Appropriateness of imaging for lung cancer staging in a national cohort
Leah M Backhus1, Farhood Farjah2, Thomas K Varghese2
1Leah M. Backhus, Xiao-Hua Zhou, and Steven B. Zeliadt, Veterans Affairs Puget Sound Health Care System; and Leah M. Backhus, Farhood Farjah, Thomas K. Varghese, Aaron M. Cheng, Douglas E. Wood, Larry Kessler, and Steven B. Zeliadt, University of Washington, Seattle, WA. lbackhus@u.washington.edu.
Purpose:
Optimizing evidence-based care to improve quality is a critical priority in the United States. We sought to examine adherence to imaging guideline recommendations for staging in patients with locally advanced lung cancer in a national cohort.
Methods:
We identified 3,808 patients with stage IIB, IIIA, or IIIB lung cancer by using the national Department of Veterans Affairs (VA) Central Cancer Registry (2004-2007) and linked these patients to VA and Medicare databases to examine receipt of guideline-recommended imaging based on National Comprehensive Cancer Network and American College of Radiology Appropriateness Criteria. Our primary outcomes were receipt of guideline-recommended brain imaging and positron emission tomography (PET) imaging. We also examined rates of overuse defined as combined use of bone scintigraphy (BS) and PET, which current guidelines recommend against. All imaging was assessed during the period 180 days before and 180 days after diagnosis.
Results:
Nearly 75% of patients received recommended brain imaging, and 60% received recommended PET imaging. Overuse of BS and PET occurred in 25% of patients. More advanced clinical stage and later year of diagnosis were the only clinical or demographic factors associated with higher rates of guideline-recommended imaging after adjusting for covariates. We observed considerable regional variation in recommended PET imaging and overuse of combined BS and PET.
Conclusion:
Receipt of guideline-recommended imaging is not universal. PET appears to be underused overall, whereas BS demonstrates continued overuse. Wide regional variation suggests that these findings could be the result of local practice patterns, which may be amenable to provider education efforts such as Choosing Wisely.
