OBJECTIVE To assess the efficacy, dose-responsiveness and safety of the newer release formulations of 5-aminosalicylic acid (5-ASA) compared to placebo or sulfasalazine (SASP) for the induction of remission in active ulcerative colitis. METHODS A computer-assisted literature search for relevant studies (1981-1998) was performed using MEDLINE, BIOS, the Cochrane Controlled Trials Register and the Science Citation Index, followed by a manual search of reference lists from previously retrieved articles, review articles, symposia proceedings, and abstracts from major gastrointestinal conferences. METHODS Studies were accepted for analysis if they were randomized, double-blinded, and controlled clinical trials of parallel design, with treatment durations of a minimum of four weeks. METHODS Based on an intention to treat principle, the outcomes of interest in the treatment of active disease were the failure to induce global/clinical remission, global/clinical improvement, endoscopic remission, or endoscopic improvement. RESULTS 5-ASA was superior to placebo with regard to all measured outcome variables. For the failure to induce global/clinical improvement or remission, the pooled Peto odds ratio was 0.51 (95% CI, 0.35 to 0.76). A dose-response trend for 5-ASA was also observed. When 5-ASA was compared to SASP, the pooled Peto odds ratio was 0.87 (CI, 0.63 to 1.21) for the failure to induce global/clinical improvement or remission, and 0.66 (CI, 0.42 to 1.04) for the failure to induce endoscopic improvement. SASP was not as well tolerated as 5-ASA. CONCLUSIONS The newer 5-ASA preparations were superior to placebo and tended towards therapeutic benefit over SASP. However, considering their relative costs, a clinical advantage to using the newer 5-ASA preparations in place of SASP appears unlikely.