Disparities in Lung Cancer Screening: A Review

Abstract

Lung cancer is the leading cause of cancer mortality in the United States. Certain groups are at increased risk of developing lung cancer and experience greater morbidity and mortality than the general population. Lung cancer screening provides an opportunity to detect lung cancer at an early stage when surgical intervention can be curative; however, current screening guidelines may overlook vulnerable populations with disproportionate lung cancer burden. This review aims to characterize disparities in lung cancer screening eligibility, as well as access to lung cancer screening, focusing on underrepresented racial/ethnic minorities and high-risk populations, such as individuals with human immunodeficiency virus. We also explore potential system- and patient-level barriers that may influence smoking patterns and healthcare access. Improving access to high-quality health care with a focus on smoking cessation is essential to reduce the burden of lung cancer experienced by vulnerable populations.

Lung cancer is the most common cause of cancer mortality in the United States (1). Overall survival for patients with lung cancer is poor, with a 5-year survival rate of 19% (1). Stage is one of the most important predictors of survival. For patients diagnosed with non–small-cell lung cancer, localized disease (stage I) has the best 5-year survival (57.4%), whereas in patients diagnosed with distant stage or metastatic disease (stage IV), 5-year survival is extremely poor (5.2%) (2). Despite the improved survival for early-stage non–small-cell lung cancer, only 16% of patients are diagnosed at localized stage (2). Lung cancer disproportionately affects certain high-risk populations, including black individuals, individuals with human immunodeficiency virus (HIV), and socioeconomically disadvantaged groups (1, 3–5)

Lung cancer screening offers an opportunity for the early detection of lung cancer when surgical options are available to improve outcomes (6). With lung cancer screening, survival may be improved for even the most vulnerable, high-risk populations (7, 8). This review characterizes disparities in lung cancer screening in U.S. populations and describes system- and patient-level barriers that may influence lung cancer screening access.

Disparities in Lung Cancer Incidence and Mortality

Lung cancer is the leading cause of cancer death for both men and women in the United States, but certain high-risk populations experience greater morbidity and mortality. Racial disparities are predominant, as black males have the highest rates of age-adjusted lung cancer incidence among all U.S. racial/ethnic groups, specifically 73.5 per 100,000, versus 63.5 per 100,000 for white males (Table 1) (2). This racial disparity in incidence persists in both smokers and never-smokers (9). Black males also have the highest lung cancer mortality compared with other racial/ethnic groups (62.1 vs. 51.7 age-adjusted overall mortality) (2). Black individuals develop lung cancer at an earlier age than white individuals (median age, 67 vs. 70 yr) (10) and are more likely to present with advanced-stage disease (53% among black individuals vs. 49% among white individuals) (11). Mortality racial differences narrow when adjusting for stage at diagnosis and equal access to care, but little progress has been made in diagnosing lung cancer at an earlier stage over past decades (12).

Table 1.

Smoking prevalence, lung cancer incidence, and mortality by race

Race/Ethnicity	Smoking Prevalence*			Lung Cancer Incidence†			Lung Cancer Mortality†
	Overall	Male	Female	Overall	Male	Female	Overall	Male	Female
White	16.6	17.2	16.0	56.7	63.5	51.8	42.7	51.7	35.6
Black	16.7	20.9	13.3	56.2	73.5	44.6	44.3	62.1	32.4
Hispanic	10.1	13.1	7.1	29.8	36.1	25.4	19.2	26.6	13.8
AI/AN	21.9	19.0	24.0	37.8	43.3	33.9	27.7	33.6	23.1
API	7.0	12.0	2.6	36.0	46.3	28.2	22.7	30.2	17.3

Definition of abbreviations: AI/AN = American Indian/Alaskan Native; API = Asian/Pacific Islander; SEER = surveillance, epidemiology, and end results.

^*Weighted percent, population adjusted; from 2015 National Health Interview Survey, United States (27).

^†Age-adjusted per 100,000; SEER 21 Area Registry, 2012–2016 (2).

Other underrepresented racial/ethnic populations have received much less attention in the published literature. Hispanics have a lower smoking prevalence and lung cancer mortality than black individuals and white individuals (Table 1) (2), whereas American Indian/Alaskan Natives have the highest overall prevalence of cigarette smoking yet have lower incidence and mortality rates than white individuals and black individuals. Research should be emphasized in other racial/ethnic groups, with the goal of reducing the burden of lung cancer; however, for the remainder of this review, we focus primarily on white individuals and black individuals, as these populations have among the highest incidence and mortality rates. Other populations, such as HIV-positive patients, are also at high risk and experience lung cancer incidence three times higher than the general population (5). Sex disparities also exist in lung cancer risk, screening, and outcomes but are outside the scope of this review.

Socioeconomic inequities contribute to cancer incidence and mortality, and the largest socioeconomic disparity has been reported in lung cancer (13). Low socioeconomic populations experience greater environmental exposures and subsequent increased lung cancer incidence and mortality (14, 15). Similarly, rurality leads to further disparities, as black individuals in rural areas have greater lung cancer incidence than those in urban areas (16).

Current State of Lung Cancer Screening in the United States

The U.S. Preventive Services Task Force (USPSTF) recommends lung cancer screening with low-dose computed tomography (LDCT) annually for individuals aged 55 to 80 years with reasonable life expectancy and at least a 30-pack-year smoking history who either currently smoke or quit within the prior 15 years (17). This USPSTF grade B recommendation is based on the 2011 randomized NLST (National Lung Screening Trial), which demonstrated a 20% relative decrease in mortality for individuals screened annually with LDCT compared with screening with chest X-ray (6). More than twice the number of lung cancers diagnosed in the LDCT arm were early-stage cancers, compared with the chest X-ray screening arm (6, 18). These NLST results were confirmed in the recently completed NELSON (Dutch-Belgian Lung Cancer Screening) trial. The NELSON trial was a large, prospective cohort of patients (N = 15,822) screened for lung cancer. Preliminary reports from the NELSON trial suggest a 26% relative reduction in lung cancer mortality with LDCT screening at 10 years in men and a greater mortality benefit (39%) among women, despite low enrollment (15%) of female patients (19). Of the screening-detected cancers, 69% were early stage, and surgical resection was three times more prevalent in the screening arm than in the nonscreening arm (19, 20). Secondary analysis of the NLST demonstrated a greater reduction in mortality in black versus white individuals (hazard ratio [HR], 0.61; 95% confidence interval [CI], 0.37–1.01 vs. HR, 0.86; 95% CI, 0.75–0.98), although not statistically significant (8). Thus, lung cancer screening has the potential to provide greater benefit to underrepresented populations.

Current uptake of lung cancer screening is universally low across the United States. Jemal and colleagues compared National Health Interview Survey data from 2010 and 2015 and found no change nationally in the prevalence of lung cancer screening (3.3% vs. 3.9%) before or after implementation of the 2013 USPSTF screening recommendation (21). Reasons for this lack of screening uptake include limited access to screening, limited patient acceptance, and lack of physician knowledge about screening guidelines (22, 23). Current lung cancer screening guidelines miss a large percentage of the high-risk population. Secondary analyses of the Framingham Heart Study found <60% of lung cancer cases met eligibility criteria for USPSTF lung cancer screening guidelines (24). In the Framingham study population, the individuals ineligible for lung cancer screening but who developed lung cancer included overlapping groups of former smokers who quit smoking >15 years ago (40.8% of former smokers), smokers who developed lung cancer outside of the eligible age range (29.7%), and smokers with <30 pack-years (15.4%) (24). The USPSTF is currently reviewing their lung cancer screening guidelines to assess appropriateness in high-risk populations that do not meet current guidelines (25).

Smoking Prevalence and Access to Cessation Interventions by Race/Ethnicity

Smoking is the primary risk factor for lung carcinogenesis (26). Black males and American Indian/Alaskan Natives have the highest prevalence of smoking (Table 1) (27). Despite overall higher cigarette smoking prevalence, black individuals diagnosed with lung cancer are more likely to be intermittent or light smokers than their white counterparts with fewer accumulated pack-years (28). Black individuals are more likely to start smoking later in life and smoke fewer cigarettes per day than white smokers (26, 29). Black smokers have greater exposure to tobacco carcinogens at lower amounts of cigarette consumption than white and Hispanic smokers, as a result of increased intensity of smoking and slower nicotine metabolism and clearance (30).

Lung cancer screening has been shown to have maximum benefit when combined with smoking cessation. Screening with LDCT in the NLST together with smoking cessation at 15 years demonstrated a 38% decrease in lung cancer–specific mortality (31). A higher lung cancer mortality (HR, 2.14; 95% CI, 1.80–2.54 and HR, 2.29; 95% CI, 1.97–2.68) and all-cause mortality (HR, 1.85; 95% CI, 1.68–2.04 and HR, 1.79; 95% CI, 1.63–1.95) was found in current smokers compared with former smokers in both the LDCT and chest radiograph control arms, respectively (31). In addition, after 7 years of smoking cessation, former smokers achieved comparable lung cancer mortality benefit as with LDCT screening in the NLST (31). Notably, black current smokers were four times more likely to die from lung cancer than black former smokers in the NLST (8). In addition to a high smoking prevalence, black individuals have decreased smoking cessation rates compared with white individuals (28, 29). One reason black individuals remain smokers may be related to the high prevalence of menthol cigarette use among black compared with white smokers. Current literature is mixed regarding the impact of menthol cigarette use on smoking cessation. Menthol use may lead to higher levels of dependence and addiction, making quitting more difficult (32). However, we and others have demonstrated no difference in smoking quit rates or lung cancer incidence and mortality between smokers of menthol and nonmenthol cigarettes (33).

The universal health benefits of smoking cessation have been repeatedly emphasized (34); however, underrepresented racial/ethnic groups are less likely to be offered smoking cessation by a healthcare provider than white smokers (34, 35). Asians, American Indians/Alaska Natives, and Hispanics receive less frequent tobacco counseling from providers (34). These lower rates of tobacco cessation counseling by a provider as well as decreased use of pharmaceutical aids among Hispanic and black smokers persist even after adjusting for socioeconomic status (35). The uninsured are also significantly less likely to be offered smoking cessation by a provider than those with insurance (34). Low health literacy is associated with decreased use of preventive services such as smoking cessation programs (36), and currently available educational materials on lung cancer screening are written at levels not accessible to those with average health literacy levels (37). Having a regular source of health care is associated with smoking cessation advice as well as use of smoking cessation aids (35). Prioritizing smoking cessation with lung cancer screening will likely improve outcomes for current smokers (38). Extensive research is underway to investigate how best to incorporate smoking cessation into a screening visit.

Factors Contributing to Disparities in Lung Cancer Screening

Race and Ethnicity

Current screening guidelines do not align similarly for black and white individuals. Racial differences in both smoking patterns and age at diagnosis contribute to racial disparities in screening eligibility. Screening provides an opportunity for earlier stage at diagnosis for all racial/ethnic groups (3, 12); however, black smokers have lower rates of lung cancer screening than white smokers (39). Black participants in the LDCT arm of the NSLT study had greater reduction in both lung cancer and all-cause mortality than white participants, despite low participation (4.4% black vs. 90.9% white) (8). In a more diverse screening population with a higher percentage of black individuals (69.6%), lung cancer screening detected a larger percentage of cancers than the NSLT study (2.6% vs. 1.1%), with the majority being early-stage lung cancer (7). Despite having greater lung cancer incidence, black smokers are less likely to be eligible for screening (40, 41), as the current lung cancer screening guidelines with the 30 pack-year inclusion criteria exclude a higher proportion of high-risk black smokers because of their lower average cigarette per day consumption compared with white smokers. This reduced cigarette smoking behavior has led researchers to suggest that expansion of lung cancer screening eligibility to individuals with any smoking history (42) or 20 to 29 pack-year smoking history (41, 43) would increase the proportion of screening-eligible black patients. Black smokers are also at greater risk of developing lung cancer at an earlier age, with the highest difference in age-adjusted incidence between black and white smokers noted in the 50- to 54-year-old age range, earlier than the currently recommended minimum screening age of 55 years (10). Reducing the minimum age criteria to 50 years, as done in the NELSON trial, would further increase the number of eligible black smokers (41). Current screening guidelines are projected to capture a higher proportion of eligible white smokers with lung cancer, potentially further exacerbating the disparity gap in lung cancer survival (40). Risk models that incorporate race/ethnicity, such as the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial model (PLCO_M2012) and the Lung Cancer Death Risk Assessment Tool (LCDRAT) demonstrate improved accuracy in predicting lung cancer risk compared with the NLST criteria (42, 44, 45). Future lung cancer screening guidelines should consider expanded eligibility criteria or risk-based approaches to address equity in screening eligibility (45).

Rurality

Disparities in lung cancer incidence between white and black men and women worsen in rural geographic regions versus urban areas (16). The prevalence of cigarette smoking is higher in rural counties (46), with more smokers consuming greater than 15 cigarettes a day than in metropolitan areas (26) and adolescents in rural communities initiating smoking at an earlier age than their urban counterparts (47). Lung cancer incidence for individuals living in rural areas is estimated to be 20% higher than those living in urban areas (48). Rural areas with poverty and socioeconomic deprivation are associated with higher lung cancer incidence and mortality rates (14, 16, 27, 48). Rural residents, including those in regions with high lung cancer burden, are less likely to have a comprehensive accredited screening facility within 30 miles of their residence (49–52). Strategies to increase access to screening should focus on these areas where geographic access to both lung cancer screening centers and high-quality treatment is limited.

Environmental and Occupational Exposures

Environmental factors other than tobacco also confer an increased risk for development of lung cancer and are considered by the National Comprehensive Cancer Network in their consensus-based guideline category 2 recommendations. These category 2 recommendations require a minimum 20 smoking pack-years and age 50 years, along with additional risk factors and PLCO_M2012 risk calculator assessment (53) for lung cancer screening eligibility. Radon is present in soil, concentrated in enclosed spaces such as mines and homes, and has been identified by the U.S. Environmental Protection Agency as the second leading cause of lung cancer after cigarette smoking (54). Secondhand smoke and other environmental hazards such as asbestos, chromium, arsenic, and air pollution also play a role in lung cancer risk, exacerbated by concurrent smoking (55). Environmental and occupational exposures are often more common in underrepresented minority populations and those with lower socioeconomic status (56), potentially contributing to existing disparities in lung cancer incidence. Screening with LDCT for asbestos-exposed workers yields detection of lung cancer at localized disease rates similar to those in the NLST (57), and lung cancer risk prediction tools, such as the Bach model, that incorporate asbestos exposure demonstrate improved performance over USPSTF eligibility criteria in national datasets (42, 58). In considering these environmental factors, challenges arise because of lack of patient awareness of exposure and low exposure in the general population; however, targeted occupational screening questions could be considered for high-risk individuals with subsequent referral for screening.

HIV Infection

Lung cancer is the leading cause of non–AIDS-defining cancer deaths, and lung cancer incidence in HIV-positive patients is significantly higher than the general population (5). In addition, age of lung cancer onset in HIV-positive patients is 25 to 30 years earlier than the general population, with average age of diagnosis between 38 and 57 years, compared with 70 years in the general population (5). Most of lung cancer cases occurring in HIV-positive patients present at late stage, with only 15% presenting at local, resectable stage, and, as a result, median survival is between 3.5 and 6.3 months (5, 59). HIV-infected patients have an estimated 52% excess lung cancer risk when compared with noninfected individuals (60). These high lung cancer rates have been attributed to high smoking prevalence among HIV-infected individuals, with smoking prevalence ranging from 25% to 80%, two to three times higher than the general population (5, 61). Prior studies also implicate the chronic inflammatory state and immunosuppressive treatment regimens in this population as well as a potential oncogenic role of the HIV virus (61). Current lung cancer screening guidelines perform poorly in individuals living with HIV (62). A modified Lung Cancer Policy Model that mirrors the distinctive aspects of lung cancer screening in HIV-infected individuals appears to provide similar mortality reduction in HIV-infected persons with a CD4⁺ cell count of ≥500 cells/μl, as in the general population (63). Because of the high incidence and mortality associated with lung cancer in HIV-infected patients, lung cancer screening should be considered in this high-risk group and must be accompanied by smoking cessation programs, although further work should be done to determine appropriate screening eligibility criteria (59).

Access to Care

Unequal access to high-quality health care may also contribute to racial disparities in lung cancer outcomes, affected by health policies that limit access to lung cancer screening. The USPSTF released lung cancer screening recommendations in December 2013 (17). In February 2015, the Center for Medicare & Medicaid Services issued a statement approving coverage for annual lung cancer screening according to USPSTF eligibility criteria for adults aged 55 to 77 years, mandating an accompanied shared decision-making visit that incorporates tobacco cessation counseling and submission of data to an approved registry (64). This expanded coverage of lung cancer screening for Medicare beneficiaries but not for eligible Medicaid beneficiaries, as Medicaid eligibility is determined at the state level. Employer-based and private health insurance plans, as well as states that have adopted the Medicaid expansion package, are required to cover this Grade B USPSTF recommendation (64). However, states that have not adopted Medicaid expansion leave high-risk patients without insurance coverage for lung cancer screening. As low-income individuals who often depend on Medicaid coverage have the highest rates of tobacco use (27), lack of coverage in states that did not adopt Medicaid expansion will likely lead to socioeconomic disparities in access to lung cancer screening. Furthermore, requirements for shared decision making and accompanied documentation create additional burden for providers that may limit likelihood of referral (65).

Underrepresented racial/ethnic populations are more likely to be uninsured or ineligible because of foreign-born nativity or citizenship status. Access to health care improves racial disparities in lung cancer survival (66–68), whereas individuals with Medicaid or the uninsured have poorer lung cancer survival than those with private health insurance (69). More than half of those who qualify for lung cancer screening on the basis of USPSTF recommendations are estimated to have Medicaid or be uninsured (70). High-risk rural populations experience additional barriers in accessing appropriate, high-quality health care, potentially contributing to disparities in access to lung cancer screening and appropriate follow-up care. Improving access to healthcare systems, both financial with insurance coverage and geographic, is important in addressing the burden of lung cancer in vulnerable populations.

Patient-Level Barriers

A generational history of discrimination and mistrust contributes to perceptions of stigma and challenges with patient–provider communication. Black and Hispanic populations report higher levels of physician mistrust than white populations, with large variability among individuals with disparate socioeconomic status, geographic location, and insurance (71). Underrepresented racial/ethnic populations and individuals with low socioeconomic status demonstrate greater beliefs of fatalism, nihilism, and the futility of medical intervention, which frustrates attempts at cancer prevention, such as smoking cessation and lung cancer screening (72). Stigma regarding smoking perpetuates fatalism and feelings of hesitancy in seeking medical care (73). Implementation of lung cancer screening and smoking cessation programs requires addressing community beliefs regarding the importance of smoking cessation and risk of lung cancer, risk that is not well understood among individuals with low socioeconomic status (18).

Conclusions

Racial differences in smoking behaviors, lung cancer incidence, age at presentation, and mortality should be considered in lung cancer screening guidelines. Black individuals have higher incidence, earlier presentation, and increased mortality of lung cancer than white individuals, despite lower overall cigarette consumption. Lung cancer screening has maximum benefit when combined with smoking cessation and should be emphasized in all populations. Current lung cancer screening eligibility guidelines using the 30-pack-year criterion exclude a large percentage of high-risk, light-smoker black individuals. Revisions to screening guidelines should consider racial/ethnic variation in cigarette smoking, additional risk factors, and overall level of risk. Individuals with HIV have a disproportionately high burden of lung cancer morbidity and mortality, and thus HIV status should be considered in screening eligibility guidelines. Consideration of screening for those with known occupational exposures also deserves further discussion. Improving access to high-quality healthcare systems, both financially with insurance coverage and geographically with access to high-volume screening facilities, is essential to address the high incidence and mortality of lung cancer. Implementation of lung cancer screening and smoking cessation programs requires addressing community beliefs regarding the importance of smoking cessation and risks of lung cancer. Outcomes for high-risk individuals and patients with lung cancer will equalize once disparities in lung cancer screening eligibility and access to care are considered. We call on professional organizations caring for patients with lung cancer to address the disparities discussed in this review and translate them into actionable policy recommendations.