Computed Tomographic Screening for Lung Cancer Trends at Leading Academic Medical Centers From 2013 to 2015

In 2013, we¹ surveyed leading US academic medical centers (AMCs) and found variability in lung cancer screening (LCS) practices. Since then, favorable policy and payment decisions have been announced by the US Preventive Services Task Force (USPSTF)² and Centers for Medicare and Medicaid Services (CMS),³ and radiology-specific nodule guidelines have been established by the American College of Radiology (ACR).⁴ We resurveyed these same leading AMCs in 2014 and 2015 to reassess their practices and hypothesized that there would be greater conformity of practice patterns and increased patient volumes in response to these developments.

Surveys were emailed in March of 2013, 2014, and 2015 to thoracic radiology division chiefs at leading US AMCs, identified from the 2012-2013 US News & World Report overall ranking of best hospitals (n = 17), top 10 cancer centers, and top 10 pulmonology centers.⁵ From the 37 listings, 21 unique sites were identified (Table 1). Thirteen sites (67%) participated in prior multicenter LCS trials. Each survey (Table 2) inquired whether the site currently offered lung cancer screening. Additional questions related to screening practices, with selected questions repeated annually. Institutional review board approval and informed consent were waived by Beth Israel Deaconess Medical Center.

Of 21 sites, 19 (91%) responded in 2013, 20 (95%) in 2014, and 18 (86%) in 2015. The percentage of sites with a LCS program increased from 79% (15 of 19) in 2013 to 95% (19 of 20) in 2014 and 94% (17 of 18) in 2015.

Whereas 11 (73%) of 15 LCS sites used National Lung Screening Trial (NLST) entry criteria in 2013, only 6 of 17 sites (35%) used them in 2015, because several sites adopted CMS (n = 4) and USPSTF criteria (n = 4).

While an exclusive self-pay model was the norm in 2013, the percentage of sites using this model decreased to 47% in 2014 (9 of 19) and 6% in 2015 (1 of 17).

The most common response was 1 to 5 patients scanned per week each survey year, although the percentage of sites in this category decreased steadily from 87% (13 of 15) in 2013 to 74% (14 of 19) in 2014 and 53% (9 of 17) in 2015. Only 1 site reported scanning more than 20 patients per week in each survey year.

From 2013 to 2015, there was an evolution from using a variety of nodule management guidelines toward using ACR Lung Imaging Reporting and Data System (Lung-RADS) at most sites (13 of 17 [75%]) in 2015. Concurrently, there was greater uniformity among sites regarding the size threshold of a solid nodule for a positive screening result, with most (12 of 17 [70%]) using a criterion of 6 mm in 2015.

Our survey demonstrates several trends in LCS from 2013 to 2015, including broad adoption of ACR Lung-RADS, with associated greater conformity regarding threshold nodule size criteria for a positive screen. These findings suggest that radiology-specific guidelines have contributed to greater uniformity in LCS practices.

Despite favorable public policy decisions in support of LCS between 2013 and 2015, we observed only a modest increase in patient volume. We emphasize, however, that the timing of the survey occurred too early to determine the impact of CMS coverage on patient volumes.

By design, we targeted a focused population of leading AMCs to determine whether there is a consensus of “best practices.” We acknowledge that our results may not be representative of all AMCs or of other practice types offering screening in the community setting.

In summary, the development of radiology-specific guidelines has likely contributed to reduced variability in LCS practices at leading AMCs. We plan to continue our longitudinal survey of these sites to determine the impact of CMS coverage on patient screening volumes.

Corresponding Author: Phillip M. Boiselle, MD, Department of Radiology, Beth Israel Deaconess Medical Center, 330 Brookline Ave, TCC-4, Boston, MA 02215 (pboisell@bidmc.harvard.edu).

Published Online: February 11, 2016. doi:10.1001/jamaoncol.2015.6419.

Author Contributions: Dr Boiselle had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Boiselle, Ravenel.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Boiselle.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Boiselle.

Administrative, technical, or material support: Boiselle, Ravenel, White.

Conflict of Interest Disclosures: None reported.