Letters
RESEARCH LETTER
Lung Cancer Screening With
Low-Dose Computed Tomography
in the United States—2010 to 2015
Lung cancer is the most preventable and leading cause of
cancer deaths in the United States, with about 155 870
deaths each year.
1
In December 2013, the United States
Preventive Services Task Force (USPSTF) recommended
annual screening for lung cancer with low-dose computed
tomography (LDCT) for asymptomatic persons aged 55
to 80 years who have a 30 pack or more per year smoking
history and currently smoke or have quit within the
past 15 years.
2
According to the 2010 National Health Inter-
view Survey (NHIS), only 2% to 4% of high-risk smokers
received LDCT for lung cancer screening in the pre-
vious year.
3
In this study, we examined whether
LDCT screening has increased following the USPSTF recom-
mendation.
Methods | We used the 2010 and 2015 NHIS, which included
2347 respondents who met the USPSTF criteria for LDCT.
2
Self-
Table 1. Prevalence of LDCT Testing for Lung Cancer in the Past Year Among Screening-Eligible and Noneligible Smokers, National Health Interview
Surveys 2010 and 2015
a,b
Characteristic
Total 2010 2015
P Value
c
No. (%) (95% CI) No. (%) (95% CI) No. (%) (95% CI)
Screening-eligible smokers
(n = 2167)
Weighted No. receiving LDCT
d
276 700 262 700
Weighted No. eligible for LDCT 8 456 800 6 819 500
Total 2167 (3.5) (2.6-4.8) 1036 (3.3) (2.3-4.7) 1131 (3.9) (2.4-6.2) .60
Smoking history
Former, ≥30 PY, quit ≤15 years
ago
1020 (4.2) (2.7-6.5) 491 (4.0) (2.6-6.1) 529 (4.6)
e
(2.1-9.4)
e
.76
Current, ≥30 PY 1147 (2.9) (1.8-4.5) 545 (2.6)
e
(1.4-4.9)
e
602 (3.2) (1.8-5.6) .64
Age, y
55-64 1119 (2.3) (1.5-3.6) 554 (2.8)
e
(1.6-5.1)
e
565 (1.7) (1.0-3.1) .29
65-80 1048 (5.0) (3.3-7.6) 482 (3.8) (2.4-6.0) 566 (6.6)
e
(3.6-11.9)
e
.19
Sex
Male 1245 (3.8) (2.6-5.4) 597 (3.8) (2.5-5.9) 648 (3.8) (2.2-6.3) .96
Female 922 (3.2)
e
(1.7-5.7)
e
439 (2.5)
e
(1.2-5.0)
e
483 (4.0)
e
(1.6-9.5)
e
.46
BMI
<25 688 (5.6) (3.4-9.3) 320 (4.4)
e
(2.4-8.0)
e
368 (7.2)
e
(3.3-14.7)
e
.36
≥25 1400 (2.6) (1.8-3.7) 673 (2.7) (1.7-4.3) 727 (2.5) (1.5-4.2) .84
Usual place for medical care
Yes 1965 (3.9) (2.9-5.3) 934 (3.6) (2.5-5.2) 1031 (4.3) (2.6-6.9) .60
No 202 (0.2)
e
(0.0-1.2)
e
102
e,f
100 (0.4)
e
(0.1-2.6)
ef
Visited PCP in past year
Yes 1726 (4.3) (3.1-5.9) 813 (4.1) (2.9-5.9) 913 (4.5) (2.7-7.4) .78
No 440 (0.6) (0.2-1.8) 223
f
217 (1.4) (0.5-4.1)
f
Insurance type
Uninsured or Medicaid 1230 (4.2) (2.8-6.3) 586 (3.2) (2.0-5.1) 644 (5.5)
e
(3.0-9.9)
e
.20
Medicare, private, or other 937 (2.8) (1.7-4.4) 450 (3.4) (1.9-6.1) 487 (2.0)
e
(1.1-3.6)
e
.20
Race
g
White 1787 (3.5) (2.5-5.0) 833 (3.1) (2.0-4.6) 954 (4.1) (2.4-6.9) .39
Nonwhite 380 (3.5) (2.0-6.2) 203 (4.7)
e
(2.3-9.5)
e
177 (2.1)
e
(1.0-4.6)
e
.18
Education level
<High school or high school
graduate
1216 (3.4) (2.4-4.9) 613 (2.6) (1.6-4.1) 603 (4.6) (2.9-7.3) .08
Some college or college
graduate
946 (3.7) (2.2-6.2) 420 (4.3) (2.5-7.3) 526 (3.0)
e
(1.1-8.3)
e
.51
(continued)
1278 JAMA Oncology September 2017 Volume 3, Number 9 (Reprinted) jamaoncology.com
© 2017 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ on 08/26/2022
reported LDCT in the past year for lung cancer screening was
the primary outcome of the study. Analyses excluded respon-
dents with unknown (n = 6) or self-reported history of lung can-
cer (n = 41) or were missing LDCT testing information (n = 133),
leaving 2167 adults available for analyses. Weighted preva-
lence of LDCT for lung cancer screening in the past year was
calculated by factors of interest. Multivariable prevalence ra-
tios of LDCT in the past year were estimated using predicted
margins. All statistical analyses accounted for complex sam-
pling design and were conducted with SAS callable SUDAAN
statistical software (version 9.0.3, SAS Institute). The study was
based on deidentified publicly available database and ex-
empt from institutional review board and informed consent.
Results | From 2010 to 2015, the percentage of eligible smok-
ers who reported LCDT screening in the past 12 months re-
mained low and constant, from 3.3% in 2010 to 3.9% in 2015
(P = .60); an even lower proportion of noneligible smokers re-
ceived LDCT (Table 1). Of the 6.8 million smokers eligible for
LDCT screening in 2015, only 262 700 received it. Further-
more, there was no significant increase in screening from 2010
to 2015 for any of the sociodemographic groups, nor were there
significant subgroup differences in screening, except be-
tween participants with or without a history of bronchitis
(Table 2). Of note, over 50% (1230/2167) of smokers meeting
USPSTF recommendations for LDCT screening were unin-
sured or Medicaid insured (Table 1).
Discussion | Screening for lung cancer using LDCT among eli-
gible current and former smokers remained low and un-
changed in 2015 following the 2013 USPSTF recommenda-
tion for annual screening. Reasons for exceptionally low uptake
Table 1. Prevalence of LDCT Testing for Lung Cancer in the Past Year Among Screening-Eligible and Noneligible Smokers, National Health Interview
Surveys 2010 and 2015
a,b
(continued)
Characteristic
Total 2010 2015
P Value
c
No. (%) (95% CI) No. (%) (95% CI) No. (%) (95% CI)
Income, $
<35 000 1130 (3.9) (2.8-5.3) 543 (3.9) (2.5-6.1) 587 (3.8) (2.3-6.2) .97
≥35 000 926 (3.3) (2.0-5.4) 446 (2.8) (1.5-5.0) 480 (3.9)
e
(1.8-8.1)
e
.51
Family history of lung cancer
Yes 362 (4.5)
e
(2.4-8.2)
e
161 (4.8)
e
(2.0-10.8)
e
201 (4.1)
e
(2.1-8.0)
e
.76
No 1709 (3.3) (2.3-4.8) 812 (2.8) (1.9-4.4) 897 (3.9) (2.1-6.9) .42
Attempted to quit smoking in the
past 12 months
h
Yes 363 (4.1)
e
(2.1-8.0)
e
164 (3.3)
e
(1.2-8.8)
e
199 (5.1)
e
(2.1-12.3)
e
.52
No 784 (2.3) (1.3-3.9) 381 (2.3)
e
(1.0-5.2)
e
403 (2.2)
e
(1.1-4.3)
e
.93
Ever diagnosed with emphysema
Yes 321 (8.9) (5.8-13.4) 169 (9.6) (5.8-15.5) 152 (7.9)
e
(3.8-15.8)
e
.64
No 1844 (2.6) (1.7-3.9) 866 (2.0) (1.2-3.4) 978 (3.2)
e
(1.7-5.9)
e
.30
Ever diagnosed with bronchitis
Yes 272 (11.2) (6.4-18.8) 135 (11.5) (6.5-19.7) 137 (10.7)
e
(3.6-27.7)
e
.90
No 1895 (2.4) (1.7-3.5) 901 (2.1) (1.3-3.3) 994 (2.9) (1.8-4.6) .30
Ever diagnosed with asthma
Yes 327 (6.2) (3.7-10.1) 184 (8.0) (4.4-14.0) 143 (3.2)
e
(1.3-7.3)
e
.08
No 1838 (3.1) (2.1-4.5) 851 (2.3) (1.5-3.7) 987 (4.0) (2.3-6.7) .16
Noneligible smokers (n = 6632)
i
Total 6632 (2.4) (1.9-2.9) 2632 (2.0) (1.5-2.9) 3989 (2.7) (2.1-3.6) .12
Former, <30 PY, quit ≤15 years
ago
932 (2.3) (1.3-4.1) 378 (3.1) (1.5-6.3) 554 (1.7) (0.7-4.4) .36
Former, ≥30 PY, quit >15 years
ago
740 (4.0) (2.5-6.2) 339 (2.5) (1.1-5.4) 401 (5.8) (2.9-11.3) .17
Former, <30 PY, quit ≥15 years
ago
3334 (1.6) (1.2-2.3) 1255 (1.5) (0.9-2.5) 2079 (1.7) (1.2-2.6) .68
Current, <30 PY 1626 (3.3) (2.3-4.6) 671 (2.0) (1.2-3.5) 955 (4.4) (2.8-6.6) .04
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided
by height in meters squared); LDCT, low-dose computerized tomography; PCP,
primary care physician; PY, pack-years.
a
The following number of respondents were missing data for these items and
are shown in parentheses: income (111), BMI (79), PCP visits (1), education (5),
immigration status (1), family history of lung cancer (96), emphysema (2),
asthma (2). Respondents with missing information were included in the
model, but data are not shown.
b
Percentages are weighted.
c
P value compares 2010 vs 2015.
d
Weighted numbers take into account the assigned sampling weights of
respondents.
e
Unreliable estimates as a result of relative standard errors exceeding 30%.
f
Unable to generate estimate owing to small denominator.
g
White includes non-Hispanic whites, nonwhite includes: Hispanic, Asian,
Black, Native American/Alaskan Native and other race and/or ethnicities.
h
Among current smokers only.
i
Includes former and current smokers who do not meet the US Preventive
Services Task Force Recommendations.
Letters
jamaoncology.com (Reprinted) JAMA Oncology September 2017 Volume 3, Number 9 1279
© 2017 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ on 08/26/2022
of screening may include gaps in smokers’ knowledge regard-
ing LDCT, lack of access to care as well as physicians’ knowl-
edge about screening recommendations
4
and reimburse-
ment. For example, according to a 2015 survey of physicians
in South Carolina, 36% of physicians correctly stated that LCDT
screening should be conducted annually in high-risk individu-
als, and 63% of physicians did not know that Medicare covers
LDCT for lung cancer screening.
4
It is also possible that phy-
sicians may be aware of LDCT screening, but have limited ac-
cess to the high-volume, and high-quality radiology centers,
a recommendation set forth by public health organizations
5
and a stipulation on Medicare reimbursement.
6
The decrease
in the number of screening-eligible smokers from 8.4 million
in 2010 to 6.8 million in 2015 reflects progress in tobacco con-
trol, and this has implications for the future provision of LDCT
screening. Receipt of LDCT and smoking history were self-
reported and subject to recall bias and the limited time fol-
lowing the USPSTF recommendation and Medicare-
reimbursement are limitations of our study. Despite this, our
study provides the first national estimate of LDCT following
the USPSTF recommendation.
In conclusion, annual LCDT screening among heavy
current and former smokers remains low and unchanged
following the USPSTF recommendation despite the poten-
tial to avert thousands of lung cancer deaths each year. This
underscores the need to educate clinicians and smokers
about the benefit and risks of lung cancer screening for
informed decision making.
Ahmedin Jemal, DVM, PhD
Stacey A. Fedewa, MPH, PhD
Author Affiliations: Surveillance & Health Services Research, American Cancer
Society, Atlanta, Georgia.
Corresponding Author: Ahmedin Jemal, DVM, PhD, Surveillance & Health
Services Research, American Cancer Society, 250 Williams St NW, Atlanta, GA
30303-1002 (ajemal@cancer.org).
Published Online: February 2, 2017. doi:10.1001/jamaoncol.2016.6416
Author Contributions: Dr Fedewa had full access to all of the data in the study
and takes responsibility for the integrity of the data and the accuracy of the data
analysis.
Concept and design: All authors.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Fedewa.
Conflict of Interest Disclosures: None reported.
Funding/Support: The American Cancer Society funded the analysis,
interpretation, and presentation of the manuscript.
Role of the Funder/Sponsor: Staff in the Surveillance and Health Services
Research of the American Cancer Society designed and conducted the study,
including analysis, interpretation, and presentation of the manuscript. No staff
at the American Cancer Society, other than the study investigators, reviewed or
approved the manuscript.
1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2017. CA Cancer J Clin.2017;67
(1):7-30.
2. Moyer VA. 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.
3. Doria-Rose VP, White MC, Klabunde CN, et al. Use of lung cancer screening
tests in the United States: results from the 2010 National Health Interview
Survey. Cancer Epidemiol Biomarkers Prev. 2012;21(7):1049-1059.
Table 2. Adjusted Prevalence Ratios and 95% CIs of LDCT Testing for
Lung Cancer in the Past Year Among Screening-Eligible Respondents,
National Health Interview Survey 2010 and 2015 (n = 2167)
a,b
Characteristic PR (95% CI)
Year
2010 1 [Reference]
2015 1.28 (0.66-2.47)
Age, y
55-64 1 [Reference]
65-80 1.34 (0.62-2.88)
Sex
Male 1 [Reference]
Female 0.61 (0.26-1.4)
BMI
<25 1 [Reference]
≥25 0.36 (0.16-0.8)
Usual place for medical care
Yes 1 [Reference]
No 0.12 (0.01-1.78)
Insurance type
Uninsured or medicaid 1 [Reference]
Medicare, private, or other 0.94 (0.43-2.06)
Race
c
White 1 [Reference]
Nonwhite 1.31 (0.51-3.33)
Education level
<High school or high school graduate 1 [Reference]
Some college or college graduate 1.13 (0.49-2.62)
Family history of lung cancer
Yes 1 [Reference]
No 0.84 (0.32-2.21)
Smoking history
Former, ≥30 PY, quit ≤15 years ago 1.27 (0.53-3.05)
Current, ≥30 PY 1 [Reference]
Attempted to quit smoking in the past 12 months
d
Yes 1 [Reference]
No 0.55 (0.17-1.71)
Ever diagnosed with emphysema
Yes 1 [Reference]
No 0.60 (0.19-1.90)
Ever diagnosed with bronchitis
Yes 1 [Reference]
No 0.27 (0.09-0.83)
Ever diagnosed with asthma
Yes 1 [Reference]
No 0.91 (0.35-2.35)
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided
by height in meters squared); LDCT, low-dose computerized tomography; PR,
prevalence ratio; PY, pack-years.
a
192 participants were not included in the model owing to missing data. Visiting
a physician in the past year and income level were not included in the model
owing to instability of estimates.
b
Prevalence Ratios are adjusted for: age, sex, race, smoking history, family of
lung cancer, chronic respiratory conditions, and BMI.
c
White includes non-Hispanic whites, nonwhite includes Hispanic, Asian, Black,
Native American/Alaskan Native and other race/Ethnicities.
d
Among current smokers only.
Letters
1280 JAMA Oncology September 2017 Volume 3, Number 9 (Reprinted) jamaoncology.com
© 2017 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ on 08/26/2022
4. Ersek JL, Eberth JM, McDonnell KK, et al. Knowledge of, attitudes toward,
and use of low-dose computed tomography for lung cancer screening among
family physicians. Cancer. 2016;122(15):2324-2331.
5. Wender R, Fontham ET, Barrera E Jr, et al. American Cancer Society lung
cancer screening guidelines. CA Cancer J Clin. 2013;63(2):107-117.
6. Centers for Medicare and Medicaid Services. Decision Memo for Screening
for Lung Cancer with Low Dose Computed Tomography (LDCT) (CAG-00439N).
2015; https://www.cms.gov/medicare-coverage-database/details/nca-decision
-memo.aspx?NCAId=274. Accessed January 13, 2017.
Association of Interactive Reminders
and Automated Messages
With Persistent Adherence to Colorectal
Cancer Screening: A Randomized Clinical Trial
The US Preventive Services Task Force recommends annual fe-
cal immunochemical test (FIT) as one of the colorectal cancer
(CRC) screening tests.
1
Adherence to yearly FIT is crucial to pro-
grammatic success.
2
However, longitudinal adherence is low
and strategies to improve
persistent adherence are
needed.
3
We evaluated the ef-
fectiveness of interactive telephone calls vs automated short
message service (SMS) on improving adherence to FIT screen-
ing compared with usual care.
Methods | We conducted a prospective randomized parallel
group study, with the setting previously described.
4
The
trial was registered on Clinicaltrials.gov (NCT02815436).
Asymptomatic patients with negative FIT results in their first
screening round from April to September 2015 due for annual
screening in 2016 were eligible. Patients who could not
understand telephone or SMS, or did not have mobile phones
were excluded. Participants were randomized by a computer-
generated sequence with an allocation ratio of 1:1:1. In the
control group, participants were told in 2015 that they should
visit the screening center for annual FIT pickup at the same
calendar month of 2016. In the SMS group, subjects received
a 1-way SMS, highlighting importance of CRC screening, and
notifying date and location of FIT pickup on their mobile. In
the telephone group, participants received a call from a trained
health care physician with the same message as the SMS, but
an interactive conversation was permitted. The interventions
were delivered 1 month before the expected date of participant
return for second round of screening. The Joint Chinese
University of Hong Kong–New Territories East Cluster Clinical
Research Ethics Committee approved the study and participant
consent was waived because the interventions were an
extension of the screening services. The trial protocol is
provided in the Supplement.
Outcomes were rate of FIT pickup within 1 month of a pa-
tient’s anticipated return, and rate of FIT return within 2 months
of anticipated return. Six hundred patients provide 80% power
(at 5% α level) for detecting an 11% increase in FIT return rate in
the intervention groups compared with control, which was as-
sumed to have a FIT return rate of 70%.
5
Associations between
study groups and outcomes were examined by backward step-
wise, binary logistic regression. Subgroup analysis for sex, mari-
tal status, household income, and educational level were per-
formed, because these factors were previously found to be
associated with screening adherence.
6
Figure. Consort Flow Diagram
630 Assessed for eligibility
1 Excluded
No mobile phone
629 Randomized
212 Allocated to SMS messages
intervention
78 Excluded
Did not pick up FIT on time at
follow-up
45 Excluded
Did not pick up FIT on time at
follow-up
21 Excluded
Did not pick up FIT on time at
follow-up
2 Excluded
1 Received colonoscopy in other
sectors before second round of FIT
1 Change in bowel habits before
second round of FIT
3 Excluded
Received colonoscopy in other
sectors before second round of FIT
3 Excluded
Received colonoscopy in other
sectors before second round of FIT
207 Allocated to interactive telephone
messages intervention
210 Allocated to control group
207 Analyzed 209 Analyzed 205 Analyzed
3 Excluded
Already received cancer screening or
not expected to return for follow-up
3 Excluded
Already received cancer screening or
not expected to return for follow-up
2 Excluded
Already received cancer screening or
not expected to return for follow-up
FIT indicates fecal immunochemical
test.
Supplemental content
Letters
jamaoncology.com (Reprinted) JAMA Oncology September 2017 Volume 3, Number 9 1281
© 2017 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ on 08/26/2022