Abstract

To recognize fully the benefit of lung cancer screening (LCS), annual adherence must approach the high levels seen in the National Lung Screening Trial. Emerging data suggest that annual adherence is poor and that a centralized approach to screening improves adherence. Do differences in adherence exist between a centralized and decentralized approach to LCS within a hybrid program and what are predictors of adherence? A retrospective evaluation of a single-center hybrid LCS program was conducted to compare outcomes including patient eligibility and adherence between the centralized and decentralized approaches. Patient demographics and outcomes were compared between those screened with a centralized and decentralized approach and between adherent and nonadherent patients using two-sample t tests, χ 2 tests, or analyses of variance, as appropriate. Annual adherence analysis was conducted using data from patients who remained eligible for screening with a baseline Lung CT Screening Reporting and Data System (Lung-RADS) score of 1 or 2. Logistic regression was used to estimate the association between adherence and the primary exposure, adjusting for potential confounders. A cohort of 1,117 patients underwent baseline low-dose CT imaging. Two hundred eleven patients (19%) were ineligible by United States Preventative Services Task Force criteria and most (90%) were screened with the decentralized approach. After exclusions, 765 patients with Lung-RADS score of 1 or 2 remained eligible for annual screening. Overall adherence was 56%; however, adherence in the centralized program was 70%, compared with 41%with the decentralized approach (P< .001). Individuals screened in a decentralized approach were 73%less likely to be adherent (OR, 0.27; 95%CI, 0.19-0.37). A greater proportion of patients with three or more comorbidities were screened outside the centralized program. Those screened using a centralized approach were more likely to meet eligibility criteria for LCS and more likely to return for annual screening than those screened using a decentralized approach.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call