Lead/Presenter: Renda Wiener, COIN - Bedford/Boston
All Authors: Wiener RS (Center for Healthcare Organization & Implementation Research, VA Boston Healthcare System), Nunez ER (Center for Healthcare Organization & Implementation Research, VA Boston Healthcare System) Qian SX (Center for Healthcare Organization & Implementation Research, VA Bedford Healthcare System) Zhang S (Center for Healthcare Organization & Implementation Research, VA Bedford Healthcare System) Boudreau J (Center for Healthcare Organization & Implementation Research, VA Bedford Healthcare System) Glickman M (Center for Healthcare Organization & Implementation Research, VA Bedford Healthcare System) Miller DR (Center for Healthcare Organization & Implementation Research, VA Bedford Healthcare System) Slatore CG (VA National Center for Lung Cancer Screening; VA Portland Healthcare System) Caverly TJ (Center for Clinical Management Research, VA Ann Arbor Healthcare System)
Objectives:
Lung cancer is the leading cause of cancer death among Veterans. The National Lung Screening Trial demonstrated a 20% reduction in lung cancer mortality with three annual lung cancer screening (LCS) exams. This mortality reduction was dependent on annual screening: lung cancer detection rates were similar throughout each round. Yet adherence to LCS in real-world practice may be suboptimal. We sought to determine how often Veterans receive three annual rounds of LCS and identify patient and facility-level factors associated with adherence. We hypothesized that Veterans receiving care at sites with an LCS coordinator would be more likely to receive 3 rounds of annual screening.
Methods:
Using the VA Corporate Data Warehouse, we assembled a national retrospective cohort of Veterans who initiated LCS between 2015-18 in any VA facility, identified by presence of a CT chest with a report including a Lung-RADS (Lung CT Screening Reporting and Data System) code. We then assessed for the presence of a second and third annual LCS exam 10-15 months from the prior CT. Disqualifying events (those for which we would not expect a subsequent annual LCS exam) included high-risk LCS findings (i.e. Lung-RADS category 3 and 4) for which closer follow-up is recommended, lung cancer diagnosis, reaching age>80 years, or death before the next annual LCS exam was due. We calculated the proportion of Veterans who received a total of 3 annual LCS exams over a period of 3 years (allowing some room for delayed LCS). We then performed hierarchical mixed effects logistic regression analyses, with patients nested within facilities, and facilities as random effects.
Results:
Among 11,467 Veterans that initiated LCS, half (46%) received 3 annual LCS exams and a quarter (26%) had no LCS follow-up. Veterans that were racially Black (OR 0.82, 95% CI 0.68-0.98) or lived farther away from a VA facility (OR 0.86, 95% CI 0.81-0.92) were less likely to receive three annual LCS exams. Meanwhile, Veterans that had more frequent outpatient visits (OR 1.38, 95% CI 1.23-1.54) and those with initial LCS results categorized as Lung-RADS 2 vs. Lung-RADS 1 (i.e. likely benign nodule vs. no nodule detected) were more likely to have 3 annual LCS exams. At the facility-level, those with medium (OR 4.63, 95% CI 1.5-14.1) or high LCS volume (OR 6.5, 95% CI 2.4-17.6), or those facilities that used a LCS coordinator (OR 6.86, 95% CI 1.1-44.6) were more likely to receive 3 annual LCS. The facility at which a Veteran received their initial LCS was responsible for 34% of the variation in receiving 3 LCS.
Implications:
In a national cohort of Veterans, there was suboptimal longitudinal adherence to annual LCS. Further, we identified disparities among Veterans who face barriers to access: Black Veterans and those that live farther away from a VA facility.
Impacts:
To achieve the life-saving potential of LCS in VA, it is critical to optimize adherence. Facility-level structures that support LCS (e.g. coordinators) can improve adherence to LCS and may be an avenue to address disparities among underserved populations.