Abstract Background. Venous thromboembolism (VTE) is a frequent surgical complication. The type of surgery and baseline patient VTE risk factors are important in defining postoperative VTE risk. The 2004 American College of Chest Physicians (ACCP) recommends implementation of pharmacologic thromboprophylaxis according to surgery type and VTE risk factors. We conducted a retrospective cohort study of surgical patients to determine adherence to the 2004 ACCP surgical thromboprophylaxis recommendations and to assess for predictors of non-adherence. Methods. Using data from the administrative healthcare database CIRESSS (Centre Informatisé de Recherche Évaluative en Services en Soins de Santé), in patients admitted for surgery between January 1st and December 31st 2006 at the Centre Hospitalier Universitaire de Sherbrooke (CHUS), a tertiary care center in Sherbrooke, Quebec, Canada, we retrospectively assembled a cohort of all consecutively admitted surgical patients who met ACCP criteria for pharmacologic thromboprophylaxis. We assessed the proportion of patients who received in the 24 hours pre- and post-surgery a prescription for thromboprophylaxis. We used conditional logistic regression to determine clinical characteristics associated with an absent prescription. The incidence of objectively-defined symptomatic postoperative VTE was assessed at three months. Results. Of 2286 surgical admissions that met ACCP recommendations for pharmacologic thromboprophylaxis, 1852 (81%) received thromboprophylaxis and 434 (19%) did not. Male sex (odds ratio (OR): 1.9, 95% confidence interval (CI) (1.4–2.4)), age <40 years (OR: 1.8, 95% CI (1.2–2.9)), absent varicose veins (OR: 3.2, 95% CI (1.1–10.0)), pregnancy within 3 months of surgery (OR: 8.5, 95% CI (3.5–20.8)), a moderate risk for post-operative VTE as per ACCP criteria (OR: 4.4, 95% CI (2.9–6.52)), non-major surgery (OR: 4.5, 95% CI (2.3–8.8)), short-duration hospitalization (2–3 days vs. 13 days or more) (OR: 15.3, 95% CI (9.6–24.4)), absence of active cancer (OR: 1.6, 95% CI (1.2–2.1)), absence of heart failure (OR: 1.8, 95% CI (1.1–2.8)), and non-orthopaedic surgery (OR: 28.2, 95% CI (15.9–49.9)) were associated with lack of pharmacologic thromboprophylaxis. At three months following surgery, 16 patients (0.7%) developed VTE, and the adjusted relative risk of VTE in patients without thromboprophylaxis was 2.1 (95% CI:0.5–7.9). Conclusions. Though surgical thromboprophylaxis was observed in 81% of surgical patients, 19% of patients who met criteria for ACCP thromboprophylaxis did not receive prophylaxis. Targeted recommendations in particular toward pregnant women undergoing non-obstetrical surgery, patients with short duration hospitalization, and patients undergoing non-orthopaedic surgery may ameliorate thromboprophylaxis compliance rates, which in turn may impact on post-operative VTE risk.