Lung cancer is the leading cause of cancer-related mortality in the United States.1,2 Lung cancer screening with annual low-dose computed tomography (LDCT) is an effective, evidence-based practice that can diagnose lung cancer at earlier stages and improve lung cancer mortality.3,4 Despite guideline support for lung cancer screening in high-risk populations, screening utilization remains low within and outside the Veterans Health Administration (VHA).5–9
Individuals residing in rural areas have higher smoking rates than those residing in non-rural areas. Further, individuals residing in rural areas face barriers to care, such as lower socioeconomic status, inadequate provider-patient communication, and limited geographic access to specialty care, that may contribute to higher lung cancer incidence, mortality, and advanced-stage disease.10–17 Rural norms and culture influence the way health care is delivered and received in rural settings.18,19 Understanding patient satisfaction with lung cancer screening care in rural settings is necessary for developing high-quality lung cancer screening practices.
The primary aim of this study was to explore the association of rurality with Veteran satisfaction with lung cancer screening care at 10 Veterans Affairs Medical Centers (VAMCs). Our secondary aims were to explore the association of other Veteran factors, such as race/ethnicity, educational attainment, and smoking status, with satisfaction with lung cancer screening care and to identify specific areas where care gaps may exist by probing Veteran agreement with individual PSCC survey items that reflect more discrete steps in care.
METHODS
Study Design and Setting
Between June and October 2022, we surveyed a stratified random sample of Veterans who had lung cancer screening exams performed at 10 VAMCs participating in the Veteran Affairs Partnership to increase Access to Lung Cancer Screening (VA-PALS).20 VA-PALS was VHA’s Enterprise-Wide Initiative to nationally implement lung cancer screening programs to increase uptake of lung cancer screening. From 2017–2022, VA-PALS provided participating VAMCs with resources, including funding for lung cancer screening program navigators, educational support, and a software program to support lung cancer screening program needs. VA-PALS provided resources at the system level, while individual program structure and process decisions were made at the individual site level. The 10 VA-PALS sites were Atlanta, Chicago-Hines, Cleveland, Denver, Indianapolis, Milwaukee, Nashville, Philadelphia, Phoenix, and St. Louis.20
The VA Tennessee Valley Health care System Research and Development Committee and VA Central Institutional Review Board (C-IRB E19-05) approved this study. Participants’ implied consent was inferred from completing the paper survey after reading the study information sheet.
Study Population
We identified potential participants (the population of screened Veterans) through the VA Corporate Data Warehouse (CDW) as those who were alive and had an LDCT for lung cancer screening ordered by a lung cancer screening program navigator or other provider designated by the lung cancer screening program at a VA-PALS VAMC between January 1, 2018, and September 30, 2021. Lung cancer screening order placement during this time frame implied eligibility for lung cancer screening by the United States Preventive Services Taskforce (USPSTF) 2013 guidelines. At that time, screening was recommended for individuals aged 55–80 years with at least 30 pack-years of cigarette smoking and who currently smoke or formerly smoked and quit within the past 15 years.5,21
To ensure inclusion of smaller demographic groups that may be at higher risk of lung cancer and to improve the precision of our inference, we used stratified sampling to select Veterans for surveys.22 We prioritized survey distribution based on 4 strata: all combinations of 2-level rural status (rural vs. nonrural) and 2-level race/ethnicity (White non-Hispanic vs. any other race/ethnicity). We oversampled those from “rural” and “other race/ethnicity” backgrounds to ensure we received feedback from a variety of Veteran groups. We selected all Veterans in the “rural, any other race/ethnicity” stratum (oversampling relative to the size of this stratum in the population) and randomly selected Veterans from the remaining strata to receive surveys. The final sample selected to receive surveys consisted of 1997 Veterans.
Survey Distribution
We mailed paper surveys to Veterans between June and October 2022, including a postage-paid return envelope. A single additional mailed reminder was sent to nonresponders. Participants were offered a $20 check for participation. The study team entered responses from the paper survey into Veteran Affairs Research Electronic Data Capture (VA REDCap) for data storage.23,24
Data Sources and Survey Content
Demographic data: We obtained Veteran demographic data from the survey (Supplemental Appendix A, Supplemental Digital Content 2, http://links.lww.com/MLR/D209) and supplemented it with data available in the CDW. The survey elicited participant self-reported demographics (age, educational attainment, and smoking status). Veteran rurality, facility, race/ethnicity, sex, and mailing address were obtained from the CDW. We defined rurality as a binary variable (rural or nonrural) based on the Rural-Urban Commuting Areas, as previously described.25
Smoking Status: Smoking status was defined through the survey question: “Do you CURRENTLY smoke cigarettes every day, some days, or not at all?” which had 4 answer choices: (1) Every Day; (2) Some Days;(3) Not at All; (4) Do not know/Not sure. Given the implied history of either current or former smoking due to lung cancer screening eligibility, we classified smoking status for respondents as a binary variable “currently smoking” versus “formerly smoked.” Those who answered Every Day, Some Days, and Do not Know/Not Sure were classified as “currently smoking.” Those who answered Not at All were classified as “formerly smoked.”
Satisfaction with Lung Cancer Screening Care: We adapted the Patient Satisfaction with Cancer Care (PSCC) items to reflect satisfaction with lung cancer screening care.26 The PSCC is an 18-item scale with responses elicited on a 5-point Likert-like scale to assess satisfaction with cancer-related care (1: strongly disagree–5: strongly agree). The PSCC has been validated in individuals from diverse socioeconomic and cultural backgrounds.26–28 The satisfaction with lung cancer screening care score (PSCC) for each participant was calculated by summing all PSCC item responses (possible range: 18–90) and analyzed as a continuous variable, with higher scores reflecting higher satisfaction with lung cancer screening care. The survey also included 3 open-ended questions for respondents to provide additional information about their lung cancer screening experience.
The datasets generated during and/or analyzed during the current study are not publicly available, but are available from the corresponding author on reasonable request.
Statistical Analysis
We excluded surveys with incomplete PSCC scales from the analysis. Descriptive statistics, including means and SD for continuous variables and counts and frequencies for categorical variables, were used to summarize participant demographics. We analyzed self-reported age as a continuous variable, self-reported education attainment (Less than High School, High School or GED, Some College or Post High School Training, and College or Post College Graduate Degree) as a categorical variable, and smoking status as a binary variable (currently smoking/formerly smoked).
To account for the original sampling scheme and differences in response rates across sampling strata, we used weights calculated by multiplying the stratum-specific sampling weights (rurality and race/ethnicity) by the response rates, assuming that stratum-specific response probabilities did not depend on other characteristics.29 We present descriptive statistics for the responder cohort and in a population-weighted cohort that resembles the population of Veterans we identified as having lung cancer screening ordered through a VA-PALS program (n=8032).
Primary Analysis: The primary outcome was Veteran satisfaction with lung cancer screening care, as measured by the PSCC score. The primary analysis compared the mean PSCC score for each respondent by rurality using t tests in both the responder cohort and the population-weighted cohort.30 The F test compared mean PSCC scores in the responder cohort by sampling strata. P-values <0.05 were considered statistically significant.
Secondary Analyses: We conducted 3 secondary analyses. In the first secondary analysis, we evaluated the association between Veteran predictors (age, sex, race/ethnicity, education attainment, smoking status, and facility location) with the overall PSCC score using an adjusted linear regression model in the population-weighted cohort. Multivariate imputation by chained equations imputed values for missing covariates in the linear regression model.31
In the second secondary analysis, we evaluated the proportion of respondents who agreed with individual PSCC items across the entire responder cohort and, to better understand the experiences of rural Veterans and those who currently smoke, we further stratified the results by rural status and smoking status using the Fisher exact test. To provide simplified and actionable feedback to our clinical partners, satisfaction was summarized by a binary variable defined as “Agree” (which included responses of “agree” and “strongly agree,” 4 and 5 on the Likert-like scale) and “Disagree” (which included “strongly disagree,” “neither agree nor disagree,” “disagree,” 1–3 on the Likert-like scale).
For the third secondary analysis, to account for potential loss of information in creation of the binary satisfaction outcome variable, we performed separate, exploratory adjusted analysis on the individual PSCC item responses in the responder cohort using the original Likert-like scale (1: strongly disagree, 2: disagree, 3: neither agree nor disagree, 4: agree and 5: strongly agree) focusing on the odds ratios for rurality and smoking status. We adjusted for age, race, ethnicity, education attainment, smoking status, and facility and estimated the odds ratios using proportional odds logistic regression and single imputation for missing values. Given the large number of proportional odds regression models and the small proportion of missing values, single imputation reduced the computational burden while still addressing missingness. For continuous variables, we imputed the mean; for categorical variables, we imputed the mode.
In exploratory analyses, we also reviewed participants’ responses to the open-ended question: “What would you change about getting screened for lung cancer?” What did you like least about lung cancer screening?
The primary and secondary analyses were performed using (R Statistical Software version 4.4.0 and version 4.4.2; The R Foundation for Statistical Computing, Vienna, Austria).32
RESULTS
Analytic Sample
We identified 8032 unique Veterans who had an LDCT for lung cancer screening ordered by a lung cancer screening program navigator or other designated provider at one of the 10 VAMCs between January 1, 2018, and September 30, 2021. Of these, 1997 were selected to receive survey mailings. Of those who received mailings, 718 Veterans returned surveys to the study team, for an overall response rate of 36.0%. After exclusion of incomplete surveys (n=29), the final analytic sample consisted of 689 surveys. Participant enrollment is depicted in Figure 1.
Respondent Characteristics
Among survey respondents in our analytic cohort (responder cohort), the median age of Veterans was 68.3 years; 95.4% were male; 34.4% identified as Black, 55.7% as White, and 5.2% as Hispanic; 55.6% reported current smoking; 50.2% were rural. Of the 689 returned surveys, 11 (1.6%) were missing age, 22 (3.2%) were missing education, and 15 (2.2%) were missing smoking status. No clear patterns among the missing data were found. After weighting to create a cohort that resembles the population of screened Veterans at a VA-PALS site (population-weighted cohort), the Veterans’ mean age was 68.6 years, 94.9% were male, 22.5% were Black, 72.4% were White, 3.2% were Hispanic, 57.2% were currently smoking, and 21.3% were rural. Respondent characteristics stratified by rural status in the responder and population-weighted cohorts are presented in Table 1. Respondent characteristics stratified by smoking status in the responder and population-weighted cohorts are presented in Supplemental Table 1, Supplemental Digital Content 1, http://links.lww.com/MLR/D208.
TABLE 1
Demographic Characteristics
Responder cohortPopulation-weighted cohortCharacteristicTotal N=689, n (%), or mean±SDRural N=346, n (%), or mean±SDNonrural N=343, n (%), or mean±SDTotal N=8032, n (%), or mean±SDRural N=1712, n (%), or mean±SDNonrural N=6320, n (%), or mean±SDAge (y)68.3±5.768.3±5.968.4±5.568.6±5.768.3±5.968.6±5.6Missing11 (1.6)7 (2)4 (1.2)Male657 (95.4)331 (95.7)326 (95)7621 (94.9)1643 (96.0)5978 (94.6)Race Black or African American237 (34.4)28 (8.1)209 (60.9)1808 (22.5)77 (4.5)1732 (27.4) Unknown51 (7.4)21 (6.1)30 (8.7)306 (3.8)58 (3.4)249 (3.9) White384 (55.7)290 (83.8)94 (27.4)5816 (72.4)1559 (91.1)4257 (67.4) Other17 (2.5)7 (2.1)10 (2.9)102 (1.2)20 (1.1)82 (1.2)Ethnicity Not Hispanic or Latino620 (90.0)319 (92.2)301 (87.7)7610 (94.7)1638 (95.7)5972 (94.5) Hispanic or Latino36 (5.2)8 (2.3)28 (8.2)254 (3.2)22 (1.3)232 (3.7) Unknown33 (4.8)19 (5.5)14 (4.1)168 (2.1)52 (3)116 (1.8)Education Less than high school48 (7)32 (9.2)16 (4.7)441 (5.5)157 (9.2)284 (4.5) High school or GED222 (32.2)111 (32.1)111 (32.3)2382 (29.7)555 (32.4)1827 (28.9) Some college or post-high school training293 (42.5)139 (40.2)154 (44.9)3621 (45.1)682 (39.8)2939 (46.5) College or post-college graduate degree104 (15.1)56 (16.2)48 (14.0)1183 (14.7)282 (16.5)902 (14.3) Missing22 (3.2)14 (4)8 (2.3)Smoking status Currently smokes383 (55.6)189 (54.6)194 (56.6)4593 (57.2)919 (53.7)3674 (58.1) Formerly smoked291 (42.2)150 (43.4)141 (41.1)3191 (39.7)762 (44.5)2428 (38.4) Missing15 (2.2)8 (2.3)7 (2.0)Facility 1141 (20.5)98 (28.3)43 (12.5)1654 (20.6)492 (28.7)1163 (18.4) 2131 (19.0)49 (14.2)82 (23.9)1471 (18.3)237 (13.9)1234 (19.5) 310 (1.5)2 (0.6)8 (2.3)75 (0.9)8 (0.5)66 (1.0) 462 (9.0)42 (12.1)20 (5.8)706 (8.8)188 (11)519 (8.2) 557 (8.3)46 (13.3)11 (3.2)318 (4)226 (13.2)91 (1.4) 664 (9.3)3 (0.9)61 (17.8)617 (7.7)11 (0.6)606 (9.6) 768 (9.9)47 (13.6)21 (6.1)876 (10.9)249 (14.5)628 (9.9) 811 (1.6)5 (1.4)6 (1.7)125 (1.6)25 (1.5)100 (1.6) 929 (4.2)20 (5.8)9 (2.6)527 (6.6)99 (5.8)427 (6.8) 10116 (16.8)34 (9.8)82 (23.9)1663 (20.7)177 (10.3)1486 (23.5)
Other races included American Indian or Alaska Native, Asian, and Native Hawaiian or other Pacific Islander.
Primary Analysis: Satisfaction With Lung Cancer Screening Care and Rurality
In the responder cohort, the mean PSCC for the overall sample was 72.4 (SD=14.2), and there was no statistically significant difference in PSCC scores by rural status (rural 72.44, nonrural 72.44, P=0.99). In the population-weighted cohort, the mean PSCC score was 71.9 (SD=14.2), and there was no statistically significant difference in PSCC by rural status (rural 72.1, nonrural 71.9, P=0.81) (Fig. 2). Supplemental Table 2, Supplemental Digital Content 1, http://links.lww.com/MLR/D208, contains the mean PSCC score in the responder cohort by the 4 sampling strata. There is no significant difference in mean score among these 4 groups.
Open multimedia modal
FIGURE 2
Open multimedia modal
Effect Estimates of Predictors on Satisfaction with Lung Cancer Screening Care. *The following text should be a note below. *These effect estimates were identified in the covariate-adjusted linear regression analysis in the population-weighted cohort. Predictors are located on the y-axis. The effect estimate on the Patient Satisfaction with Cancer Care (PSCC) score is located on the x-axis. In the adjusted linear regression, age, sex, race, ethnicity, and educational attainment were not strongly associated with satisfaction with lung cancer screening care (P-values=0.28, 0.47, 0.38, 0.38, 0.64, 0.76, and 0.76, respectively; reference: High School or GED). The facility also demonstrated no strong association with satisfaction with lung cancer care (P-value=0.45, 0.41, 0.92, 0.95, 0.95, 0.14, 0.82, 0.62, and 0.87, reference Facility 1). We note a negative association with satisfaction with lung cancer screening care and “currently smoking” status [regression coefficient=−3.20 (−6.42 to 0.02), P-value=0.05].
Secondary Analysis: Predictors of Satisfaction With Lung Cancer Screening Care and Agreement With Individual Patient Satisfaction With Cancer Care Questionnaire Items
The effect estimates from the covariate-adjusted linear regression analysis in the population-weighted cohort are shown in Figure 2. In the adjusted linear regression, age, sex, race/ethnicity, education attainment, and facility were not strongly associated with satisfaction with lung cancer screening care. “Currently smoking” status was negatively associated with lung cancer screening care satisfaction [regression coefficient=−3.20 (−6.42 to 0.02), P=0.05].
Agreement with individual PSCC survey items for the entire responder cohort and by rurality and smoking status is presented in Table 2. The items with the highest proportion agreement for the responder cohort were “I felt that I was treated with courtesy and respect” (93.6% agree), followed by “I felt that my health concerns were understood” (85.6% agree). The item with the lowest proportion agreement was “I knew what my next step in my care would be” (65.2% agree). The item with the second lowest proportion agreement was “I knew who to contact when I had a question” (67.9% agree). Of those who currently smoke, 87.5% agreed with the statement “I felt informed about smoking cessation,” while only 80.1% of those who formerly smoked agreed with this statement (P=0.01). Of those who currently smoke, 80.7% agreed with the statement “I received high-quality care from my health care providers,” while 85.9% of those who formerly smoked agreed with this statement (P=0.079). Since items focusing on communication received the lowest scores, we provide sample quotations from the free-response survey items in Table 3, along with the next steps.
TABLE 2
Individual PSCC Survey Item Responses (Responder Cohort)
Survey ItemOverall agree (%) (n=689)Rural agree (%) (n=346)Nonrural agree (%) (n=343)PCurrently amokes agree (%) (n=383)*Formerly smoked agree (%) (n=291)*PI felt that my health concerns were understood.590 (85.6)288 (83.2)302 (88.0)0.082321 (83.8)258 (88.7)0.075I felt that I was treated with courtesy and respect.645 (93.6)324 (93.6)321 (93.6)1.000354 (92.4)276 (94.8)0.270I felt included in decisions about my health.567 (82.3)281 (81.2)286 (83.4)0.486312 (81.5)244 (83.8)0.474I felt informed about lung cancer screening.553 (80.3)276 (79.8)277 (80.8)0.774299 (78.1)242 (83.2)0.118I felt informed about smoking cessation.580 (84.2)293 (84.7)287 (83.7)0.755335 (87.5)233 (80.1)0.010†I felt encouraged to talk about my health concerns.558 (81.0)277 (80.1)281 (81.9)0.561205 (79.6)240 (82.5)0.375I felt I had enough time with my healthcare providers.531 (77.1)269 (77.7)262 (76.4)0.717288 (75.2)232 (79.7)0.195My questions were answered to my satisfaction.552 (80.1)274 (79.2)278 (81.0)0.568307 (80.2)233 (80.1)1.000Making an appointment was easy.521 (75.6)266 (76.9)255 (74.3)0.478284 (74.2)227 (78.0)0.276I knew what the next step in my care would be.449 (65.2)225 (65)224 (65.3)1.000250 (65.3)190 (65.3)1.000I feel confident in how I deal with the health care system.523 (75.9)269 (77.7)254 (74.1)0.285288 (75.2)224 (77.0)0.649I was able to get the advice I needed about my health issues.525 (76.2)265 (76.6)260 (75.8)0.858287 (74.9)227 (78.0)0.362I knew who to contact when I had a question.468 (67.9)243 (70.2)225 (65.6)0.221257 (67.1)200 (68.7)0.678I received all the services I needed.514 (74.6)258 (74.6)256 (74.6)1.000277 (72.3)227 (78.0)0.107I am satisfied with the care I received.571 (82.9)282 (81.5)289 (84.3)0.363316 (82.5)245 (84.2)0.603The healthcare providers seemed to communicate well about my care.542 (78.7)267 (77.2)275 (80.2)0.353298 (77.8)232 (79.7)0.570I received high-quality care from my healthcare providers.571 (82.9)285 (82.4)286 (83.4)0.762309 (80.7)250 (85.9)0.079My regular health care provider was informed of the test results.492 (71.4)248 (71.7)244 (71.1)0.933272 (71.0)207 (71.1)1.000
Items with the highest agreement: I felt that I was treated with courtesy and respect (93.6%); I felt that my health concerns were understood (85.6%).
Items with the lowest agreement: I knew what the next step in my care would be (65.2%); I knew who to contact when I had a question (67.9%).
*
Fifteen individuals did not report smoking status.
†
statistically significant (P<0.05).
PSCC indicates patient satisfaction with cancer care.
TABLE 3
Sample Veteran Quotations
Veteran responses to questions: “What would you change about getting screened for lung cancer?” and “What did you like least about lung cancer screening?” Communication of results and next stepsMore information. I don’t always get feedback about my CT scans.
Better contact info on when scheduling next scan.
Better explanation on what the CT scan revealed.
The screening went well, but I don’t feel that I was fully informed regarding the results.
Making sure you are contacted with results after the test.
Get the results explained in layman’s terms promptly. That was promised but not done.
Fairly poor job of explaining the results. It was months afterwards before another doctor told me about the pulmonary fibrosis.
More follow-up on scan. It was my first or second scan the radiologist thought there was a couple of non-related problems to cancer on the report. The following year the report…more thoroughly checked those areas and found they were not existent.
If the screening shows an abnormality- I think that the pulmonary doctor should notify you and advise you of his/her plan of action, instead of you having to contact him/her.
Making sure you are contacted with results after the test. Communication of general information about screeningNeed more info about screening.
I would like to have been informed as to the reason for the screening, and why it was necessary.
To be told exactly what would be done.
Reason for testing explained before day of screening.
More information to me in normal language to understand.
Supplemental Table 3, Supplemental Digital Content 1, http://links.lww.com/MLR/D208, contains the results of the exploratory, adjusted analysis of the individual PSCC items by rurality and smoking status. The results suggest similar levels of satisfaction across rurality except for a single item (I felt informed about smoking cessation, OR 1.60 (1.10–1.79), P=0.014), with rural respondents indicating higher levels of satisfaction. However, there is a significant difference in satisfaction by smoking status for 10 individual PSCC items. For each of these items, those currently smoking indicated lower levels of satisfaction with care.
DISCUSSION
This study broadens current understanding of Veterans’ satisfaction with lung cancer screening care within an enterprise-wide lung cancer screening initiative (VA-PALS). Overall, we found that satisfaction with lung cancer screening care was not significantly different between rural and nonrural Veterans. However, Veterans characterized as currently smoking had lower satisfaction with lung cancer screening care than those who formerly smoked. Also, gaps in communication about lung cancer screening care emerged as an area for improvement.
Prior studies found that rural populations have worse lung cancer outcomes than nonrural populations.11,14,16,33 More Veterans eligible for lung cancer screening under USPSTF criteria reside in rural areas than in non-rural areas. Understanding rural Veterans’ satisfaction with lung cancer screening care may help overcome barriers that this population faces.12,22 Our study suggests that once rural Veterans choose to pursue lung cancer screening and complete the screening exam, their overall satisfaction with the experience may be similar to that of nonrural Veterans. This finding can inform future strategies to increase awareness of and access to lung cancer screening.
Of note, we found that Veterans who currently smoke have lower satisfaction with lung cancer screening care than those who formerly smoked. With adjusted analysis, this lower satisfaction with care persisted across multiple individual PSCC items, including those reflective of feeling treated with courtesy and respect, feeling included in decisions about health, feeling encouraged to talk about health concerns, and being satisfied with the care received. This lower satisfaction could result from heightened concern about lung cancer screening, which has been observed in people who currently smoke.34 ∼50% of individuals screened for lung cancer currently smoke.35 A critical component of improving lung cancer screening care and increasing lung cancer screening utilization is understanding the screening experience for the many individuals being screened who currently smoke.
Communication about lung cancer screening processes and results emerged as an area for improvement in lung cancer screening care. We identified low levels of agreement with the statements “I knew what my next step in my care would be” (65.2%) and “I knew who to contact when I had a question” (67.9%). This aligns with previous research findings that clarifying insurance coverage, improving communication, and enhancing administrative coordination are necessary.36–38 A single-site qualitative study exploring patient satisfaction and experiences with lung cancer screening found a lack of communication, particularly regarding the receipt and understanding of screening results and the necessary follow-up.38 Improvement in communication throughout the lung cancer screening process remains an opportunity for continued, focused efforts.
Study strengths include surveying a national sample of participants, oversampling certain demographic groups to enhance representativeness, and conducting analyses in a weighted cohort that resembles the population of Veterans screened at VA-PALS sites in terms of race and rurality. We selected the PSCC, an established survey that has been validated in multiple populations,26–28 to further strengthen the study.26 However, study limitations should be noted. The results may have been influenced by misclassification, nonresponse bias, social desirability bias, and recall bias. Finally, this study may not generalize to populations at other VHA sites or outside VHA.
While Veterans in rural and nonrural areas reported similar satisfaction with lung cancer screening care, Veterans who currently smoke reported lower levels of satisfaction with care than those who formerly smoked. Veterans identified a limited understanding of lung cancer screening and poor communication of next steps in the lung cancer screening process as areas needing improvement. This study provides important insight into Veteran satisfaction with lung cancer screening care and guides future directions to improve lung cancer screening care and communication. Understanding Veteran satisfaction with lung cancer screening care across populations and identifying areas for improvement can inform future health care improvement initiatives.
Keywords:
cancer detection; cancer screening; lung cancer; patient satisfaction; veterans
