Intensity of Diagnostic Chest CT Imaging and False Positive Lung Cancer Rates Before the Introduction of Lung Cancer Screenings in VA

Zeliadt SB, Backhus LM, Reinke LF, Hebert PL, Liu C, Hu E, Au DH. Intensity of Diagnostic Chest CT Imaging and False Positive Lung Cancer Rates Before the Introduction of Lung Cancer Screenings in VA. Poster session presented at: AcademyHealth Annual Research Meeting; 2014 Jun 10; San Diego, CA.

Related HSR&D Project(s)

• IIR 12-065 – A Novel Approach to Measuring Costs and Efficiency: Lung Nodules as a Case Study

Following the 20% reduction in lung cancer mortality observed in the National Lung Screening Trial (NLST), annual lung cancer screening with chest CT has been recommended for long term smokers ages 55-80 by the US Preventive Services Task Force and many (but not all) provider organizations. Chest CT imaging is highly sensitive at detecting suspicious findings/nodules but has low specificity with many findings being insignificant. Radiology quality and aggressiveness of care following suspicious findings from both screening CTs and chest CTs ordered for diagnostic purposes may potentially alter the balance between the benefits and harms of imaging due to complications associated with invasive diagnostic procedures such as transthoracic needle biopsy and lung resections. Variability among radiologists was extremely high in the context of the NLST; little information is available about variability and radiology quality in the community. A cohort of VA-users who received a chest CT in FY2011 were identified from electronic medical records. Subjects who had a prior diagnosis of cancer or for whom the CT was a follow-up from a prior imaging study were excluded. Patients from facilities who performed fewer than 100 chest CTs were excluded. The frequency of chest CTs coded as positive/suspicious was determined using ICD9 codes 793.1 (pulmonary coin lesion) and 786.6 (chest mass). Patients were followed for up to 24 months from the date of the chest CT to identify the frequency of true positive/false positive findings. To determine the frequency of background diagnostic chest CT imaging performed in the VA prior to the introduction of lung cancer screening and explore radiology quality. 159,453 (2.5%) patients between ages 30-79 from among 6.3 million VA users in FY2011 had at least one newly ordered diagnostic chest CT; 199,958 additional patients received a chest CT for follow-up of a prior CT or for staging/surveillance for cancer. Frequency of newly ordered chest CT imaging varied across the 126 facilities from 5 per 1000 patients to 49 per 1000 patients that could not be explained by age differences or smoking differences across facilities. 23,768 (15%) of patients with a newly ordered chest CT were considered 'positive' based on ICD9 codes. Of those with a positive finding, 6,362 were determined to be true positive for an overall positive predicted value (PPV) of 26.8%. 17,406 (10.9%) patients had false positive findings, which ranged from 4.2% to 22.3% across the 126 VA facilities. Positive predictive values (PPV) ranged from 12.9% to 55.3% across VA facilities. PPV rates were higher among current smokers (34.4%) compared to former smokers (22.2%) and never smokers (10.1%). Based on data available to date, we have not identified any facility-level predictors associated with false positive rates such as volume of chest CTs.