Background: Pediatric adenotonsillectomies are common and carry known risks of potentially severe complications. Complications that require a revisit, to either the emergency department or hospital readmission, increase costs and may be tied to lower reimbursements by federal programs. In 2011 and 2012, recommendations by pediatric and surgical organizations regarding selection of candidates for ambulatory procedures were issued. We hypothesized that guideline-associated changes in practice patterns would lower the odds of revisits. The primary objective of this study was to assess whether the odds of a complication-related revisit decreased after publication of guidelines after accounting for preintervention temporal trends and levels. The secondary objective was to determine whether temporal associations existed between guideline publication and characteristics of the ambulatory surgical population.
Methods: This study employs an interrupted time series design to evaluate the longitudinal effects of clinical guidelines on revisits. The outcome was defined as revisits after ambulatory tonsillectomy for privately insured patients. Data were sourced from the Truven Health Analytics MarketScan database, 2008-2015. Revisits were defined by the most prevalent complication types: hemorrhage, dehydration, pain, nausea, respiratory problem, infection, and fever. Time periods were defined by surgeries before, between, and after guidelines publication. Unadjusted odds ratios estimated associations between revisits and clinical covariates. Multivariable logistic regression was used to estimate the impact of guidelines on revisits. Differences in revisit trends among pre-, peri-, and postguideline periods were tested using the Wald test. Results were statistically significant at P < .005.
Results: A total of 326,993 surgeries met study criteria. The absolute revisit rate increased over time, from 5.9% (95% confidence interval [CI], 5.8-6.0) to 6.7% (95% CI, 6.6-6.9). The proportion of young children declined slightly, from 6.4% to 5.9% (P < .001). The proportion of patients having a tonsillectomy in an ambulatory surgery center increased (16.5%-31%; P < .001), as did the prevalence of obstructive sleep apnea (7.0%-14.0%; P < .001) and sleep-disordered breathing (20.6%-35.0%; P < .001). In a multivariable logistic regression model adjusted for age, sex, comorbidities, and surgical location, odds of a revisit increased during the preguideline period (0.4% increase per month; 95% CI, 0.24%-0.54%; P < .001). This monthly increase did not continue after guidelines (P = .002).
Conclusions: While odds of a postoperative revisit did not decline after guideline publication, there was a significant difference in trend between the pre- and postguideline periods. Changes in the ambulatory surgery population also suggest at least partial adherence to guidelines.