To the Editor:
The United States Preventive Services Task Force (USPSTF) guidelines recommend yearly low-dose CT (LDCT) screening for high-risk smokers. These guidelines are based on the National Lung Screening Trial (NLST), which showed that yearly LDCT screening for high-risk smokers decreased lung cancer-related mortality.1 Despite strong supporting evidence, national data indicate low screening uptake, with < 6% of USPSTF criteria-eligible smokers being screened in 2015.2 However, it is unclear if screening uptake is increasing, and very little is known about predictors of screening uptake.
The current study estimated 2017 rates of LDCT screening among USPSTF criteria-eligible smokers in Florida, Nevada, and Georgia and investigated factors associated with utilization of screening.
Materials and Methods
The 2017 Behavioral Risk Factor Surveillance System (BRFSS) dataset was used to identify respondents across three states who met the USPSTF criteria for lung cancer screening. BRFSS comprises health survey data collected via random digit-dialed probability sampling across all 50 states. Self-reported LDCT imaging for lung cancer screening was the primary outcome. Weighted percentages of self-reported covariates were used to summarize the data, and logistic regression was corrected for sampling weight. All analyses were conducted by using SAS version 9.4 (SAS Institute, Inc.). This study was based on a de-identified, publicly available database and was exempt from institutional review board review.
Results
Of the estimated 866,305 smokers eligible for lung cancer screening according to USPSTF criteria, 141,161 (95% CI, 107,392-174,930) or 16.3% (95% CI, 12.7-19.9) received it. Among the screened and unscreened, the majority were white (85.2% and 83.2%, respectively) (Table 1). Black race and Hispanic ethnicity were not associated with different screening rates, nor was sex or income. Lack of insurance and annual income less than $15,000 were associated with lower screening rates; self-reported COPD diagnosis was associated with higher screening rates. The proportion of participants with income less than $15,000 and self-reported COPD were similar across the three states, whereas the proportion of uninsured participants was highest in Georgia (10.2%) and lowest in Nevada (2.3%) (Table 2).
Table 1.
Baseline Characteristics, Prevalence of LDCT Testing, and Adjusted ORs for Predictors of LDCT Screening in USPSTF Criteria-Eligible Respondents
| Characteristic | Received Screening (n = 141,161) | Did Not Receive Screening (n = 725,144) | OR (95% CI) |
|---|---|---|---|
| Age, y | 66.9 ± 0.6 | 65.8 ± 0.4 | 1.16 (0.69-1.96) |
| Race/ethnicity | |||
| White (16.6% screened) | 120,213 (85.2) | 603,389 (83.2) | Reference |
| Black (18.6% screened) | 11,793 (8.4) | 51,608 (7.1) | 1.95 (0.49-7.8) |
| Hispanic (9.5% screened) | 2,234 (1.6) | 21,224 (2.9) | 0.60 (0.16-2.26) |
| Other (8.8% screened) | 6,921 (4.9) | 48,923 (6.7) | 0.73 (0.31-1.75) |
| Male | 85,479 (60.6) | 430,733 (59.4) | 1.37 (0.80-2.35) |
| Married | 50,207 (35.6) | 338,121 (46.6) | 0.52 (0.31-0.88) |
| College (less than high school, high school, college) | 74,346 (52.7) | 355,195 (49.0) | 1.18 (0.70-1.99) |
| LGB | 9,956 (7.1) | 22,174 (3.1) | 2.36 (0.95-5.82) |
| Income, $ | |||
| < 15,000 | 9,318 (6.6) | 89,015 (12.3) | Reference |
| > 15,000 | 131,843 (93.4) | 636,129 (87.7) | 0.33 (0.16-0.68) |
| No insurance | 269 (0.2) | 59,737 (8.2) | 0.02 (0.01-0.10) |
| COPD | 90,613 (64.2) | 230,161 (31.7) | 4.61 (2.56-8.30) |
| Pack years | 58.0 ± 3.1 | 53.0 ± 1.5 | 1.00 (0.99-1.01) |
Data are presented as mean ± SD or No. (%). LDCT = low-dose CT; LGB = lesbian, gay, bisexual; USPSTF = United States Preventive Services Task Force.
Table 2.
State-Based Breakdown of Baseline Characteristics and Prevalence of LDCT Testing for Lung Cancer Among USPSTF-Criteria Screening-Eligible Smokers
| Characteristic | Florida (n = 597,830) | Nevada (n = 82,036) | Georgia (n = 186,438) |
|---|---|---|---|
| Age, y | 66.2 ± 0.4 | 65.7 ± 2.9 | 65.4 ± 0.5 |
| Race/ethnicity | |||
| White | 497,067 (83.1) | 69,444 (84.6) | 157,092 (84.3) |
| Black | 49,121 (8.2) | 2,366 (2.9) | 11,914 (6.4) |
| Hispanic | 20,158 (3.4) | 2,342 (2.9) | 959 (0.5) |
| Other | 31,484 (5.3) | 7,885 (9.6) | 16,474 (8.8) |
| Male | 346,056 (57.9) | 49,883 (60.8) | 120,273 (64.5) |
| Married | 259,084 (43.4) | 34,505 (42.1) | 94,739 (50.8) |
| College (less than high school, high school, college) | 303,310 (5.7) | 42,299 (51.6) | 83,932 (45.0) |
| LGB | 19,656 (3.3) | 1,537 (1.9) | 10,937 (5.9) |
| Income, $ | |||
| < 15,000 | 71,120 (11.9) | 6,795 (8.3) | 20,419 (11.0) |
| 15,000-50,000 | 306,751 (51.3) | 33,074 (40.3) | 74,972 (40.2) |
| > 50,000 | 219,959 (36.8) | 42,167 (51.4) | 91,047 (48.8) |
| No insurance | 39,023 (6.5) | 1,925 (2.3) | 19,059 (10.2) |
| COPD | 224,774 (37.6) | 31,140 (38.0) | 64,859 (34.8) |
| Pack years | 55.0 ± 1.8 | 52.0 ± 2.9 | 51.0 ± 2.0 |
| Received LDCT screening | 114,914 (19.2) | 5,646 (6.9) | 20,600 (11.0) |
Data are presented as mean ± SD or No. (%). See Table 1 legend for expansion of abbreviations.
Discussion
Compared with a previous study that reported low national rates of lung cancer screening from 2010 to 2015,2 we found that in the states of Florida, Nevada, and Georgia, the collective rate in 2017 was much higher at 16.3%. This increase in lung cancer screening rates may reflect recent campaigns to identify high-risk smokers for screening and raise physician and public awareness, as well as progressive uptake as observed following introduction of other cancer screening tests such as mammography.3 Negative predictors of screening include lack of insurance and low income, expected findings consistent with drivers of access to health care.4 Notably, the rates of uninsured participants were higher in states that did not expand Medicaid (Florida and Georgia), although on the state level this fact did not correlate with increased lung cancer screening.5 Self-reported COPD correlated with increased lung cancer screening, perhaps reflecting more aggressive screening in participants with this well-known risk factor for lung cancer. It may also reflect the inclusion of LDCT imaging as part of the initial COPD evaluation.
A limitation of this study is that only three of the 11 states that adopted the lung cancer screening module also adopted the respiratory module, limiting the generalizability of the findings among screening eligible individuals with COPD. Efforts should be made to include lung cancer screening questions into the core set of BRFSS questions. Data regarding demographic characteristics, smoking history, and lung cancer screening were all self-reported with limited validation, which may contribute to biased responses such as the underestimation of rates of the uninsured. Despite this, our study is the first to provide an updated prevalence of LDCT screening and identify a predictor of screening among USPSTF criteria-eligible populations.
Footnotes
FINANCIAL/NONFINANCIAL DISCLOSURES: None declared.
References
- 1.The National Lung Screening Trial Research Team Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011;365(5):395–409. doi: 10.1056/NEJMoa1102873. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Jemal A., Fedewa S.A. Lung cancer screening with low-dose computed tomography in the United States—2010 to 2015. JAMA Oncol. 2017;3(9):1278–1281. doi: 10.1001/jamaoncol.2016.6416. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Blackman D.K., Bennett E.M., Miller D.S. Trends in self-reported use of mammograms (1989-1997) and Papanicolaou tests (1991-1997)—Behavioral Risk Factor Surveillance System. MMWR CDC Surveill Summ. 1999;48(6):1–22. [PubMed] [Google Scholar]
- 4.Carter-Harris L., Slaven J.E., Monahan P.O., Shedd-Steele R., Hanna N., Rawl S.M. Understanding lung cancer screening behavior: racial, gender, and geographic differences among Indiana long-term smokers. Prev Med Rep. 2018;10:49–54. doi: 10.1016/j.pmedr.2018.01.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Collins S.R., Gunja M.Z., Doty M.M. Following the ACA repeal-and-replace effort, where does the U.S. stand on insurance coverage? Findings from the Commonwealth Fund Affordable Care Act Tracking Survey, March-June 2017. Issue Brief (Commonw Fund). 2017. 2017:1–21. [PubMed] [Google Scholar]