Abstract
BACKGROUND
In response to the National Lung Screening Trial, numerous professional organizations published guidelines recommending annual lung cancer screening with low-dose computed tomography (LDCT) for high-risk patients. Prior studies found that physician attitudes and knowledge about lung cancer screening directly impacts the number of screening exams ordered.
METHODS
In 2015, we surveyed 34 pulmonologists and 186 primary care providers (PCPs) to evaluate opinions and practices of lung cancer screening in a large academic medical center. We compared PCP and pulmonologist responses using t-tests and ᵪ2 tests.
RESULTS
The overall survey response rate was 40% (39% for PCPs and 50% for pulmonologists). Pulmonologists were more likely than PCPs to report lung cancer screening as beneficial for patients (88.2% versus 37.7%, P < .0001) and as being cost-effective (47.1% versus 14.3%, P = .02). More pulmonologists (76%) reported ordering a LDCT for screening in the past 12 months compared to PCPs (41%, P = .012). Pulmonologists and PCPs reported similar barriers to referring patients for lung cancer screening, including patient costs (82.4% versus 77.8%), potential for emotional harm (58.8% versus 58.3%), high false positive rate (47.1% versus 69.4%), and likelihood for medical complications (47.1% versus 59.7%).
LIMITATIONS
Our results are generalizable to academic medical centers and responses may be susceptible to recall bias, non-response bias, and social desirability bias.
CONCLUSION
We found significant differences in opinions and practices between PCPs and pulmonologists regarding lung cancer screening referrals and perceived benefits. As lung cancer screening continues to emerge in clinical practice, it is important to understand these differences across provider specialty to ensure screening is implemented and offered to patients appropriately.
In 2017, an estimated 222,500 new cases of lung cancer were diagnosed and more than 155,800 people died from lung cancer [1]. In 2011, the National Lung Screening Trial (NLST) found a 20% reduction in lung cancer mortality when high-risk patients, defined as those aged 55–74 years who were current or former (quit within the past 15 years) smokers with a 30 pack-year smoking history, were screened annually with low-dose computed tomography (LDCT) as compared to chest radiography [2]. This mortality benefit led to a Grade B endorsement for lung cancer screening with LDCT by the US Preventive Services Task Force (USPSTF) in December of 2013 [3]. Recommendations for lung cancer screening vary among professional societies with respect to their criteria for identifying high-risk patients based on age and smoking status, but it is widely accepted to screen patients who meet the NLST criteria. The National Comprehensive Cancer Network (NCCN) [4], the American Lung Association (ALA) [5], the American Cancer Society (ACS) [6], the American Thoracic Society (ATS) [7], the American Society of Clinical Oncologists (ASCO) [8], and the American College of Chest Physicians (ACCP) [7] recommend that individuals at high-risk for lung cancer consider annual screening with LDCT. In contrast, the American Academy of Family Physicians (AAFP) states that there is insufficient evidence to recommend for or against screening for lung cancer with LDCT [9]. Since January 2015, private insurance payers have covered lung cancer screening and in February 2015, the Centers for Medicare and Medicaid Services (CMS) began providing reimbursement for lung cancer screening in high-risk individuals [6, 5, 10]. Under the USPSTF guidelines, an estimated 8.7 million US adults fit within the screening eligible population [11].
Prior research has found that physician attitudes, beliefs, and knowledge about lung cancer screening directly impact the number of reported screening exams ordered [12–14]. Barriers to screening commonly cited by physicians include uncertainty of patient benefits, patient financial burdens, and concern regarding potential harms [15–18]. Although lung cancer screening attitudes, beliefs, and knowledge have been examined among primary care providers (PCPs) and specialists, no studies to date have compared physician opinions and practices by specialty in the same medical center. Evaluating differences by specialty will allow us to identify the extent to which our lung cancer screening program is being implemented equally across the health care system and points for further education and communication among physicians. Therefore, we sought to understand the practices and attitudes of pulmonologists and PCPs in a large academic medical center.
Materials and Methods
Survey Development
Using the Tailored Design Method, we conducted a web-based survey (using Qualtrics survey software) of physicians in Family Medicine, Internal Medicine, and Pulmonology in 2015 [19]. The Tailored Design Method is an approach to designing surveys that emphasizes all aspects of questionnaire and survey implementation procedures and focuses on reducing 4 dominant sources of survey error: sampling, coverage, measurement, and nonresponse. Survey questions were developed through collaboration with a multidisciplinary Advisory Group, which included 7 members from the following fields of study: Internal Medicine, Family Medicine, Thoracic Radiology, Pulmonary Medicine, Pathology, Survey Methodology, and Epidemiology. Over the course of 6 conference calls and follow-up correspondence, we discussed survey themes, developed specific questions, and revised our survey instrument. We included 23 survey items focused on opinions about lung cancer screening, lung cancer screening practices, perceived barriers to lung cancer screening, and physician demographics. Survey questions were comprised of Likert scale items, clinical vignettes, and multiple choice. The survey was pre-tested with 5 North Carolina physicians in primary care or pulmonology who practice outside of the participating academic center. These physicians provided feedback with regard to survey flow, length, design, and ease of understanding and responding to the survey questions. Responses from the pretesting led to modifications to clarify the intent of several questions. The survey was conducted after the USPSTF guideline was issued. During the time of this survey, lung cancer screening had been adopted at our institution by some providers and clinics. Preliminary data indicated that lung cancer screening was fragmented, and few patients were being screened. The survey was designed to help further understand physician’s behaviors and screening patterns.
Physician lung cancer screening opinions were evaluated via a 5-point Likert scale that ranged from strongly agree to strongly disagree. The survey asked physicians to rank their opinion on 6 statements: 1) I am convinced that screening for lung cancer is beneficial for patients; 2) inconsistent recommendations about lung cancer screening make it difficult to decide whether or not to screen; 3) screening for lung cancer is cost-effective; 4) I rely on the recommendations of local specialists regarding lung cancer screening in my practice; 5) I have enough knowledge to explain the pros and cons of lung cancer screening to my patients; and 6) time restrictions during a patient’s clinic visit mean other conditions have higher priority over screening for lung cancer.
Physician lung cancer screening practices were assessed through 5 clinical vignettes of hypothetical patient scenarios to assess knowledge of screening guidelines and to show the complexities involved in making screening decisions. Vignette 1 consisted of a healthy 60-year-old former smoker with a 30 pack-year history who quit smoking 1 year ago. Vignette 2 included a healthy 55-year-old former smoker with a 15 pack-year history who quit smoking 20 years ago. Vignette 3 consisted of a healthy 62-year-old former smoker with a 40 pack-year history who quit smoking 16 years ago and was exposed to asbestos in the workplace. For each vignette, physicians were asked if they would or would not recommend lung cancer screening with LDCT, or if they would request additional information. Since differing organizations have varying recommendations for lung cancer screening (in terms of age at which to screen, smoking history, and other factors), there is no 1 correct answer for each vignette. For vignette 1, all organizations except the AAFP recommend screening. In vignette 2, the fact that the patient quit smoking 20 years ago indicates that they do not meet USPSTF, ALA, ACS, ATS, ASCO, or ACCP guidelines for screening, although they do meet the NCCN criteria. For vignette 3, the patient is outside the 15 year-since-quit window so does not meet most organizations criteria for screening, but the patient has exposure to asbestos so may be suitable for screening if it were risk-based.
Additional survey questions included whether the physician had ordered a LDCT for lung cancer screening in the past 12 months, if they had initiated discussion regarding the risks and benefits of screening, if they had discussed the results of an LDCT screening exam with a patient, or if they had referred a patient for lung cancer screening. These questions were based on those from the 2006 National Cancer Institute Colorectal and Lung Cancer Screening Questionnaire [20]. Questions also included whether or not a physician had received requests from patients for lung cancer screening in the past 12 months, and if so, how many requested screening. Lastly, the survey included a question regarding patient referrals for a smoking cessation program.
In order to determine awareness of guidelines concerning lung cancer screening, physicians were given a list of professional organizations that have published statements concerning lung cancer screening and were asked if these organizations recommend for or against lung cancer screening.
To ascertain physician perceived barriers to lung cancer screening, we provided a list related to test performance issues, patient related issues, health care access issues, and medical complications. Physicians were asked to select all perceived barriers. Lastly, physician demographics included age, race, ethnicity, gender, medical division, the number of years in clinical practice, the percent of effort dedicated to patient care, and the number of patients typically seen in their clinic.
Survey Deployment
We identified physicians in Family Medicine, Internal Medicine, and Pulmonary Medicine groups through our academic online directories. We mailed each participant a pre-notification postcard to introduce the study and survey. One week later, a survey link was emailed to each participant. At 1, 4, 8, and 9 weeks post survey delivery, reminder emails were sent to those who had not yet responded. A reminder postcard was mailed 2 weeks after survey deployment. The survey was conducted online with responses recorded directly in Qualtrics at the time of survey completion, then exported for analysis in SAS v9.4 (SAS Institute, Inc., Cary, NC). Consent for the study was determined by return of the survey. As an incentive, those that participated were given the opportunity to enter into a random drawing for an iPad. We limited our survey to 23 items and included an incentive, as prior work has shown this to increase physician response rates [21–23]. This study was reviewed and approved by the University of North Carolina at Chapel Hill Institutional Review Board.
Statistical Analysis
We combined Family Medicine and Internal Medicine physicians into a PCP group, as the majority of patients eligible for screening will be identified from these 2 groups and both groups provide primary care. We compared PCP responses with physicians in Pulmonary Medicine using t-tests for continuous outcomes and ᵪ2 tests for categorical outcomes.
Results
The overall survey response rate was 40% and by physician group was 50.0% for pulmonologists (17 of 34) and 38.7% for PCPs (72 of 186). Within PCP, 37 were from Family Medicine and 35 were from Internal Medicine. Approximately 58.3% of PCPs were attending physicians and 41.7% were resident physicians, whereas 64.7% of pulmonologists were attending physicians and 35.3% were resident physicians. The mean age of pulmonologists and PCPs was similar (see Table 1). Compared with PCPs, pulmonologists spent less time in outpatient care (mean time 45% versus 29% respectively) and saw fewer patients per week. Physicians in Family Medicine and Internal Medicine were similar with respect to age, sex, race, years in clinical practice, percent of time spent in outpatient care, and the average number of patients seen per week.
Table 1.
Characteristics of Survey Respondents by Provider Specialty
| Pulmonologists N = 17 |
Primary care providers N = 72 |
|||
|---|---|---|---|---|
| N or Mean |
% (Range) |
N or Mean |
% (Range) |
|
| Age, mean | 40.8 | (29–67) | 39.7 | (28–64) |
| # Missing | 2 | 9 | ||
| Sex | ||||
| Female | 5 | 31.3 | 34 | 53.1 |
| Male | 11 | 68.8 | 30 | 46.9 |
| Missing | 1 | -- | 8 | -- |
| Race | ||||
| White | 12 | 75.0 | 57 | 89.1 |
| Non-white | 4 | 25.0 | 7 | 10.9 |
| Missing | 1 | -- | 8 | -- |
| Years in clinical practice, mean | 13.4 | (2–35) | 11.0 | (1–35) |
| # Missing | 1 | 8 | ||
| % Time in outpatient care | ||||
| Mean | 29.4 | (10–70) | 44.5 | (5–100) |
| Median | 25.0 | 40.0 | ||
| # Missing | 1 | 7 | ||
| Average # patients seen in outpatient/week | 16.7 | (6–50) | 28.0 | (3–100) |
| # Missing | 1 | 7 | ||
The majority of respondents agreed that they had enough knowledge to explain the pros and cons of lung cancer screening to their patients. Most pulmonologists agreed that lung cancer screening is beneficial for patients, while most PCPs disagreed or were undecided (P < .001) (see Table 2). Pulmonologists were more likely to report screening as cost-effective (P = .02) and that they rely on recommendations of local specialists when making lung cancer screening decisions compared with PCPs (P < .001). It is important to note that within pulmonology there are screening specialists. PCPs were significantly more likely than pulmonologists to report that time restrictions during a patient’s visit led to other presenting problems having higher priority than lung cancer screening (P = .012). Within the PCP group, Family Medicine and Internal Medicine physicians had similar answers except for the question “I have enough knowledge to explain the pros and cons of lung cancer screening to my patient,” for which Family Medicine physicians were more likely to report being undecided than Internal Medicine (31% versus 3%).
Table 2.
Lung Cancer Screening Opinions by Provider Specialty
| Statement | Pulmonologists N = 17 |
Primary Care Providers N = 72 |
P value |
||
|---|---|---|---|---|---|
| N | % | N | % | ||
| I am convinced that screening for lung cancer is beneficial for patients | |||||
| Strongly agree | 6 | 35.3 | 4 | 5.8 | < .001 |
| Agree | 9 | 52.9 | 22 | 31.9 | |
| Undecided | 2 | 11.8 | 33 | 47.8 | |
| Disagree | 0 | 0 | 10 | 14.5 | |
| Strongly disagree | 0 | 0 | 0 | 0 | |
| Missing | 0 | – | 3 | – | |
| Inconsistent recommendations about lung cancer screening make it difficult to decide whether or not to screen | |||||
| Strongly agree | 2 | 11.8 | 8 | 11.4 | .575 |
| Agree | 5 | 29.4 | 34 | 48.6 | |
| Undecided | 2 | 11.8 | 7 | 10.0 | |
| Disagree | 7 | 41.2 | 18 | 25.7 | |
| Strongly disagree | 1 | 5.9 | 3 | 4.3 | |
| Missing | 0 | – | 2 | – | |
| Screening for lung cancer is cost-effective | |||||
| Strongly agree | 0 | 0 | 1 | 1.4 | .026 |
| Agree | 8 | 47.1 | 9 | 12.9 | |
| Undecided | 8 | 47.1 | 41 | 58.6 | |
| Disagree | 1 | 5.9 | 11 | 15.7 | |
| Strongly disagree | 0 | 0 | 8 | 11.4 | |
| Missing | 0 | – | 2 | – | |
| I rely on the recommendations of local specialists regarding lung cancer screening in my practice | |||||
| Strongly agree | 3 | 19.7 | 0 | 0 | < .001 |
| Agree | 9 | 52.9 | 16 | 23.2 | |
| Undecided | 0 | 0 | 12 | 17.4 | |
| Disagree | 5 | 29.4 | 35 | 50.7 | |
| Strongly disagree | 0 | 0 | 6 | 8.7 | |
| Missing | 0 | – | 3 | – | |
| I have enough knowledge to explain the pros and cons of lung cancer screening to my patients | |||||
| Strongly agree | 4 | 25.0 | 9 | 13.0 | .221 |
| Agree | 9 | 56.3 | 33 | 47.8 | |
| Undecided | 3 | 18.8 | 12 | 17.4 | |
| Disagree | 0 | 0 | 14 | 20.3 | |
| Strongly disagree | 0 | 0 | 1 | 1.5 | |
| Missing | 1 | – | 3 | – | |
| Time restrictions during a patient’s clinic visit mean other presenting problems have higher priority than screening for lung cancer | |||||
| Strongly agree | 2 | 11.8 | 10 | 14.7 | .017 |
| Agree | 5 | 29.4 | 36 | 52.9 | |
| Undecided | 7 | 41.2 | 8 | 11.8 | |
| Disagree | 2 | 11.8 | 14 | 20.6 | |
| Strongly disagree | 1 | 5.9 | 0 | 0 | |
| Missing | 0 | – | 4 | – | |
The majority of pulmonologists and PCPs correctly identified that the USPSTF and the ACS recommend lung cancer screening (see Table 3). However, a higher proportion of pulmonologists knew that the ACCP recommends screening compared with PCPs (76.5% versus 38.5%, respectively). Very few pulmonologists or PCPs knew that the AAFP does not endorse lung cancer screening. Among PCPs, Family Medicine and Internal Medicine physicians had similar responses except for the AAFP organization, for which, Family Medicine physicians were more likely to report yes than Internal Medicine (34% versus 9%), more likely to report no (25% versus 6%), and less likely to report not sure (41% versus 84%).
Table 3.
Professional Organization Lung Cancer Screening Guideline Knowledge by Provider Specialty
| Professional organization |
Pulmonologists’ response |
Primary care providers’ response |
P value |
||||
|---|---|---|---|---|---|---|---|
| Recommend screening? | Recommend screening? | ||||||
| Yes N (%) |
No N (%) |
Not Sure N (%) |
Yes N (%) |
No N (%) |
Not Sure N (%) |
||
| US Preventive Services Task Force |
13 (76.5)* |
0 (0) |
4 (23.5) |
57 (87.7)* |
4 (6.2) |
4 (6.2) |
.079 |
| American Cancer Society |
10 (58.8)* |
0 (0) |
7 (41.2) |
35 (53.9)* |
2 (3.1) |
28 (43.1) |
1.00 |
| American College of Chest Physicians |
13 (76.5)* |
0 (0) |
4 (23.5) |
25 (38.5)* |
1 (1.5) |
39 (60.0) |
.0146 |
| American Academy of Family Physicians |
2 (11.8) |
1 (5.9)* |
14 (82.4) |
14 (21.9) |
10 (15.6)* |
40 (62.5) |
.387 |
Note. Indicates correct choice
In the clinical vignettes (see Table 4), pulmonologists and PCPs reported similar responses. Both recommended LDCT most often for vignette 1 (81% of PCPs and 100% of pulmonologists), the scenario that met recommended screening guideline criteria for all organizations except the AAFP. For vignette 2, the majority of both groups recommended no screening in accordance with the USPSTF, ACS, ALA, ATS, ASCO, and ACCP guidelines. In vignette 3, providers were split, with 47.1% of pulmonologists and 51.5% of PCPs recommending LDCT. In this vignette, a similar proportion said they would not screen and slightly more pulmonologists said they needed more information than PCPs (35% versus 25% respectively). The responses for vignette 3 indicate the complexities involved in making decisions around lung cancer screening. Family Medicine and Internal Medicine reported similar responses for all clinical vignettes.
Table 4.
Recommended Lung Cancer Screening Strategies for 3 Clinical Vignettes
| Pulmonologists | Primary care providers | |||||||
|---|---|---|---|---|---|---|---|---|
| Recommended screening strategy | Recommended screening strategy | |||||||
| Vignette | Description | No Screening N (%) |
LDCT N (%) |
Need additional information N (%) |
No screening N (%) |
LDCT N (%) |
Need additional information N (%) |
P value |
| 1 | Healthy 60-year-old, former smoker that has a 30 pack-year history and quit smoking 1 year ago | 0 (0) |
17 (100) |
0 (0) |
6 (8.8) |
55 (80.9) |
7 (10.3) |
.218 |
| 2 | Healthy 55-year-old, former smoker that has a 15 pack-year history and quit smoking 20 years ago | 12 (70.6) |
1 (5.9) |
4 (23.5) |
59 (86.8) |
5 (7.4) |
4 (5.9) |
.085 |
| 3 | Healthy 62-year-old, former smoker that has a 40 pack-year history, quit smoking 16 years ago, and was exposed to asbestos in the workplace | 3 (17.7) |
8 (47.1) |
6 (35.3) |
16 (23.5) |
35 (51.5) |
17 (25.0) |
.724 |
Note. LDCT, low-dose computed tomography.
During the 12 months preceding the survey, 76.5% of pulmonologists reported ordering a LDCT for lung cancer screening, compared to 41.2% of PCPs (P = .002) (see Table 5). A similar proportion of both pulmonologists (70.6%) and PCPs (41.8%) who ordered a LDCT reported discussing results of the scan with their patient. Less than 15% of physicians reported referring a patient to another provider for evaluation of lung cancer screening. Over 90% of pulmonologists and PCPs reported referring patients to various smoking cessation programs during the prior 12 months. Within PCPs, Internal Medicine physicians were more likely to report yes than Family Medicine (58% versus 42%) for the question “In last 12 months, did you discuss results of LDCT for lung cancer screening?”.
Table 5.
Lung Cancer Screening Practices by Pulmonologists and Primary Care Providers
| Statement | Pulmonologists N = 17 |
Primary care providers N = 72 |
P value | ||
|---|---|---|---|---|---|
| N | % | N | % | ||
| In last 12 months, did any of your patients ask if they should be screened for lung cancer with LDCT? | 8 | 50.0 | 18 | 27.3 | .126 |
| Yes | 7 | 43.8 | 44 | 66.7 | |
| No | 1 | 6.3 | 4 | 6.1 | |
| Don’t recall | 1 | – | 6 | – | |
| Missing | |||||
| In last 12 months, did you order LDCT for LCS? | 13 | 76.5 | 28 | 41.2 | .002 |
| Yes | 3 | 17.6 | 40 | 58.8 | |
| No | 1 | 5.9 | 0 | 0 | |
| Don’t recall | 0 | – | 4 | – | |
| Missing | |||||
| In last 12 months, did you initiate LCS discussion re benefits/harms? | |||||
| Yes | 14 | 82.4 | 42 | 61.8 | .076 |
| No | 2 | 11.8 | 25 | 36.8 | |
| Don’t recall | 1 | 5.9 | 1 | 1.5 | |
| Missing | 0 | – | 4 | – | |
| In last 12 months, did you discuss results of LDCT for lung cancer screening? | |||||
| Yes | 12 | 70.6 | 28 | 41.8 | .099 |
| No | 5 | 29.4 | 35 | 52.2 | |
| Don’t recall | 0 | 0 | 4 | 6.0 | |
| Missing | 0 | – | 5 | – | |
| In last 12 months, did you refer patient to another provider for evaluation of lung cancer screening? | 1 | 5.9 | 10 | 14.7 | .448 |
| Yes | 16 | 94.1 | 57 | 83.8 | |
| No | 0 | 0 | 1 | 1.5 | |
| Don’t recall | 0 | – | 4 | – | |
| Missing | |||||
| Have you referred patients to any of the following smoking cessation programs? | |||||
| QuitlineNC | 13 | 76.5 | 55 | 76.4 | 1.000 |
| 1-800-Quit NOW | 13 | 76.5 | 52 | 72.2 | 1.000 |
| Local Nicotine Dependence Program | 13 | 76.5 | 45 | 62.5 | .398 |
| ACS Freedom from Smoking | 1 | 5.9 | 0 | 0 | .191 |
| Alternative Provider | 3 | 17.6 | 14 | 19.4 | 1.000 |
| Any Program | 16 | 94.1 | 66 | 91.7 | 1.000 |
Note. LCS, lung cancer screening; LDCT, low-dose computed tomography.
Pulmonologists and PCPs reported similar barriers for recommending or referring patients for lung cancer screening. The most commonly cited barrier was cost to patients (82.4% of pulmonologists and 77.8% of PCPs). Other barriers to screening included potential for emotional harm (58.8% pulmonologists and 58.3% PCPs), potential for complications (47.1% pulmonologists and 59.7% PCPs), too many false positives (47.1% pulmonologists and 69.4% PCPs), cost to the health care system (47.1% pulmonologists and 54.2% PCPs), and low patient acceptance (41.2% pulmonologists and 37.5% PCPs).
Discussion
Based on our survey, significant differences existed in opinions and practices of pulmonologists and PCPs regarding lung cancer screening. Pulmonologists were more likely than PCPs to report positive aspects of lung cancer screening, such as the benefit to patients and cost-effectiveness. A higher proportion of pulmonologists reported ordering a LDCT for lung cancer screening for an asymptomatic patient during the 12 months before the survey compared to PCPs. This may be due to the higher prevalence of chronic heavy smokers in pulmonology clinics than in PCP practices. It is also possible that PCPs may be referring patients to specialists for further evaluation and the specialists may be ordering the LDCT exam. In our study, factors likely to contribute to LDCT ordering rates were similar between groups, such as having sufficient knowledge to initiate conversation and explain the risks and benefits of screening. One important difference that may explain some of the variation in ordering lung cancer screening is that PCPs more often reported time limitations during the patients visit compared with pulmonologists. In addition, patient visits with pulmonologists are focused on lung and breathing related health, allowing for more time to consider lung screening as part of overall lung health.
Several prior studies surveyed family physicians and other PCPs regarding their attitudes and knowledge of lung cancer screening. A 2015 survey of family physicians in South Carolina identified similar barriers and perceived risks of screening as in our study [16]. Specifically, concerns about patient stress or anxiety was a perceived risk among 52% of family physicians, compared to the 58% of PCPs in our survey who reported emotional harm to patients as a barrier. In 2 studies conducted among family physicians and PCPs, 83%–88% expressed concern over the potential for unnecessary diagnostic procedures and false positive findings [12, 16], which is similar to PCPs in our study indicating concern over false positives (69%) and potential for complications (60%) [16].
We found that pulmonologists and PCPs identified similar barriers to screening. Cost to the patient was the most often cited concern, which is similar to the 2013 data from the Lewis study (87%) [12]. Although CMS began covering the cost of lung cancer screening in 2015, it is possible that at the time of our survey, physicians may not have been aware of this coverage or they may have been concerned that additional follow-up would increase out-of-pocket expenses. While insurance may cover the cost of the initial LDCT screening test, follow-up diagnostic testing is likely subject to deductibles and further out-of-pocket expenses. Providers and patients are both increasingly aware of these expenses and some clinicians may be concerned about the financial strain of diagnostic tests [24]. Policy interventions aimed at reducing uncertainty regarding out-of-pocket expenses may help patients and providers make more informed decisions about screening.
In our study, 38% of PCPs agreed or strongly agreed that screening for lung cancer is beneficial for patients, which was similar to findings of prior studies examining the proportion of physicians who believe that screening reduces lung cancer mortality (41% and 42%, respectively) [12, 16]. Approximately 62% of PCPs in our study had initiated a discussion regarding the benefits and harms of screening, which is similar to the 72% of PCPs in the Los Angeles County study by Raz and colleagues [16]. In our study, 41% of PCPs reported ordering a LDCT for lung cancer screening in the prior 12 months, which is similar to other studies that reported 52%–53% [15, 16].
Prior work examining pulmonologists’ attitudes toward lung cancer screening include a national survey to pulmonologists in Veterans Health Administration clinics from July 2013 to February 2014 and an international sample of ATS clinicians from March to April of 2014 [13, 18]. Among pulmonologists in Veterans clinics, the most commonly cited barriers to screening were poor infrastructure and lack of clinical personnel. Respondents of the ATS survey who said they would screen an NLST-eligible patient reported the following barriers to screening: false positives (52%), over-diagnosis (44%), and high cost to the health care system (33%) [13].
The AAFP does not currently recommend lung cancer screening, stating that the evidence is insufficient to recommend for or against screening [9]. The AAFP has expressed concerns that lung cancer screening recommendations are based on a single study and that NLST results have not been replicated in community settings. While initial findings from the Lahey Hospital & Medical Center are similar to the cancer detection and incidental finding rates of the NLST [25], there remains a paucity of data on LDCT screening in real-world settings. We found that most PCPs and pulmonologists were unsure of AAFP recommendations around lung cancer screening. While it is not surprising that pulmonologists lack this knowledge, less than 20% of PCPs knew the AAFP stance on lung cancer screening, which suggests that most PCPs are more aware of guidelines from the USPSTF.
In our sample of providers, both pulmonologists and PCPs reported spending less than 50% of their time in outpatient care (mean time 29% and 45%, respectively). The relatively low proportion of time spent in outpatient care is likely reflective of: 1) the inclusion of resident physicians who spend a significant amount of their time on inpatient work; 2) academic pulmonologists often spending more time in the inpatient setting, covering the intensive care unit and the pulmonary inpatient service; and 3) academic physicians having teaching and research responsibilities that limit their practice time, as compared to community physicians.
Limitations
Our study has several limitations. Surveyed physicians were part of a large academic medical center, and our results may not be generalizable to other practice settings. However, to our knowledge, our study is the first to survey both PCPs and pulmonologists in the same institution, allowing for a direct comparison. The responses are self-reported and some of the survey questions asked practitioners about events from the past 12 months, so it is possible that responses may be susceptible to recall bias. While our survey response rate of 40% was not ideal, it is in line with prior web-based surveys to physicians regarding lung cancer screening, which ranged from 4.8%–60% [12, 15, 16, 18]. In addition, there is the potential for non-response bias given our overall response rate. Future studies should collect information on non-respondents or include auxiliary data on the sampling frame to address nonresponse bias. It is also possible that survey responses could be subject to social desirability bias.
Conclusion
As lung cancer screening is now a covered preventive service in the United States, the role of the PCP will be critical for screening in the setting of patient care. Our survey showed that the majority of PCPs are uncertain about referring patients for lung cancer screening, citing concerns regarding potential harms and cost. Understanding the opinions, beliefs, knowledge, and practice patterns of PCPs and pulmonologists may result in improved educational programs and development of comprehensive infrastructures that strengthen support and collaboration among the various physicians who are charged with providing this preventive service.
Acknowledgments
Financial support. This work was supported by the National Cancer Institute at the National Institutes of Health (grant number R21CA175983). The views expressed in this article are the authors and do not necessarily represent those of the National Institutes of Health. The funding source did not provide any input or contributions to the development of the research and manuscript beyond funding.
Footnotes
Potential conflicts of interest. All authors have no relevant conflicts of interest.
Contributor Information
Louise M. Henderson, Department of Radiology; Department of Epidemiology; Member, Lineberger Comprehensive Cancer Center, The University of North Carolina, Chapel Hill, North Carolina..
Mary W. Marsh, Department of Radiology, The University of North Carolina, Chapel Hill, North Carolina..
Thad S. Benefield, Department of Radiology, The University of North Carolina, Chapel Hill, North Carolina..
Laura M. Jones, Department of Radiology, The University of North Carolina, Chapel Hill, North Carolina..
Daniel S. Reuland, Department of Medicine, Division of General Medicine & Clinical Epidemiology; Member, Lineberger Comprehensive Cancer Center, The University of North Carolina, Chapel Hill, North Carolina..
Alison T. Brenner, Department of Medicine, Division of General Medicine & Clinical Epidemiology; Member, Lineberger Comprehensive Cancer Center, The University of North Carolina, Chapel Hill, North Carolina..
Adam O. Goldstein, Department of Family Medicine, The University of North Carolina, Chapel Hill, North Carolina..
Paul L. Molina, Department of Radiology, The University of North Carolina, Chapel Hill, North Carolina..
Susan J. Maygarden, Department of Pathology and Laboratory Medicine, The University of North Carolina, Chapel Hill, North Carolina..
M. Patricia Rivera, Department of Medicine, Division of Pulmonary Disease and Critical Care Medicine; Member, Lineberger Comprehensive Cancer Center, The University of North Carolina, Chapel Hill, North Carolina..
References
- 1.National Cancer Institute. CANCER Stat Facts: Lung and Bronchus Cancer. Surveillance, Epidemiology, and End Results Program website. http://seer.cancer.gov/statfacts/html/lungb.html. Accessed January 26, 2018. [Google Scholar]
- 2.Aberle DR, Adams AM, Berg CD, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011;365(5):395–409. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.US Preventive Services Task Forage. Final recommendation Statement. Lung cancer: Screening. USPSTF website. https://www.uspreventiveservicestaskforce.org/Page/Document/RecommendationStatementFinal/lung-cancer-screening. Accessed January 26, 2018.
- 4.Wood DE. National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines for Lung Cancer Screening. Thorac Surg Clin. 2015;25(2):185–197. [DOI] [PubMed] [Google Scholar]
- 5.Providing Guidance on Lung Cancer Screening to Patients and Physicians: An Update from the American Lung Association Lung Cancer Screening Committee. American Lung Association; 2015. http://www.lung.org/assets/documents/lung-cancer/lung-cancer-screening-report.pdf. Accessed January 26, 2018. [Google Scholar]
- 6.Wender R, Fontham ET, Barrera E Jr, et al. American Cancer Society lung cancer screening guidelines. CA Can J Clin. 2013;63(2):107–117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wiener RS, Gould MK, Arenberg DA, et al. An official American Thoracic Society/American College of Chest Physicians policy statement: implementation of low-dose computed tomography lung cancer screening programs in clinical practice. Am J Respir Crit Care Med. 2015;192(7):881–891. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bach P, Mirkin JN, Oliver TK, et al. Lung Screening: The Role of CT Screening for Lung Cancer in Clinical Practice. The Evidence Based Practice Guideline of the American College of Chest Physicians and the American Society for Clinical Oncology. https://pilotguidelines.atlassian.net/wiki/spaces/LS/overview. Accessed February 6, 2018. [Google Scholar]
- 9.American Academy of Family Physicians. Summary of Recommendations for Clinical Preventive Services. Leawood KS: American Academy of Family Physicians; 2017. https://www.aafp.org/dam/AAFP/documents/patient_care/clinical_recommendations/cps-recommendations.pdf. Accessed January 26, 2018. [Google Scholar]
- 10.Jaklitsch MT, Jacobson FL, Austin JH, et al. 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] [PubMed] [Google Scholar]
- 11.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. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lewis JA, Petty WJ, Tooze JA, et al. Low-dose CT lung cancer screening practices and attitudes among primary care providers at an academic medical center. Cancer Epidemiol Biomarkers Prev. 2015;24(4):664–670. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Simmons J, Gould MK, Woloshin S, Schwartz LM, Wiener RS. Attitudes about low-dose computed tomography screening for lung cancer: a survey of American Thoracic Society Clinicians. Am J Respir Crit Care Med. 2015;191(4):483–486. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Klabunde CN, Marcus PM, Han PK, et al. Lung cancer screening practices of primary care physicians: results from a national survey. Ann Fam Med. 2012;10(2):102–110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Raz DJ, Wu GX, Consunji M, et al. Perceptions and utilization of lung cancer screening among primary care physicians. J Thorac Oncol. 2016;11(11):1856–1862. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.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. [DOI] [PubMed] [Google Scholar]
- 17.Hoffman RM, Sussman AL, Getrich CM, et al. Attitudes and beliefs of primary care providers in New Mexico about lung cancer screening using low-dose computed tomography. Prev Chronic Dis. doi: 10.5888/pcd12.150112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Iaccarino JM, Clark J, Bolton R, et al. A national survey of pulmonologists’ views on low-dose computed tomography screening for lung cancer. Ann Am Thorac Soc. 2015;12(11):1667–1675. [DOI] [PubMed] [Google Scholar]
- 19.Dillman DA, Smyth JD, Christian LM. Internet, Phone, Mail, and Mixed-Mode Surveys: The Tailored Design Method, 4th Edition 4th ed Hoboken NJ: Wiley; 2014. [Google Scholar]
- 20.National Cancer Institute. National Survey of Primary Care Physicians’ Cancer Screening Recommendations and Practices; Colorectal and Lung Cancer Screening Questionnaire. Rockville MD: Westat; 2006. http://healthservices.cancer.gov/surveys/screening_rp/screening_rp_colo_lung_inst.pdf. Accessed January 26, 2018. [Google Scholar]
- 21.VanGeest JB, Johnson TP, Welch VL. Methodologies for improving response rates in surveys of physicians: a systematic review. Eval Health Prof. 2007;30(4):303–321. [DOI] [PubMed] [Google Scholar]
- 22.Klabunde CN, Willis GB, McLeod CC, et al. Improving the quality of surveys of physicians and medical groups: a research agenda. Eval Health Prof. 2012;35(4):477–506. [DOI] [PubMed] [Google Scholar]
- 23.Thorpe C, Ryan B, McLean SL, et al. How to obtain excellent response rates when surveying physicians. Fam Pract. 2009;26(1):65–68. [DOI] [PubMed] [Google Scholar]
- 24.Harris RP, Sheridan SL, Lewis CL, et al. The harms of screening: a proposed taxonomy and application to lung cancer screening. JAMA Intern Med. 2014;174(2):281–285. [DOI] [PubMed] [Google Scholar]
- 25.McKee BJ, McKee AB, Flacke S, Lamb CR, Hesketh PJ, Wald C. Initial experience with a free, high-volume, low-dose CT lung cancer screening program. J Am Col Radiol. 2013;10(8):586–592. [DOI] [PubMed] [Google Scholar]