Pregnant women may take non-steroidal anti-inflammatory drugs (NSAIDs), selective cyclooxygenase (COX)-2 inhibitors or biologic agents to relieve symptoms or manage disease flares in late pregnancy. We aimed to quantify the risk of prematurity associated with late pregnancy exposure to non-selective NSAIDs, selective COX-2 inhibitors and biologic agents. Using data from Quebec Pregnancy Cohort, we performed a population-based cohort study. We included all women who were covered by the Quebec Drug Plan and had a singleton livebirth between 1/1/1998 and 31/12/2009. Late pregnancy exposure was defined as having filled at least one prescription for non-selective NSAIDs, selective COX-2 inhibitors or biologic agents in the 3 months before delivery. Prematurity was defined as <37 weeks of gestation. Crude and adjusted odds ratios (OR) were obtained using generalized estimation equation models. Covariates included maternal autoimmune diseases, demographics, concomitant drug use, history of pregnancy complications, and other comorbidities. A total of 156,531 pregnancies met inclusion criteria and were considered for analyses. In the 3 months before delivery, 391 pregnancies were exposed to non-selective NSAIDs, 55 to COX-2 inhibitors, and 12 to biologic agents. After adjustment for maternal autoimmune diseases, concomitant medication use, and other risk factors, COX-2 inhibitor use in late pregnancy was associated with a 2.46 fold increased risk of prematurity (adjusted OR, 2.46; 95% confidence interval, 1.28-4.72) compared to non-use; only late pregnancy exposure to celecoxib was found to increase the risk (adjusted OR, 3.41; 95% confidence interval, 1.29-9.02). In conclusion, celecoxib use during late pregnancy may increase the risk of prematurity.