Observed-to-expected ratio for adherence to treatment guidelines as a quality of care indicator for laryngeal cancer.

Abstract

To examine associations between survival and adherence to National Comprehensive Cancer Network (NCCN) treatment guidelines using an observed-to-expected (O/E) ratio for greater adherence as a risk-adjusted hospital measure of quality care in elderly patients treated for larynx cancer. Retrospective analysis of Surveillance, Epidemiology, and End Results (SEER)-Medicare data. Patients diagnosed with larynx cancer from 2004 to 2007 were evaluated using multivariate regression and survival analysis. A fit logistic regression model was used to calculate an O/E ratio for guideline adherence for each hospital using quality indicators derived from NCCN guidelines for recommended treatment and stratified by hospital volume. Of 1,721 patients treated at 395 hospitals, 43.0% of patients received NCCN guideline-adherent care. Low-volume hospitals (N = 295) treating six or fewer cases treated 765 patients (44.5%), with a mean O/E of 0.96 ± 0.45. Hospitals treating more then six cases with an O/E <1 (N = 32) treated 284 patients (16.5%), with a mean O/E of 0.77 ± 0.18. Hospitals treating more than six cases with an O/E ≥1 (N = 68) treated 672 patients (39.1%), with a mean O/E of 1.17 ± 0.11. Treatment at hospitals with an O/E ≥1 was associated with improved survival (hazard ratio [HR] = 0.83 [95% confidence interval [CI]: 0.70 to 0.98]) and lower mean incremental treatment-related costs (-$3,009 [-$5,226 to -$791]) compared with hospitals with an O/E <1 (HR = 1.00 [0.80 to 1.24]) and the reference group of low-volume hospitals. A hospital-specific O/E for NCCN treatment guideline adherence, combined with a minimum case volume criterion, is associated with survival and treatment-related costs in elderly patients with larynx cancer, and may be a feasible measure of larynx cancer quality of care. NA Laryngoscope, 130:672-678, 2020.