Smoking is a known risk factor for inferior health outcomes. Retrospective analyses of large datasets may assess whether such risk is manifested or mitigated in clinical practice. Although many risk factor analyses use the National Surgical Quality Improvement Program (NSQIP) and National (Nationwide) Inpatient Sample (NIS), such investigations have seldom been directly compared.NIS and NSQIP datasets were used. Primary outcome measures were perioperative complications. NSQIP complications were stratified based on occurrence before versus after discharge. Multiple logistic regression was employed in adjusted analyses.Among NSQIP (N = 56,145) and NIS (N = 1,311,426) patients, 24.0% and 31.8% were identified as tobacco users, respectively. Before discharge, NSQIP smokers had increased odds of pneumonia (adjusted odds ratio [aOR] = 1.43), postoperative intubation (aOR = 1.40), and sepsis (aOR = 1.71) and decreased odds of pulmonary embolism (aOR = 0.57) (all P < 0.05). After discharge, NSQIP smokers exhibited increased odds of cardiac arrest (aOR = 2.53) and surgical site infection (aOR = 1.25) and decreased odds of urinary tract infection (aOR = 0.68) and deep venous thrombosis (aOR = 0.61) (all P < 0.05). In adjusted analysis of NIS data, tobacco users exhibited increased odds of inpatient pneumonia (aOR = 1.57), myocardial infarction (aOR = 1.29), and postoperative intubation (aOR = 1.15) and decreased odds of pulmonary embolism (aOR = 0.80) and deep venous thrombosis (aOR = 0.78) (all P < 0.05).Patients with a history of tobacco use undergoing spinal fusion were at increased risk of pneumonia and intubation but decreased risk of deep venous thrombosis and pulmonary embolism during the inpatient postoperative stay. After discharge, smokers exhibited increased odds of cardiac arrest and surgical site infection. NIS and NSQIP results were similar, with 9 of 12 outcomes exhibiting identical inpatient conclusions. Qualitative comparison of NIS and NSQIP yields results that offer increased validity over single-source studies.
