Reducing Disparities in Lung Cancer Screening: It’s Not So Black and White

Lung cancer is the leading cause of cancer death in the United States (1). This is true for Black, Asian, and White men and women as well as Hispanic men (2‐4). To reduce lung cancer death, lung cancer screening with low-dose computed tomography was recommended by the US Preventative Task Force (USPSTF) in 2013 for high-risk individuals (5). These recommendations were based on the National Lung Cancer Screening Trial that showed a 20% relative reduction in lung cancer death among participants (6). This marked the first time that cancer screening was recommended for only a high-risk group. Researchers then began to question if these simple criteria adequality selected patients, especially given that Black populations smoke fewer cigarettes, are diagnosed at a younger age, and have a higher risk of developing and dying from lung cancer (7‐9). In July 2020, the USPSTF issued draft recommendations expanding the pool of eligible candidates by lowering the age at which to begin screening and pack-year thresholds (10). One of USPSTF’s goals was to reduce disparities and to include more racial and ethnic minorities, specifically Black individuals.

In this issue of the Journal, Landy and colleagues (11) tackled the important question “Does the 2020 USPSTF draft recommendations truly reduce racial/ethnic disparities?” They did so by using nationally representative data and flexible predictive models that compute the risk of lung cancer death and life-years gained (LYG) under 3 scenarios: 1) 2013 USPSTF recommendations, 2) 2020 USPSTF draft recommendations, and 3) 2020 USPSTF draft recommendations plus risk-based modeling. These models incorporate empirical data on the chances of dying from lung cancer and number of life-years lost using inputs including race and ethnicity, age, sex, pack-years, life expectancies, underlying comorbidities, family history, and screening efficacy, along with other assumptions.

Under the 2020 USPSTF recommendations, an additional 6.5 million people will be eligible for lung cancer screening. The authors report that 18% of people who ever smoked were eligible according to 2013 recommendations and 33% were according to 2020 recommendations, a 15% increase. This increase was enjoyed by all racial and ethnic groups but was greater in magnitude among White (16%) persons vs Black (14%), Asian (8%), and Hispanic (10%) persons. The proportion of lung cancer deaths averted, and LYG also increased across all race and ethnicities when 2020 recommendations were applied, but again, the gains did not eliminate disparities. For example, more than half (55%) of lung cancer deaths among White persons could have been prevented with 2013 USPSTF recommendations; this increased to 67% with the 2020 recommendations, a difference of 12%. About 30%-40% of Black, Hispanic, and Asian lung cancer deaths would have been prevented under the 2013 recommendations and, because of parallel or even smaller gains (9%-14%), in the proportion lung cancer deaths averted, disparities remained. For example, 41%-54% Black, Hispanic, and Asian lung cancer deaths would be prevented under the new 2020 recommendations, which is the same or a lower proportion of deaths averted among White persons (55%) in the 2013 recommendation.

These results are in line with previous studies that looked at the proportion of lung cancers that could have been captured with the 2013 recommendations (7,12). The power of the current study is the use of nimble risk-based models to examine simultaneous scenarios and outcomes (deaths, LYG). Landy and colleagues (11) found that 2020 recommendations plus risk-based screening could close potential disparities in eligibility and outcomes. Under this scenario, more than 80% of White and Black persons’ lung cancer deaths would be averted, eliminating the gap. However, only 59% of Hispanic and 63% of Asian persons’ lung cancer deaths would be averted.

The current study answers some questions but raises others about the promise of lung cancer screening to reduce disparities. The authors make a cogent argument that expanding the eligibility might increase disparities in absolute terms. Although we conjecture, if a rising tide lifts all boats but not to the same level, it’s certainly not achieving equity but is an incremental step toward it. More White, Black, Hispanic, and Asian persons will be included under the new recommendations, and 10 368 more lung cancer deaths in these racial and ethnic groups can be prevented with the expanded recommendations. Adding the risk-based model could potentially prevent an additional 2657 lung cancer deaths among Black persons.

The question we ask is: Even if recommendations are honed to perfectly select eligible candidates for screening, are they enough to reduce disparities in lung cancer deaths? Even with simple pack-year criteria, identification of eligible adults is just one of the many barriers to lung cancer screening, and currently only about 5%-6% of eligible US adults received low-dose computed tomography for lung cancer screening in 2018 (13,14). Implementing risk-based screening may prove even more difficult because providers may not have the time or the systems in place to compute a patients’ individualized risk and life expectancy. This may be especially challenging in the underresourced healthcare settings in which Black and Hispanic patients more frequently receive care (15). Further research is needed on how best to incorporate risk-based screening into practice. Another challenge in addressing lung cancer disparities is the pervasiveness of them from prevention to screening to treatment to survivorship. It appears that Black persons who received lung cancer screening may have better outcomes than White persons, so the potential is there (16). However, this observation was within a controlled setting, with rigorous follow-up. Although not demonstrated yet for lung cancer screening in the community setting, Black persons experience disparities during they screening process. For example, Black persons receive lower quality of breast and colorectal cancer screening and they are also less likely to receive follow up after abnormal tests (17‐19). When it comes to treatment, Black lung cancer patients are less likely to receive guideline-concordant care than White patients, so even if all groups are diagnosed at an earlier lung cancer stage, the same benefits may not be reaped (20,21). Last, lowering the screening age will also increase the number of eligible adults who are uninsured, have Medicaid insurance, or experience disruptions in insurance coverage, all of which are related to lower receipt of cancer screening and are more common among racial and ethnic minorities (22,23). This could widen disparities even in the face of recommendations aimed at reducing them. Unfortunately, equitable screening recommendations are necessary but not sufficient steps in reducing the entrenched lung cancer disparities. You see, it’s not so Black and White.

Funding

None.

Notes

Role of the funder: Not applicable.

Disclosures: Dr Fedewa is employed by the American Cancer Society, which receives grants from private and corporate foundations, including foundations associated with companies in the health sector for research outside of the submitted work. Dr Fedewa is not funded by any of these grants, and her salary is solely funded through American Cancer Society funds. Dr Silvestri has nothing to disclose regarding this manuscript.

Author contributions: Both authors contributed equally to writing and editing this manuscript and approved the final product.

Data Availability

Not applicable.