Background Non-alcoholic steatohepatitis (NASH) is a common type of chronic liver disease that can lead to cirrhosis. Obeticholic acid, a farnesoid X receptor agonist, has been shown to improve the histological features of NASH. Here we report results from a planned interim analysis of an ongoing, phase 3 study of obeticholic acid for NASH. Methods In this multicentre, randomised, double-blind, placebo-controlled study, adult patients with definite NASH, non-alcoholic fatty liver disease (NAFLD) activity score of at least 4, and fibrosis stages F2–F3, or F1 with at least one accompanying comorbidity, were randomly assigned using an interactive web response system in a 1:1:1 ratio to receive oral placebo, obeticholic acid 10 mg, or obeticholic acid 25 mg daily. Patients were excluded if cirrhosis, other chronic liver disease, elevated alcohol consumption, or confounding conditions were present. The primary endpoints for the month-18 interim analysis were fibrosis improvement (≥1 stage) with no worsening of NASH, or NASH resolution with no worsening of fibrosis, with the study considered successful if either primary endpoint was met. Primary analyses were done by intention to treat, in patients with fibrosis stage F2–F3 who received at least one dose of treatment and reached, or would have reached, the month 18 visit by the prespecified interim analysis cutoff date. The study also evaluated other histological and biochemical markers of NASH and fibrosis, and safety. This study is ongoing, and registered with ClinicalTrials.gov, NCT02548351, and EudraCT, 20150-025601-6. Findings Between Dec 9, 2015, and Oct 26, 2018, 1968 patients with stage F1–F3 fibrosis were enrolled and received at least one dose of study treatment; 931 patients with stage F2–F3 fibrosis were included in the primary analysis (311 in the placebo group, 312 in the obeticholic acid 10 mg group, and 308 in the obeticholic acid 25 mg group). The fibrosis improvement endpoint was achieved by 37 (12%) patients in the placebo group, 55 (18%) in the obeticholic acid 10 mg group (p=0·045), and 71 (23%) in the obeticholic acid 25 mg group (p=0·0002). The NASH resolution endpoint was not met (25 [8%] patients in the placebo group, 35 [11%] in the obeticholic acid 10 mg group [p=0·18], and 36 [12%] in the obeticholic acid 25 mg group [p=0·13]). In the safety population (1968 patients with fibrosis stages F1–F3), the most common adverse event was pruritus (123 [19%] in the placebo group, 183 [28%] in the obeticholic acid 10 mg group, and 336 [51%] in the obeticholic acid 25 mg group); incidence was generally mild to moderate in severity. The overall safety profile was similar to that in previous studies, and incidence of serious adverse events was similar across treatment groups (75 [11%] patients in the placebo group, 72 [11%] in the obeticholic acid 10 mg group, and 93 [14%] in the obeticholic acid 25 mg group). Interpretation Obeticholic acid 25 mg significantly improved fibrosis and key components of NASH disease activity among patients with NASH. The results from this planned interim analysis show clinically significant histological improvement that is reasonably likely to predict clinical benefit. This study is ongoing to assess clinical outcomes. Funding Intercept Pharmaceuticals.