJECTIVE To examine the effect of the introduction of laparoscopic cholecystectomy (LC) on patterns of practice (number of cholecystectomy procedures, case-mix and length of hospital stay) and patient outcomes in Ontario. METHODS Cross-sectional population-based time trends using hospital discharge data. METHODS All acute care hospitals in Ontario where cholecystectomy was provided. METHODS All 119,821 Ontario residents who underwent cholecystectomy between 1989-90 and 1993-94. After exclusions (initial bile duct exploration, cancer, incidental cholecystectomy, or missing codes for age, sex or residence) 108,442 patients remained. METHODS Number of cholecystectomy procedures, proportion of patients with acute or chronic gallstone disease, length of hospital stay, and rates of death, readmission, and bile duct injury and other in-hospital complications after cholecystectomy by year. RESULTS The number of cholecystectomy procedures increased by 30.4% between 1989-90 and 1993-94. The number of patients with chronic gallstone disease increased by 33.6%, and the number who underwent elective surgery increased by 48.3%. The proportion of procedures performed as LC increased from 1.0% in 1990-91 to 85.6% in 1993-94. Patients who received LC tended to be younger female patients with chronic gallstone disease with no coexisting conditions undergoing elective operations. The mean length of stay, adjusted for case-mix differences, was significantly lower in 1993-94 than in 1989-90 (2.6 days v. 7.5 days) (p < 0.05); the values for LC and open cholecystectomy in 1993-94 were 1.8 days and 7.3 days respectively. The decrease in the crude death rate over the study period (0.3% to 0.2%) was not significant (relative odds 1.10, 95% confidence interval [CI] 0.72 to 1.69). In 1993-94 the adjusted risk of readmission to hospital within 30 days was 1.38 (95% CI 1.19 to 1.58) as compared with 1989-90. Over the 5 years the rate of bile duct injuries tripled (0.3% in 1989-90 v. 0.9% in 1993-94). The adjusted risk of having at least one complication after cholecystectomy in 1993-94 was 1.90 (95% CI 1.75 to 2.07) as compared with 1989-90. CONCLUSIONS LC has had a substantial effect on the number of cholecystectomy procedures performed, the type of patient having the gallbladder removed and the length of hospital stay. Death rates are unchanged, but the odds of readmission and in-hospital complications are both increased. Future research should be directed toward determining the reasons for the overall increase in rates, developing methods to reduce bile duct injuries and identifying other relevant outcomes, such as patient satisfaction with the procedure.