Alex A Balekian1, Nichole T Tanner2,3, Joshua M Fisher1, Gerard A Silvestri3, Michael K Gould4. 1. 1 Division of Pulmonary, Critical Care, and Sleep Medicine, Keck School of Medicine, University of Southern California, Los Angeles, California. 2. 2 Health Equity and Rural Outreach Innovation Center, Ralph H. Johnson Veterans Affairs Hospital, Charleston, South Carolina. 3. 3 Division of Pulmonary and Critical Care Medicine, Medical University of South Carolina, Charleston, South Carolina; and. 4. 4 Department of Research and Evaluation, Kaiser Permanente Southern California, Pasadena, California.
Abstract
RATIONALE: Lung cancer screening with low-dose computed tomography (LDCT) has been shown to decrease mortality in eligible high-risk patients. However, this mortality benefit comes with a high rate of false-positive findings, which require further evaluation. OBJECTIVES: To identify patient- and center-specific factors associated with having a pulmonary nodule on baseline LDCT, and to develop a prediction rule to help in shared decision making. METHODS: We identified individuals who underwent baseline LDCT screening as part of the National Lung Screening Trial. A positive screen was defined as a nodule 4 mm or greater in largest dimension. Using multiple logistic regression, we identified variables independently associated with having a positive screen. MEASUREMENTS AND MAIN RESULTS: Among the 26,004 patients with complete data who underwent baseline LDCT, 7,123 patients (27%) had a positive screen. In a multivariate analysis, older age (odds ratio [OR] = 1.03 per 1-year increase, 95% confidence interval [CI] = 1.03-1.04), female sex (OR = 1.08, 95% CI = 1.01-1.14), white race (OR = 1.39, 95% CI = 1.25-1.55), heavier smoking history (OR = 1.02 per 5 pack-years smoked over 30, 95% CI = 1.00-1.04), history of chronic obstructive pulmonary disease (OR = 1.08, 95% CI = 1.01-1.17), being married (OR = 1.08, 95% CI = 1.02-1.15), hard rock mining (OR = 1.40, 95% CI = 1.04-1.89), and farm work (OR = 1.13, 95% CI = 1.03-1.23) were independently associated with having a positive screen, whereas having a college degree (OR = 0.94, 95% CI = 0.86-1.00) and abstinence from smoking (OR = 0.98 per year, 95% CI = 0.98-0.99) were associated with not having a positive screen. Patients enrolled at a site in an area highly endemic for histoplasma were 30% more likely to have a positive baseline LDCT screen (OR = 1.30, 95% CI = 1.21-1.40). The area under the receiver operator characteristic curve for the full model was 0.57 (0.56-0.58); including enrollment center as a random effect increased the area under the receiver operator characteristic curve to 0.65. CONCLUSIONS: In the National Lung Screening Trial, both patient- and center-specific factors were associated with having a positive baseline screen. Although the model does not have sufficient accuracy to provide personalized risk estimates to guide shared decision making on an individual basis, it can nonetheless inform screening centers of the likelihood of further follow-up testing for their populations at large when allocating resources. Data collected from centers as broad-based screening is implemented can be used to improve model accuracy further.
RCT Entities:
RATIONALE: Lung cancer screening with low-dose computed tomography (LDCT) has been shown to decrease mortality in eligible high-risk patients. However, this mortality benefit comes with a high rate of false-positive findings, which require further evaluation. OBJECTIVES: To identify patient- and center-specific factors associated with having a pulmonary nodule on baseline LDCT, and to develop a prediction rule to help in shared decision making. METHODS: We identified individuals who underwent baseline LDCT screening as part of the National Lung Screening Trial. A positive screen was defined as a nodule 4 mm or greater in largest dimension. Using multiple logistic regression, we identified variables independently associated with having a positive screen. MEASUREMENTS AND MAIN RESULTS: Among the 26,004 patients with complete data who underwent baseline LDCT, 7,123 patients (27%) had a positive screen. In a multivariate analysis, older age (odds ratio [OR] = 1.03 per 1-year increase, 95% confidence interval [CI] = 1.03-1.04), female sex (OR = 1.08, 95% CI = 1.01-1.14), white race (OR = 1.39, 95% CI = 1.25-1.55), heavier smoking history (OR = 1.02 per 5 pack-years smoked over 30, 95% CI = 1.00-1.04), history of chronic obstructive pulmonary disease (OR = 1.08, 95% CI = 1.01-1.17), being married (OR = 1.08, 95% CI = 1.02-1.15), hard rock mining (OR = 1.40, 95% CI = 1.04-1.89), and farm work (OR = 1.13, 95% CI = 1.03-1.23) were independently associated with having a positive screen, whereas having a college degree (OR = 0.94, 95% CI = 0.86-1.00) and abstinence from smoking (OR = 0.98 per year, 95% CI = 0.98-0.99) were associated with not having a positive screen. Patients enrolled at a site in an area highly endemic for histoplasma were 30% more likely to have a positive baseline LDCT screen (OR = 1.30, 95% CI = 1.21-1.40). The area under the receiver operator characteristic curve for the full model was 0.57 (0.56-0.58); including enrollment center as a random effect increased the area under the receiver operator characteristic curve to 0.65. CONCLUSIONS: In the National Lung Screening Trial, both patient- and center-specific factors were associated with having a positive baseline screen. Although the model does not have sufficient accuracy to provide personalized risk estimates to guide shared decision making on an individual basis, it can nonetheless inform screening centers of the likelihood of further follow-up testing for their populations at large when allocating resources. Data collected from centers as broad-based screening is implemented can be used to improve model accuracy further.
Entities:
Keywords:
lung cancer screening; lung nodule; predictive model
Authors: Andrea N Burnett-Hartman; Nikki M Carroll; Stacey A Honda; Caroline Joyce; Nandita Mitra; Christine Neslund-Dudas; Oluwatosin Olaiya; Katharine A Rendle; Mitchell D Schnall; Anil Vachani; Debra P Ritzwoller Journal: Ann Am Thorac Soc Date: 2022-03
Authors: Peter J Mazzone; Gerard A Silvestri; Lesley H Souter; Tanner J Caverly; Jeffrey P Kanne; Hormuzd A Katki; Renda Soylemez Wiener; Frank C Detterbeck Journal: Chest Date: 2021-07-13 Impact factor: 9.410
Authors: Peter J Mazzone; Gerard A Silvestri; Lesley H Souter; Tanner J Caverly; Jeffrey P Kanne; Hormuzd A Katki; Renda Soylemez Wiener; Frank C Detterbeck Journal: Chest Date: 2021-07-13 Impact factor: 9.410