ASPO Abstracts
Rate and Predictors of Lung Cancer Screening Utilization in a Large Health System in South Carolina
Category: Early Detection & Risk Prediction
Conference Year: 2022
Abstract Body:
Purpose: To examine the rate and predictors of lung cancer screening (LCS) utilization in a large health system in South Carolina.Methods: Data on patients meeting LCS eligibility criteria (aged 55-77 years, current smokers or former smokers who quit smoking within the past 15 years, smoking ‚â•30 pack years) in 2019 were extracted from the electronic medical record system. The outcome was LCS utilization. Potential predictors included: age, sex, race, marital status, insurance, zip code level median income, zip code level urbanicity, travel time from the centroid of zip code area to the nearest screening site (<15, ‚â•15 minutes), smoking status, body mass index, chronic pulmonary disease (COPD), and Charlson Comorbidity Index (CCI) (0, 1, 2, ‚â•3). Chi-square test, univariate and multivariable logistic regression were employed in the analysis.Results: A total of 6,930 patients were included. About half of patients were age 65-77, male, married or living with a partner, current smokers, had COPD, or CCI ‚â•3. The majority of patients were white, insured by Medicare or Medicaid, living in a metropolitan area, or had <15 minutes of driving time to the nearest LCS site. Among all LCS-eligible patients, 1,432 (20.66%) received LCS. Compared to patients who did not receive LCS, those who received screening were more likely to be older, female, married or living with partner, covered with Medicare or Medicaid, former smokers, have COPD, or live in a metropolitan area or an area with income >$60,000 or <15 minutes of driving time to the nearest LCS site. After adjusting for covariates, significant predictors of receiving LCS included: aged 65-77 (adjusted odds ratio [aOR]: 1.27 [1.10-1.46]), female (1.18 [1.04-1.34]), African American race (1.45 [1.20-1.76]), single or other marital status (0.83 [0.73-0.94]), living in a non-metropolitan area (0.32 [0.26-0.40]), with ‚â•15 minutes of travel time to a LCS site (0.70 [0.58-0.84]), with COPD (2.35 [2.03-2.71]), and with CCI‚â•3 (0.74 [0.59-0.93]).Conclusions: Disparities exist in LCS utilization, especially among patients living in non-metropolitan areas or areas with less accessibility to a screening site. Further examination of social determinants and structural factors related to LCS is needed.
Keywords: lung cancer screening, predictors, large health system