Social Determinants of Preoperative Polysomnography and Surgical Outcomes in Pediatric Tonsillectomy.

OBJECTIVE: To evaluate how sociodemographic and neighborhood factors are associated with use of preoperative polysomnography (PSG) in children undergoing tonsillectomy, and to assess whether these factors are associated with return for postoperative PSG, surgical success, or postoperative complications. METHODS: We analyzed 940 children undergoing tonsillectomy at a tertiary children's hospital (2016-2024). Multivariable logistic regression estimated associations between sociodemographic factors (race, ethnicity, insurance, language), neighborhood opportunity (Child Opportunity Index [COI]), and preoperative PSG. Sequential models examined pathways of association. Doubly robust models estimated the association between PSG and postoperative complications. RESULTS: Preoperative PSG was performed in 389 of 940 children (41.4%). In adjusted models, Black race (adjusted odds ratio [aOR], 1.79; 95% CI, 1.18-2.72), non-English language (aOR, 2.26; 95% CI, 1.45-3.52), Medicaid insurance (aOR, 1.52; 95% CI, 1.06-2.17), and lower COI (aOR per 1-SD increase, 0.81; 95% CI, 0.67-0.99) were associated with higher odds of having preoperative PSG prior to tonsillectomy. Obesity was also strongly associated with preoperative PSG (aOR, 2.49; 95% CI, 1.77-3.51), second only to medical complexity. The association for Hispanic ethnicity attenuated by approximately 80% in sequential models. Sociodemographic and neighborhood factors were not associated with return for postoperative PSG or surgical success; those outcomes were associated with preoperative apnea-hypopnea index (AHI) and medical complexity. In doubly robust models, preoperative PSG was associated with higher complication odds (aOR, 2.07; 95% CI, 1.09-3.93), consistent with confounding by indication rather than a causal effect of testing. CONCLUSION: Social and neighborhood factors were associated with preoperative PSG use but not with surgical success. The critical equity question in pediatric tonsillectomy may lie upstream, in who reaches tertiary surgical evaluation at all. LEVEL OF EVIDENCE: 4.