OBJECTIVE Previous Australian studies reported that postoperative pulmonary complications affect 13% of patients undergoing upper abdominal laparotomy. This study measured the incidence of postoperative pulmonary complications, risk factors for the diagnosis of postoperative pulmonary complications and barriers to physiotherapy mobilisation in a cohort of patients undergoing high-risk abdominal surgery. METHODS Prospective, observational cohort study. METHODS Two surgical wards in a tertiary Australian hospital. METHODS Seventy-two patients undergoing high-risk abdominal surgery (participants in a larger trial evaluating a novel model of medical co-management). METHODS Incidence of, and risk factors for, postoperative pulmonary complications, barriers to mobilisation and length of stay. RESULTS The incidence of postoperative pulmonary complications was 39%. Incision type and time to mobilise away from the bed were independently associated with a diagnosis of postoperative pulmonary complications. Patients were 3.0 (95% confidence interval 1.2 to 8.0) times more likely to develop a postoperative pulmonary complication for each postoperative day they did not mobilise away from the bed. Fifty-two percent of patients had a barrier to mobilisation away from the bed on the first postoperative day, with the most common barrier being hypotension, although cessation criteria were not defined objectively by physiotherapists. Development of a postoperative pulmonary complication increased median hospital length of stay (16 vs 13 days; P=0.046). CONCLUSIONS This study demonstrated an association between delayed postoperative mobilisation and postoperative pulmonary complications. Randomised controlled trials are required to test the role of early mobilisation in preventing postoperative pulmonary complications in patients undergoing high-risk upper abdominal surgery.