Abstract
Background
Criteria for low-dose CT scan lung cancer screening vary across guidelines. Knowledge of the eligible pool across demographic groups can enable policy and programmatic decision-making, particularly for disproportionately affected populations.
Research Question
What are the eligibility rates for low-dose CT scan screening according to sex and race or ethnicity and how do these rates relate to corresponding lung cancer incidence rates?
Study Design and Methods
This was a cross-sectional study using data from the 2015 National Health Interview Survey adult and cancer control supplement files. In addition to eligibility rates, the ratio of the eligibility rate to the lung cancer incidence rate in a given population group (eligibility to incidence [E-I] ratio) also was determined. Guidelines assessed were: Centers for Medicare and Medicaid Services, National Comprehensive Cancer Network, and US Preventive Services Task Force current or with expansion of age and smoking or quit thresholds. We also assessed a risk model (PLCOM2012 risk model).
Results
Total numbers eligible based on current guidelines ranged from 8.3 to 13.3 million, representing 8.3% to 13.4% of the US population 50 to 80 years of age, and up to 17.5 million with expanded criteria. Overall eligibility rates on average were about 10 percentage points higher for men than women. For both men and women, and both overall and among ever smokers, non-Hispanic Whites had the highest eligibility rates across all guidelines, followed generally by non-Hispanic Blacks, and then Asians and Hispanics. Among both men and women, non-Hispanic Whites had the highest E-I ratios across all guidelines; non-Hispanic Black men had higher lung cancer incidence, but 30% to 50% lower E-I ratios, than non-Hispanic White men.
Interpretation
Screening eligibility rates vary widely across guidelines, with disparities evident in E-I ratios, including among non-Hispanic Black men, despite higher lung cancer burden. Consideration of smoking duration in risk assessment criteria may address current disparities.
Key Words: disparities, eligibles, guidelines, lung cancer, screenings
Abbreviations: CMS, Center for Medicare and Medicaid Services; E-I, eligibility to incidence; NCCN, National Comprehensive Cancer Network; Expanded50|20|25, expansion of 2020 USPSTF proposed recommendation: lower age limit of 50, 20+ pack-y; NHIS, National Health Interview Survey; USPSTF, US Preventive Services Task Force; USPSTF50|20|15, 2020 USPSTF proposed recommendation: lower age limit of 50, 20+ pack-years, and current smoker or quit within 15 years; USPSTF2013, 2013 USPSTF recommendation
FOR EDITORIAL COMMENT, SEE PAGE 31
Screening for lung cancer with low-dose CT scanning is recommended by major guidelines.1, 2, 3, 4, 5, 6, 7, 8, 9 However, the total number of screening-eligible people in populations is not well known in part because of limited population-based data systems with accurate information (eg, smoking history) for risk assessment.10 Also, criteria for screening eligibility vary across guidelines and considerable debate exists about the adequacy of recommended risk assessment criteria in current guidelines in identifying those at risk for lung cancer developing.11, 12, 13 Research suggest that many people who die of lung cancer do not meet screening-eligible criteria in United States Preventive Services Task Force (USPSTF) recommendations or as set by the Center for Medicare and Medicaid Services (CMS).1,14
In 2013, the USPSTF recommended screening in 55 to 80-year-old people with 30-pack-year smoking histories who are current smokers or quit < 15 years previously.1 Eligibility criteria across guidelines encompass the 50- to 80-year-old age group and a smoking history of 20 pack-years or more. A variety of proposed risk-stratification approaches, such as the PLCOM2012, define eligibility using several known risk factors to generate model-derived risk thresholds.12,15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27 However, it is unclear how the eligible pool compares across guidelines and how screening eligibility mirrors the burden of lung cancer in specific population groups.
Changes to existing guidelines, such as broadening of USPSTF eligibility criteria to a lower minimum age (50 years) and reduced smoking history (20 pack-years) to initiate screening as proposed in 2020, will increase the pool of people eligible for screening.28,29 Low-dose CT screening uptake is very low,30,31 and the novel coronavirus disease 2019 pandemic further decreased screening delivery. In this study, we evaluated the number of people eligible for lung cancer screening according to sex and race or ethnicity across several current and proposed eligibility criteria. We also assessed the number of eligible people as a proportion of incidence as a measure of potential disparities in eligibility in relationship to lung cancer burden. The goal was to inform policy and programmatic decisions regarding screening eligibility and to address disparities in deaths resulting from lung cancer.
Methods
This was a cross-sectional analysis using the adult and cancer control supplement files of the 2015 National Health Interview Survey (NHIS). We restricted our study to 50- to 80-year-old people, which encompasses the age criteria used across all guidelines. Although NHIS data currently are available through 2018, the 2016 through 2018 surveys did not include a cancer control supplement, which had items required for assessing pack-year information in former smokers.32 For current smokers, we derived pack-years from age started smoking regularly, age at date of interview, and number of cigarettes per day smoked. Among former smokers, pack-year also considered years since quitting and number of cigarettes smoked per day when last smoked regularly. Race and ethnicity were categorized as non-Hispanic White, non-Hispanic Black, Hispanic or Latino (hereafter Hispanic), and Asian.
Analysis
Numbers and percentages of the population 50 to 80 years of age who were eligible by criteria used in various guidelines were computed by sex and race or ethnicity. NHIS sampling weights were used to reflect the US population. Multiple imputation was used for missing data, using PROC MI (SAS version 9.4 software, SAS Institute).33 Details of missing data and the imputation approach are given in e-Appendix 1. The variables included in the imputation model included race or ethnicity, sex, smoking history variables, and COPD status.
We based our analyses on eligibility criteria in CMS coverage decisions, the 2013 USPSTF recommendations, and the National Comprehensive Cancer Network (NCCN) guidelines (version 1.2020) (Table 1). The NCCN guidelines incorporate the PLCOM2012 ever smokers risk model. Therefore, we also include estimates based solely on the PLCOM2012 risk model using a 6-year risk cutoff of 1.3% as the threshold for eligibility. Additional alternative criteria expands 2013 USPSTF criteria to a lower age of 50 years and a 20-pack-year smoking history threshold while maintaining the 15-years-since-quitting threshold in former smokers as proposed by the USPSTF in 2020 (USPSTF50|20|15).28 Because a threshold for when lung cancer risk returns to levels in never smokers is not clear and risk remains high up to three decades or more after smoking cessation,34 we further assessed eligibility based on extended quit-year criterion of 25 years (Expanded50|20|25). To clarify differences in eligibility rates, we distinguished between percentage point and proportional differences. For instance, the difference between 25% and 20% was stated as “5 percentage points higher” or a “25% proportional increase.”
Table 1.
Lung Cancer Screening Guidelines
| Guideline | Year Issued | Age, y | Smoking History |
|---|---|---|---|
| US Preventive Services Task Force | 2013 | 55-80 | ≥ 30 Pack-years, current smoker or quit smoking within 15 y |
| National Comprehensive Cancer Network | 2020 | 50+a | ≥ 30 Pack-years, current smoker or quit smoking within 15 y; or ≥ 20 pack-years and PLCOM2012 6-y risk of ≥ 1.3%b |
| Centers for Medicare and Medicaid Services | 2015 | 55-77 | ≥ 30 Pack-years, current smoker or quit smoking within 15 y |
| PLCOM2012 model | Proposed | 50+ | 6-y risk ≥ 1.3%b |
| USPSTF50|20|15 | Proposed | 50-80 | ≥20 Pack-years, current or quit smoking within 15 y |
| Expanded50|20|25 | Proposed | 50-80 | ≥20 Pack-years, current smoker or quit smoking within 25 y |
Expanded50|20|25 = expanded 2020 USPSTF proposed recommendation: lower age limit of 50, 20+ pack-years, current smoker or quit within 25 years; USPSTF50|20|15 = 2020 USPSTF proposed recommendation: lower age limit of 50, 20+ pack-years, and current smoker or quit within 15 years.
Age range 55-77 y with ≥ 30 pack-year criterion and 50+ y with ≥ 20 pack-year and 6-y risk criterion.
Based on 6-y risk calculated using the PLCOM2012 model.
All criteria except PLCOM2012 and NCCN form a nested sequence of increasingly eligible populations, from CMS, USPSTF2013, and USPSTF50|20|15 through Expanded50|20|25. The CMS and USPSTF2013 criteria differ only in the upper age limit (77 years vs 80 years). Therefore, the primary analysis included only USPSTF2013; corresponding CMS percentages are slightly lower. However, because CMS criteria are most relevant for people 65 years of age or older, we performed a separate analysis on the population 65 to 80 years of age based on the CMS, USPSTF2013, and USPSTF50|20|15 criteria.
To assess potential disparities in eligibility across population groups better, we evaluated eligibility rates across criteria and, separately, the corresponding lung cancer incidence in each group. For each group and criteria, we computed an eligibility to incidence (E-I) ratio by dividing the percent eligible by the annual incidence rate, such that an eligibility rate of 10% (or 10,000 per 100,000) with an incidence rate of 100 per 100,000 yields an E-I ratio of 100 (10,000/100). Lung cancer incidence by sex and race or ethnicity for the most recent years available (2015-2017) for the group 50 to 80 years of age or 65 to 80 years of age, as appropriate, were obtained from the Surveillance, Epidemiology and End Results (SEER) database using the SEERSTAT program (National Cancer Institute). A lower E-I ratio in one group compared with another suggests a possible disparity, in that the group with the lower ratio may be underserved with respect to screening. Under certain assumptions, the E-I ratio has the property that, for any given eligibility criteria, two population groups having the same E-I ratio also will have the same proportion of people with lung cancer who are screening eligible. More generally, the ratio of E-I ratios in group A vs B will equal the ratio of the proportion of lung cancer patients who are screening eligible in group A vs B (see e-Appendix 2 for details). We estimated SEs for eligibility rates taking into account the design features of the NHIS.35 For SEs of the E-I ratio, incidence rates were taken as fixed (no sampling error).
Results
Smoking Prevalence
Among the population 50 to 80 years of age, 45.5% were ever smokers and 14.5% were current smokers (Table 2). Among men, ever smoking rates varied little by race or ethnicity. In contrast, rates among women were markedly lower for Asians and Hispanics compared with non-Hispanic Whites. Among men, current smoking rates were higher for non-Hispanic Black men compared with the other groups. Current smokers constituted nearly one-third of ever smokers among both men (32.9%) and women (31.1%), but the percentage was highest among non-Hispanic Black men (44.0%) and women (39.9%) and lowest in Asian women (16.8%) (Table 3).
Table 2.
Estimates of Percent of Participants Eligible for Lung Cancer Screening by Various Criteria in the Total Population and Among Ever Smokers
| Criteria | USPSTF2013 | USPSTF50|20|15 | NCCN | Expanded50|20|25 | PLCOM2012 |
|---|---|---|---|---|---|
| Total population | |||||
| Both sexes, all race groups | 8.3 | 14.8 | 13.3 | 17.5 | 12.8 |
| Both sexes, all race groups | 8.3 (0.3) | 14.9 (0.4) | 13.4 (0.3) | 17.6 (0.4) | 12.9 (0.3) |
| Men | |||||
| All race groups | 10.2 (0.5) | 17.6 (0.6) | 16.4 (0.6) | 21.1 (0.7) | 15.7 (0.6) |
| NH Whites | 11.7 (0.6) | 19.7 (0.8) | 18.7 (0.7) | 23.7 (0.8) | 17.8 (0.7) |
| NH Blacks | 7.4 (0.9) | 15.1 (1.2) | 14.8 (1.1) | 17.2 (1.2) | 15.7 (1.1) |
| Hispanics | 3.6 (0.6) | 7.5 (1.0) | 4.8 (0.71) | 10.7 (1.2) | 3.8 (0.7) |
| Asians | 6.6 (1.8) | 9.4 (2.2) | 7.3 (1.8) | 10.6 (2.3) | 4.6 (0.9) |
| Women | |||||
| All race groups | 6.6 (0.4) | 12.4 (0.5) | 10.7 (0.4) | 14.4 (0.5) | 10.4 (0.4) |
| NH Whites | 8 (0.4) | 14.9 (0.6) | 13.1 (0.6) | 17.2 (0.7) | 12.8 (0.6) |
| NH Blacks | 3.9 (0.6) | 7.6 (0.7) | 6.4 (0.7) | 8.5 (0.8) | 7.4 (0.6) |
| Hispanics | 2.2 (0.4) | 4.7 (0.5) | 2.5 (0.4) | 5.2 (0.5) | 1 (0.2) |
| Asians | 0.7 (0.3) | 2.2 (0.4) | 1.1 (0.3) | 2.5 (0.5) | 0.9 (0.3) |
| Ever smokers | |||||
| Both sexes, all race groups | 18.3 (0.6) | 32.7 (0.7) | 29.4 (0.7) | 38.7 (0.8) | 28.3 (0.7) |
| Men | |||||
| All race groups | 20 (0.8) | 34.3 (1) | 31.9 (1.0) | 41.2 (1.1) | 30.5 (1.0) |
| NH Whites | 22.3 (1.0) | 37.5 (1.2) | 35.5 (1.2) | 45.1 (1.3) | 34 (1.1) |
| NH Blacks | 15 (1.6) | 30.5 (2.1) | 30 (2.0) | 34.8 (2.1) | 31.8 (1.9) |
| Hispanics | 7.8 (1.1) | 16.2 (1.4) | 10.5 (1.2) | 23.2 (1.8) | 8.3 (1.1) |
| Asians | 14.5 (3.6) | 20.9 (4.3) | 16.2 (3.7) | 23.5 (4.3) | 10.2 (1.7) |
| Women | |||||
| All race groups | 16.4 (0.8) | 30.9 (1.0) | 26.5 (0.9) | 35.7 (1.1) | 25.8 (0.9) |
| NH Whites | 17.8 (0.9) | 33.1 (1.2) | 29.2 (1.1) | 38.4 (1.2) | 28.5 (1.1) |
| NH Blacks | 10.8 (1.3) | 21.2 (1.6) | 17.7 (1.3) | 23.4 (1.8) | 20.4 (1.4) |
| Hispanics | 8.8 (1.6) | 18.5 (1.8) | 9.7 (1.6) | 20.5 (1.8) | 3.8 (0.7) |
| Asians | 7 (1.1) | 20.1 (2.0) | 10.6 (1.3) | 23.6 (2.1) | 8.2 (1.2) |
Data are presented as No. eligible in millions or percent eligible (SE). Expanded50|20|25 = expansion of 2020 USPSTF proposed recommendation: lower age limit of 50, 20+ pack-y; NCCN = National Comprehensive Cancer Network; NH = non-Hispanic; USPSTF50|20|15 = 2020 USPSTF proposed recommendation: lower age limit of 50, 20+ pack-years, and current smoker or quit within 15 years; USPSTF2013 = 2013 USPSTF recommendation.
Table 3.
Eligible Estimates of Percent of Participants Eligible for Lung Cancer Screening by Various Criteria: Current Smokers and Former Smokers
| Criteria | USPSTF2013 | USPSTF50|20|15 | NCCN | Expanded50|20|25 | PLCOM2012 |
|---|---|---|---|---|---|
| Current smokers | |||||
| Both sexes, all race groups | 27.5 (1.2) | 56.8 (1.3) | 40.5 (1.2) | 56.8 (1.3) | 41.3 (1.2) |
| Men | |||||
| All race groups | 32.8 (1.4) | 61.7 (1.4) | 45.6 (1.6) | 61.7 (1.4) | 46 (1.6) |
| NH Whites | 39.5 (1.5) | 71.3 (1.7) | 53.4 (1.8) | 71.3 (1.7) | 53.5 (1.7) |
| NH Blacks | 20.7 (2.2) | 44.1 (3.6) | 39 (3.4) | 44.1 (3.6) | 43.1 (3.4) |
| Hispanics | 9.5 (1.4) | 27.4 (1.7) | 11.5 (1.6) | 27.4 (1.7) | 9.6 (1.6) |
| Asians | 11 (1.3) | 24.6 (2.6) | 12.7 (1.3) | 24.6 (2.6) | 8.5 (1.3) |
| Women | |||||
| All race groups | 21.3 (1.3) | 51.5 (1.8) | 34.8 (1.4) | 51.5 (1.8) | 36.1 (1.5) |
| NH Whites | 24.2 (1.5) | 57.1 (2.0) | 39.5 (1.5) | 57.1 (2.0) | 40.7 (1.6) |
| NH Blacks | 10.8 (2.2) | 30 (2.8) | 21.5 (2.2) | 30 (2.8) | 25.9 (2.3) |
| Hispanics | 5 (0.2) | 26.2 (1.9) | 5 (0.2) | 26.2 (1.9) | 4.9 (0.9) |
| Asians | 13.5 (0.6) | 33.7 (3.3) | 13.5 (0.6) | 33.7 (3.3) | 6.4 (0.3) |
| Former smokers | |||||
| Both sexes, all race groups | 14 (0.6) | 21.4 (0.8) | 24.3 (0.8) | 30.2 (0.8) | 22.3 (0.7) |
| Men | |||||
| All race groups | 14.2 (0.9) | 21.9 (1.1) | 25.8 (1.1) | 32 (1.2) | 23.5 (1) |
| NH Whites | 15.1 (1.0) | 23.3 (1.2) | 28.1 (1.2) | 34.1 (1.4) | 25.9 (1.2) |
| NH Blacks | 10.5 (1.7) | 19.8 (1.7) | 22.9 (2.0) | 27.5 (2.0) | 22.9 (1.8) |
| Hispanics | 7.1 (1.2) | 11.3 (1.4) | 10.5 (1.3) | 21.4 (1.7) | 7.7 (1.1) |
| Asians | 16.1 (5.4) | 19.3 (5.4) | 17.7 (5.4) | 23 (5.4) | 10.9 (2.5) |
| Women | |||||
| All race groups | 13.8 (0.9) | 20.8 (1.1) | 22.5 (1.1) | 28 (1.2) | 20.7 (1.1) |
| NH Whites | 14.4 (1.1) | 21.7 (1.2) | 24.3 (1.3) | 29.6 (1.4) | 22.7 (1.3) |
| NH Blacks | 10.6 (1.2) | 15.4 (1.3) | 15.1 (1.1) | 19.3 (1.4) | 16.8 (1.4) |
| Hispanics | 10.7 (2.4) | 14.6 (2.4) | 12 (2.4) | 17.7 (2.5) | 3.3 (0.9) |
| Asians | 5.6 (0.4) | 17.4 (1.4) | 10 (0.8) | 21.6 (1.5) | 8.6 (0.7) |
Data are presented as percent eligible (SE). Expanded50|20|25 = expansion of 2020 USPSTF proposed recommendation: lower age limit of 50, 20+ pack-y; NCCN = National Comprehensive Cancer Network; NH = non-Hispanic; USPSTF50|20|15 = 2020 USPSTF proposed recommendation: lower age limit of 50, 20+ pack-years, and current smoker or quit within 15 years; USPSTF2013 = 2013 USPSTF recommendation.
Screening Eligibility
The numbers eligible across criteria were 8.3 million for USPSTF2013, 13.3 million for NCCN, and 12.8 million for PLCOM2012 criteria (Table 2). About 14.8 million and 17.5 million people were eligible with USPSTF50|20|15 and Expanded50|20|25 criteria, respectively. In analysis by sex, the overall percent eligible across the five criteria ranged from 10.2% to 21.1% for men and from 6.6% to 14.4% for women; among ever smokers, it was 20.0% to 41.2% for men and 16.4% to 35.7% for women (Table 2).
By race or ethnicity, eligibility rates consistently were higher across criteria for non-Hispanic Whites than the other groups (Table 2). Among ever smokers, compared with non-Hispanic Whites, rates were substantially lower for Asians and Hispanics, with smaller differences for non-Hispanic Blacks.
Table 3 shows percent eligible in current and former smokers. Across all criteria examined, overall eligibility rates were 14 to 35 percentage points higher in current than former smokers. However, analysis by sex and race or ethnicity showed greater variability in eligibility across race or ethnicity groups for current than former smokers. For example, USPSTF50|20|15 criteria eligibility among male current smokers was 71.3% in non-Hispanic Whites, 44% in non-Hispanic Blacks, 27.4% in Hispanics, and 24.6% in Asians. In contrast, for male former smokers, the range was 11.3% to 23.3%, with the lowest rate among Hispanics. For NCCN criteria among women, the percent eligible by race or ethnicity range was 5.0% to 39.5% for current smokers and 10.0% to 24.3% among former smokers.
Among former smokers, using the Expanded50|20|25 criteria increased the percent eligible by 8.8 percentage points over USPSTF50|20|15 criteria (30.2% vs 21.4%), with a larger difference in men than in women (10.1% vs 7.2%). Among men, the difference was largest among non-Hispanic Whites (10.8%) and smallest among Asians (3.7%). Among women, the difference was largest for non-Hispanic Whites (7.9%) relative to non-Hispanic Blacks (3.9%), Hispanics (3.1%), and Asians (4.2%).
Figure 1 shows E-I ratios by sex and race or ethnicity for each eligibility criteria. Across all criteria, the E-I ratio was lower for women than men in all racial and ethnic groups (data not shown). The E-I ratio was modestly lower in women among non-Hispanic Whites (11%-34% lower) and non-Hispanic Blacks (12%-21% lower), but the difference was larger among Asians (62%-83% lower).
Figure 1.
A, B, Bar graphs showing the ratio of percent eligible to annual incidence rate (eligibility to incidence [E-I] ratio) by race or ethnicity for men (A) and women (B). An eligibility rate of 10% (or 10,000 per 100,000) and an incidence rate of 100 per 100,000 results in an E-I ratio of 100. Incidence rates (per 100,000) for non-Hispanic Whites, non-Hispanic Blacks, Hispanics, and Asians were 190, 239, 87, and 124, respectively, for men and 163, 146, 68, and 78, respectively, for women. Black vertical bars indicate SE of estimates. Expanded50|20|25 = expansion of 2020 USPSTF proposed recommendation: lower age limit of 50, 20+ pack-y; NCCN = National Comprehensive Cancer Network; NH = non-Hispanic; USPSTF50|20|15 = 2020 USPSTF proposed recommendation: lower age limit of 50, 20+ pack-years, and current smoker or quit within 15 years; USPSTF2013 = 2013 USPSTF recommendation.
Across all criteria, among men, non-Hispanic Whites had the highest E-I ratios. Compared with non-Hispanic White men, the E-I ratio was 30% to 50% lower for non-Hispanic Black men, 1% to 53% lower among Hispanic men, and 12% to 61% lower for Asian men. Among women, the ratio also was highest among non-Hispanic Whites across all criteria. Compared with non-Hispanic Whites, the ratio was 35% to 45% lower for non-Hispanic Blacks, 24% to 81% lower for Hispanics, and 70% to 85% lower for Asians across all criteria.
Subanalysis of People 65 to 80 Years of Age
In a subanalysis of people 65 to 80 years of age based on CMS, USPSTF2013, and USPSTF50|20|15 criteria, overall eligibility rates were about 4 percentage points higher in men than women; sex differences were smaller in ever smokers (Table 4). The E-I ratio was modestly higher in men than in women for all criteria (Table 5). Using CMS and USPSTF2013 criteria, in both men and women, eligibility rates were highest for non-Hispanic Whites overall and among ever smokers; the E-I ratio was also generally highest for non-Hispanic Whites among both men and women. A similar pattern was seen in women for the USPSTF50|20|15 criteria. In contrast, for the USPSTF50|20|15 criteria among men, non-Hispanic Blacks showed higher eligibility than non-Hispanic Whites, both overall (21.7% vs 16.4%) and among ever smokers (37.2% vs 27.1%), and the E-I ratio was slightly higher for non-Hispanic Blacks than non-Hispanic Whites (50.9% vs 46.7%).
Table 4.
Estimates of Percent of Participants Eligible for Lung Cancer Screening by Various Criteria: Population 65 to 80 Years of Age
| Criteria | Total Population |
Ever Smokers |
||||
|---|---|---|---|---|---|---|
| CMS | USPSTF2013 | USPSTF50|20|15 | CMS | USPSTF2013 | USPSTF50|20|15 | |
| Both sexes, all race groups | 9.0 (0.5) | 9.7 (0.5) | 13.6 (0.6) | 18.1 (0.9) | 19.6 (0.9) | 27.5 (1.0) |
| Men | ||||||
| All race groups | 11.2 (0.8) | 12.2 (0.8) | 16.0 (0.9) | 18.6 (1.3) | 20.3 (1.2) | 26.9 (1.3) |
| NH Whites | 11.9 (0.9) | 12.9 (0.9) | 16.4 (1.0) | 19.6 (1.3) | 21.2 (1.3) | 27.0 (1.4) |
| NH Blacks | 9.9 (1.7) | 11.5 (1.7) | 21.5 (2.1) | 16.6 (2.1) | 19.6 (2.1) | 36.9 (2.6) |
| Hispanics | 5.6 (1.2) | 7.6 (1.4) | 9.8 (1.5) | 9.5 (1.9) | 12.9 (1.9) | 16.6 (2.1) |
| Asians | 6.8 (1.9) | 7.1 (1.8) | 8.9 (2.1) | 13.3 (3.2) | 13.8 (3.2) | 17.5 (3.9) |
| Women | ||||||
| All race groups | 7.0 (0.6) | 7.6 (0.6) | 11.5 (0.7) | 17.3 (1.3) | 18.6 (1.3) | 28.2 (1.5) |
| NH Whites | 8.5 (0.7) | 9.1 (0.7) | 13.7 (0.9) | 18.9 (1.5) | 20.4 (1.5) | 30.8 (1.7) |
| NH Blacks | 4.6 (0.9) | 5.0 (1.0) | 7.0 (1.0) | 11.7 (1.7) | 12.9 (1.6) | 17.8 (1.8) |
| Hispanics | 1.2 (1.0) | 1.5 (1.0) | 2.4 (1.1) | 4.5 (0.5) | 5.7 (0.5) | 9.5 (2.0) |
| Asians | 0.2 (0.2) | 0.2 (0.2) | 2.3 (0.3) | 2.3 (0.1) | 2.3 (0.1) | 24.7 (1.1) |
Data are presented as percent eligible (SE). CMS = Center for Medicare and Medicaid Services; NH = non-Hispanic; USPSTF50|20|15 = 2020 USPSTF proposed recommendation: lower age limit of 50, 20+ pack-years, and current smoker or quit within 15 years; USPSTF2013 = 2013 USPSTF recommendation.
Table 5.
Estimates of E-I Ratio by Various Criteria: Population 65 to 80 Years of Age
| Criteria | CMS | USPST2013 | USPSTF50|20|15 |
|---|---|---|---|
| Both sexes, all race groups | 29.8 (1.7) | 32.4 (1.7) | 45.5 (1.9) |
| Men | |||
| All race groups | 32.5 (2.4) | 35.7 (2.4) | 47.2 (2.6) |
| NH Whites | 33.9 (2.6) | 36.8 (2.6) | 46.7 (2.8) |
| NH Blacks | 23.2 (3.9) | 27.0 (4.0) | 50.5 (4.8) |
| Hispanics | 30.9 (6.6) | 43.4 (8.2) | 56.0 (8.6) |
| Asians | 29.2 97.9) | 30.5 (7.9) | 38.2 (8.9) |
| Women | |||
| All race groups | 26.8 (2.2) | 28.8 (2.2) | 43.6 (2.7) |
| NH Whites | 29.1 (2.5) | 31.2 (2.5) | 46.9 (3.0) |
| NH Blacks | 18.8 (3.6) | 20.5 (4.0) | 28.7 (4.2) |
| Hispanics | 9.4 (7.5) | 11.7 (7.6) | 18.8 (8.4) |
| Asians | 1.5 (1.6) | 1.5 (1.6) | 16.9 (2.3) |
Data are presented as E-I ratio (SE). E-I = eligibility to incidence; CMS = Center for Medicare and Medicaid Services; NH = non-Hispanic; USPSTF50|20|15 = 2020 USPSTF proposed recommendation: lower age limit of 50, 20+ pack-years, and current smoker or quit within 15 years; USPSTF2013 = 2013 USPSTF recommendation.
Discussion
Under the criteria used in the guidelines examined in those 50 to 80 years of age, eligibility for lung cancer screening ranged from 8.3 to 13.3 million people, representing 8.3% to 13.4%, overall, 18.3% to 29.4% of ever smokers, and 27.5% to 40.5% of current smokers. The highest proportion eligible was observed consistently among non-Hispanic Whites. Lowering the age or pack-year history or extending the quit-year cutoff increased eligibility for lung screening, but substantial differences remained by race or ethnicity. Differences in overall eligibility rates, in part, reflect smoking patterns. However, even among ever smokers, substantial differences exist across groups. Among women who were ever smokers, compared with non-Hispanic Whites, proportionally substantially fewer eligible non-Hispanic Blacks, Hispanics, and Asians were found using USPSTF2013, NCCN, and CMS criteria. Among ever smoking men, similar differences were found between non-Hispanic Whites and other racial or ethnic groups, although White-to-Black differences were of smaller magnitude. Other studies using the NHIS data showed, for men and women combined, lower eligibility rates for non-Hispanic Black ever smokers compared with non-Hispanic White ever smokers using both USPSTF2013 and USPSTF50|20|15 criteria.12,13
Differences in eligibility across groups may not necessarily denote disparities because lower eligibility rates may reflect lower smoking exposure. A difference could be considered a potential disparity if a group has a lower eligibility rate given similar lung cancer incidence rates. We assessed that possibility using group-specific ratios of screening eligibility to lung cancer incidence rates (E-I ratio). Non-Hispanic Whites consistently showed the highest E-I ratios of the racial or ethnic groups across all criteria examined. Compared with non-Hispanic White men, non-Hispanic Black men had higher lung cancer incidence rates, but had lower eligibility rates and E-I ratios. Using USPSTF50|20|15 criteria, 23% proportionally fewer eligible non-Hispanic Black men than non-Hispanic White men were found and a 39% lower E-I ratio for non-Hispanic Black men was found. For some criteria, Asian and Hispanic men and women showed markedly lower E-I ratios than their non-Hispanic White counterparts.
Among those 65 years of age and older, the group for whom the CMS coverage criteria are most relevant, the eligibility rate and E-I ratio also were lower for non-Hispanic Black men than non-Hispanic White men. However, using USPSTF50|20|15 criteria, both the percent eligibility and E-I ratio were higher for non-Hispanic Black men than non-Hispanic White men. Yet, the non-Hispanic Black vs non-Hispanic White discrepancy in the E-I ratio among men was greater in those 50 to 64 years of age (59% lower E-I ratio for non-Hispanic Blacks) than the difference in the overall group 50 to 80 years of age (39% lower). The higher E-I ratio among non-Hispanic Blacks 65 to 80 years of age using USPSTF50|20|15 criteria related to higher eligibility among Blacks in those 65 to 80 years of age (21.7%) than those younger than 65 years of age (12.2%). The reverse was found in non-Hispanic Whites (21.7% in those younger than 65 years vs 16.4% in those 65 to 80 years of age). The contrast is likely because of lower quit rates in older non-Hispanic Black men, resulting in higher current smoking prevalence than in non-Hispanic White men. However, average reported cigarettes per day rates are lower in non-Hispanic Black men than in non-Hispanic White men. Taking the median age at smoking initiation and cigarettes per day for currently smoking non-Hispanic White men and currently smoking non-Hispanic Black men, the former would reach 20 pack-years by 41 years of age, compared with 57 years of age for the latter group. This also explains the finding that among current smokers who are 50 to 80 years of age, non-Hispanic White men showed substantially higher eligibility rates using most criteria than non-Hispanic Black men (71% vs 44% using USPSTF50|20|15 criteria). The discordance between smoking rates and lung cancer incidence suggests that existing risk-stratification approaches do not identify at-risk populations adequately among those who have disproportionately high mortality rates, particularly Black men. To address disparities, it will be important to consider both smoking duration and intensity in risk assessment, especially among current smokers, including those with a history of fewer than 20 pack-years.36, 37, 38
Race or ethnicity is considered in the PLCOM2012 model, but disparity in eligibility between non-Hispanic Black men and non-Hispanic White men was not reduced appreciably with NCCN criteria, despite incorporating the PLCOM2012 model. Consistent with other studies, disparities between non-Hispanic Blacks and non-Hispanic Whites were lower with the PLCOM2012 model thresholds alone.11,39 However, even with that model, eligibility rates were lower for non-Hispanic Black men than for non-Hispanic White men. Even a longer quit-year cutoff for former smokers than currently recommended leads to higher eligibility rates among non-Hispanic White men than non-Hispanic Black men. This persistent gap in the E-I ratio may be driven by incomplete capture of lung cancer risk among Black men by current eligibility criteria.
Our study has some limitations. The E-I ratio is not an exact measure of disparities or of the proportion of people with lung cancer who are screening eligible in each group. For instance, we observed a lower E-I ratio among women compared with men, which is consistent with a study that found a higher proportion of lung cancers among never smokers in women compared with men (15.6% vs 9.6%).40 Because causes beyond smoking exist, it is possible for two groups to have similar incidence, but differential eligibility rates and E-I ratios. In such cases, a higher E-I ratio could mean a greater potential for harms from screening. However, with an assumption of equal lung cancer incidence across groups among those eligible, a hypothetical E-I ratio that is twice as high in one group (group A) than another group (group B) would reflect twice the proportion of lung cancer patients who are eligible in group A vs B. Although the equal incidence assumption is unlikely to hold strictly, to the extent that it holds approximately, the E-I ratios across groups also will reflect the disparity in eligibility approximately among lung cancer patients across groups.
We were unable to determine directly which lung cancer patients met various eligibility criteria because our analysis used separate sources of data for incidence (SEER) and eligibility rates (NHIS). Although relatively large cohorts exist with both smoking exposure and cancer incidence data, such as the Prostate, Lung, Colorectal and Ovarian (PLCO) cancer screening trial cohort, they are not representative of the overall US population.41
Another limitation is the use of data from 2015. However, the percentage of current and former smokers in our analysis (14.3% and 31.0%, respectively) was similar to the prevalence in the 2018 NHIS data (13.9% and 30.0%, respectively). The percentage of current smokers meeting the new eligibility criteria also was similar in 2015 (56.9%) and 2018 (55.2%). Because of low representation of Native Americans among respondents in the NHIS, we were unable to generate reliable estimates for this group. Self-reported smoking history may be subject to recall and social desirability biases that may differ by race or ethnicity. We did not assess other risk factors that may affect the E-I ratio, such as non-smoking-related environmental exposure.
This study also has important implications for delivery to lung cancer screening. Not all eligible people, after shared decision-making, will undergo screening, although a reasonable goal is an acceptance rate of more than 50%. Before the COVID-19 pandemic, less than 10% of eligible people were undergoing low-dose CT scan screening.30 Therefore, the goal to decrease the mortality rate from, as well as disparities associated with, lung cancer by enlarging the eligible pool will require even greater efforts to increase screening delivery and uptake among those eligible, particularly among non-Hispanic Black men, who are at highest risk, but traditionally have lagged in the receipt of cancer prevention and early detection services.
Conclusions
A wide range exists in the percent of people eligible for lung cancer screening across the criteria used for risk assessment across guidelines according sex and race or ethnicity, even when controlling for the lung cancer incidence rate. Risk assessment criteria may need to consider smoking duration as well as intensity to address more fully discordance in eligibility relative to disease burden, which is particularly relevant for non-Hispanic Black men, who have the highest risk of lung cancer and who have longer smoking duration, but lower reported intensity.
Take-home Points.
Study Question: Do the eligibility criteria for lung cancer screening identify equal proportions of people by sex and race or ethnicity and do eligibility rates track according to lung cancer risk within each group?
Results: For most guidelines, the percentage of ever smokers eligible for screening is proportionally lower in women and particularly for Black men, who have higher a risk of lung cancer than White men.
Interpretation: Current criteria used for selecting people for lung cancer screening create inequities in eligibility according to race or ethnicity and sex.
Acknowledgments
Author contributions: P. F. Pinsky takes responsibility for the content of the manuscript, including the data and analysis. P. F. P., C. A. D., and Y. K. L. contributed to the study design, data analysis and interpretation, and writing of the manuscript.
Financial/nonfinancial disclosures: None declared.
Additional information: The e-Appendixes can be found in the Supplemental Materials section of the online article.
Footnotes
FUNDING/SUPPORT: The authors have reported to CHEST that no funding was received for this study.
DISCLAIMER: C. A. D. is a member of the US Preventive Services Task Force. The views expressed are those of the authors and do not represent the views of policies of the USPSTF.
Supplementary Data
References
- 1.Moyer V.A., U. S. Preventive Services Task Force Screening for lung cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2014;160(5):330–338. doi: 10.7326/M13-2771. [DOI] [PubMed] [Google Scholar]
- 2.Wood D.E., Kazerooni E.A., Baum S.L. Lung cancer screening, version 3.2018, NCCN clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2018;16(4):412–441. doi: 10.6004/jnccn.2018.0020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Jaklitsch M.T., Jacobson F.L., Austin J.H. The American Association for Thoracic Surgery guidelines for lung cancer screening using low-dose computed tomography scans for lung cancer survivors and other high-risk groups. J Thorac Cardiovasc Surg. 2012;144(1):33–38. doi: 10.1016/j.jtcvs.2012.05.060. [DOI] [PubMed] [Google Scholar]
- 4.Wender R., Fontham E.T., Barrera E., Jr. American Cancer Society lung cancer screening guidelines. CA Cancer J Clin. 2013;63(2):107–117. doi: 10.3322/caac.21172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Detterbeck F.C., Mazzone P.J., Naidich D.P., Bach P.B. Screening for Lung Cancer: Diagnosis and Management of Lung Cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2013;143(5 suppl):e78S–e92S. doi: 10.1378/chest.12-2350. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.American Lung Association Lung Cancer Screening Committee Providing guidance on lung cancer screening to patients and physicians. 2015. American Lung Association website. http://www.lung.org/assets/documents/lung-cancer/lung-cancer-screening-report.pdf
- 7.Canadian Task Force on Preventive Health Care Recommendations on screening for lung cancer. CMAJ. 2016;188(6):425–432. doi: 10.1503/cmaj.151421. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Oudkerk M., Devaraj A., Vliegenthart R. European position statement on lung cancer screening. Lancet Oncol. 2017;18(12):e754–e766. doi: 10.1016/S1470-2045(17)30861-6. [DOI] [PubMed] [Google Scholar]
- 9.Mazzone P.J., Silvestri G.A., Patel S. Screening for lung cancer: CHEST guideline and expert panel report. Chest. 2018;153(4):954–985. doi: 10.1016/j.chest.2018.01.016. [DOI] [PubMed] [Google Scholar]
- 10.Doubeni C.A., Wilkinson J.M., Korsen N., Midthun D.E. Lung cancer screening guidelines implementation in primary care: a call to action. Ann Fam Med. 2020;18(3):196–201. doi: 10.1370/afm.2541. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Han S.S., Chow E., Ten Haaf K. Disparities of national lung cancer screening guidelines in the U.S. population. J Natl Cancer Inst. 2020 doi: 10.1093/jnci/djaa013. 112:1136-1142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Katki H.A., Kovalchik S.A., Berg C.D., Cheung L.C., Chaturvedi A.K. Development and validation of risk models to select ever-smokers for CT lung cancer screening. JAMA. 2016;315(21):2300–2311. doi: 10.1001/jama.2016.6255. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Rivera M.P., Katki H.A., Tanner N.T. Addressing disparities in lung cancer screening eligibility and healthcare access. An official American Thoracic Society statement. Am J Respir Crit Care Med. 2020;202(7):e95–e112. doi: 10.1164/rccm.202008-3053ST. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Centers for Medicare and Medicaid Services Final national coverage determination on screening for lung cancer with low dose computed tomography (LDCT). 2015. Centers for Medicare and Medicaid Services website. https://www.cms.gov/medicare-coverage-database/details/nca-decision-memo.aspx?NCAId=274
- 15.Pinsky P.F., Berg C.D. Applying the National Lung Screening Trial eligibility criteria to the US population: what percent of the population and of incident lung cancers would be covered? J Med Screen. 2012;19(3):154–156. doi: 10.1258/jms.2012.012010. [DOI] [PubMed] [Google Scholar]
- 16.National Lung Screening Trial Research Team. Aberle D.R., Adams A.M. 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]
- 17.Jett J.R., Midthun D.E. Screening for lung cancer: for patients at increased risk for lung cancer, it works. Ann Intern Med. 2011;155(8):540–542. doi: 10.7326/0003-4819-155-8-201110180-00367. [DOI] [PubMed] [Google Scholar]
- 18.Kovalchik S.A., Tammemagi M., Berg C.D. Targeting of low-dose CT screening according to the risk of lung-cancer death. N Engl J Med. 2013;369(3):245–254. doi: 10.1056/NEJMoa1301851. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Raji O.Y., Duffy S.W., Agbaje O.F. Predictive accuracy of the Liverpool Lung Project risk model for stratifying patients for computed tomography screening for lung cancer: a case-control and cohort validation study. Ann Intern Med. 2012;157(4):242–250. doi: 10.7326/0003-4819-157-4-201208210-00004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.McRonald F.E., Yadegarfar G., Baldwin D.R. The UK Lung Screen (UKLS): demographic profile of first 88,897 approaches provides recommendations for population screening. Cancer Prev Res (Phila) 2014;7(3):362–371. doi: 10.1158/1940-6207.CAPR-13-0206. [DOI] [PubMed] [Google Scholar]
- 21.Cassidy A., Myles J.P., van Tongeren M. The LLP risk model: an individual risk prediction model for lung cancer. Br J Cancer. 2008;98(2):270–276. doi: 10.1038/sj.bjc.6604158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Tammemagi M.C., Katki H.A., Hocking W.G. Selection criteria for lung-cancer screening. N Engl J Med. 2013;368(8):728–736. doi: 10.1056/NEJMoa1211776. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Maisonneuve P., Bagnardi V., Bellomi M. Lung cancer risk prediction to select smokers for screening CT—a model based on the Italian COSMOS trial. Cancer Prev Res (Phila) 2011;4(11):1778–1789. doi: 10.1158/1940-6207.CAPR-11-0026. [DOI] [PubMed] [Google Scholar]
- 24.Bach P.B., Kattan M.W., Thornquist M.D. Variations in lung cancer risk among smokers. J Natl Cancer Inst. 2003;95(6):470–478. doi: 10.1093/jnci/95.6.470. [DOI] [PubMed] [Google Scholar]
- 25.Ten Haaf K., Jeon J., Tammemagi M.C. Risk prediction models for selection of lung cancer screening candidates: a retrospective validation study. PLoS Med. 2017;14(4) doi: 10.1371/journal.pmed.1002277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Roe O.D., Markaki M., Tsamardinos I. ‘Reduced’ HUNT model outperforms NLST and NELSON study criteria in predicting lung cancer in the Danish screening trial. BMJ Open Respir Res. 2019;6(1) doi: 10.1136/bmjresp-2019-000512. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Markaki M., Tsamardinos I., Langhammer A., Lagani V., Hveem K., Roe O.D. A validated clinical risk prediction model for lung cancer in smokers of all ages and exposure types: a HUNT study. EBioMedicine. 2018;31:36–46. doi: 10.1016/j.ebiom.2018.03.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.U. S. Preventive Services Task Force Lung cancer screening: draft recommendation statement. 2020. U. S. Preventive Services Task Force website. https://www.uspreventiveservicestaskforce.org/uspstf/draft-recommendation/lung-cancer-screening-2020 Accessed October 6, 2020.
- 29.Richards T.B., Doria-Rose V.P., Soman A. Lung cancer screening inconsistent with U.S. Preventive Services Task Force recommendations. Am J Prev Med. 2019;56(1):66–73. doi: 10.1016/j.amepre.2018.07.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Pham D., Bhandari S., Pinkston C., Oechsli M., Kloecker G. Lung cancer screening registry reveals low-dose CT screening remains heavily underutilized. Clin Lung Cancer. 2019 doi: 10.1016/j.cllc.2019.09.002. 2020;21(3):e206-e211. [DOI] [PubMed] [Google Scholar]
- 31.Odahowski C.L., Zahnd W.E., Eberth J.M. Challenges and opportunities for lung cancer screening in rural America. J Am Coll Radiol. 2019;16(4 Pt B):590–595. doi: 10.1016/j.jacr.2019.01.001. [DOI] [PubMed] [Google Scholar]
- 32.National Center for Health Statistics National Health Interview Survey. Centers for Disease Control and Prevention website. http://cdc.gov/nchs/nhis/index.htm
- 33.SAS Institute . SAS Institute; Cary, NC: 2020. Multiple imputation for missing data: concepts and new development [computer program]. Version 9.0. Cary, NC: SAS Institute; 2020. [Google Scholar]
- 34.Pinsky P.F., Zhu C.S., Kramer B.S. Lung cancer risk by years since quitting in 30+ pack year smokers. J Med Screen. 2015;22(3):151–157. doi: 10.1177/0969141315579119. [DOI] [PubMed] [Google Scholar]
- 35.Centers for Disease Control and Prevention Variance estimation guidance, NHIS 2006-2015 (adapted from the 2006-2015 NHIS survey description documents). 2016. Centers for Disease Control and Prevention website. https://www.cdc.gov/nchs/data/nhis/2006var.pdf
- 36.Remen T., Pintos J., Abrahamowicz M., Siemiatycki J. Risk of lung cancer in relation to various metrics of smoking history: a case-control study in Montreal. BMC Cancer. 2018;18(1):1275. doi: 10.1186/s12885-018-5144-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Rachet B., Siemiatycki J., Abrahamowicz M., Leffondre K. A flexible modeling approach to estimating the component effects of smoking behavior on lung cancer. J Clin Epidemiol. 2004;57(10):1076–1085. doi: 10.1016/j.jclinepi.2004.02.014. [DOI] [PubMed] [Google Scholar]
- 38.Meza R., Hazelton W.D., Colditz G.A., Moolgavkar S.H. Analysis of lung cancer incidence in the Nurses’ Health and the Health Professionals’ follow-up studies using a multistage carcinogenesis model. Cancer Causes Control. 2008;19(3):317–328. doi: 10.1007/s10552-007-9094-5. [DOI] [PubMed] [Google Scholar]
- 39.Pasquinelli M.M., Tammemagi M.C., Kovitz K.L. Risk prediction model versus United States Preventive Services Task Force lung cancer screening eligibility criteria: reducing race disparities. J Thorac Oncol. 2020;15(11):1738–1747. doi: 10.1016/j.jtho.2020.08.006. [DOI] [PubMed] [Google Scholar]
- 40.Siegel D.A., Fedewa S.A., Henley S.J., Pollack L.A., Jemal A. Proportion of never smokers among men and women with lung cancer in 7 US states. JAMA Oncol. 2020;7:302–304. doi: 10.1001/jamaoncol.2020.6362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Black A., Huang W.Y., Wright P. PLCO: evolution of an epidemiologic resource and opportunities for future studies. Rev Recent Clin Trials. 2015;10(3):238–245. doi: 10.2174/157488711003150928130654. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.