Objectives: To compare rates of typmanostomy tube insertions for otitis media with effusion with estimates of need in 2 countries.
Study design: This cross-sectional analysis used all-payer claims to calculate rates of tympanostomy tube insertions for insured children ages 2-8 years (2007-2010) across pediatric surgical areas (PSA) for Northern New England (NNE; Maine, Vermont, and New Hampshire) and the English National Health Service Primary Care Trusts (PCT). Rates were compared with expected rates estimated using a Monte Carlo simulation model that integrates clinical guidelines and published probabilities of the incidence and course of otitis media with effusion.
Results: Observed rates of tympanostomy tube placement varied >30-fold across English PCT (N = 150) and >3-fold across NNE PSA (N = 30). At a 25 dB hearing threshold, the overall difference in observed to expected tympanostomy tubes provided was -3.41 per 1000 child-years in England and -0.01 per 1000 child-years in NNE. Observed incidence of insertion was less than expected in 143 of 151 PCT, and was higher than expected in one-half of the PSA. Using a 20 dB hearing threshold, there were fewer tube insertions than expected in all but 2 England and 7 NNE areas. There was an inverse relationship between estimated need and observed tube insertion rates.
Conclusions: Regional variations in observed tympanostomy tube insertion rates are unlikely to be due to differences in need and suggest overall underuse in England and both overuse and underuse in NNE.
Keywords: Child; appropriateness; clinical guidelines; geography; otitis media with effusion; overuse and underuse; tympanostomy tube; unwarranted variation.
Copyright © 2016 Elsevier Inc. All rights reserved.