Abstract
Primary care physicians (PCPs) may be involved in the evaluation of solitary pulmonary nodules (SPNs) detected through lung cancer screening. Little is known about their perspectives on the management or the referral of SPN. Using the American Medical Association’s Physician Masterfile, we randomly surveyed 1384 PCPs between January and October 2015 with an 18% response rate. A subset analysis was performed on SPN management and referral practices of PCP These results and those relating to practice characteristics were compared between family practice and internal medicine physicians. Responders and nonresponders did not differ by demographic characteristics. A total of 137 (55.5%) PCPs reported feeling confident in managing the workup of imaging-detected SPN. However, only 53 PCPs (21.3%) were inclined to manage the evaluation and follow-up of SPN. There was no significant difference between family practice and internal medicine physicians with regard to years in practice, size of practice, or referral to specialists. Family practitioners and internists similarly disagreed or were neutral to self-managing SPN (P = 0.60). Internists were twice as likely to express confidence as family practitioners (odds ratio 1.95, 95% confidence interval 1.09-3.48). Among all PCPs, 75.4% would refer management of these patients to a pulmonologist, 28.9% to a surgeon, and 24.2% to an oncologist. Confidence did not predict lung cancer screening practices. Although more than half of PCPs expressed confidence in the workup of SPN, most preferred referral to specialists. Additional research is needed to understand barriers to PCP management of incidental SPN in the effort to facilitate lung cancer screening.
Keywords: solitary pulmonary nodules, primary care physicians
Graphical Abstract
Primary care physicians' confidence and preferences in the management of pulmonary nodules.
INTRODUCTION
Solitary pulmonary nodules (SPNs) are commonly detected incidental findings in patients undergoing computed tomography for other reasons, and more recently as a result of lung cancer screening. Lung cancer screening using low-dose computed tomography (LDCT) is effective at reducing lung cancer-related mortality1 and is recommended by major health organizations including the United States Preventive Services Task Force,2 the National Comprehensive Cancer Network,3 the American Cancer Society,4 the American Lung Association,5 and the American Association for Thoracic Surgery.6 LDCT is approved by the Centers for Medicare Services for Medicare beneficiaries,7 and is a mandated benefit for plans under the Affordable Care Act.
Lung cancer screening criteria include current and former smokers ages 55 to 77-80 years, depending on the organization with ≥30-pack-year smoking history and if a former smoker, who have quit within 15 years. Based on these criteria, it is estimated that 8.7 million adults in the United States are eligible for lung cancer screening.8 SPNs are detected at an average rate of 20% per round of screening with >90% of nodules being benign.9 The use of standardized reporting protocols such as Lung Imaging Reporting and Data System, which uses a 6-mm threshold for lung nodule detection (compared to 4 mm used by the National Lung Screening Trial), has reduced the positivity of LDCT to 10%.10,11 In addition to the large number of SPNs detected incidentally on computed tomography scans performed for reasons other than lung cancer screening, further adoption of LDCT will undoubtedly increase the numbers of SPN detected that need to be evaluated and followed.
Primary care physicians (PCPs) are stewards of preventative care and are integral to the successful implementation of lung cancer screening.12 PCPs may decide to manage SPN by themselves or may refer incidental SPN (detected through LDCT or otherwise) to specialists, including thoracic surgeons, pulmonologists, and even oncologists. Little is known about how comfortable PCPs feel on managing SPN and their perceptions on who should best follow up SPN. In addition, it is unknown whether PCP confidence in managing detected SPN is a barrier to adherence to lung cancer screening guidelines. Understanding these issues will be critical to working with PCP to properly manage the large numbers of patients with SPN identified on LDCT as adherence to lung cancer screening guidelines improves.
These findings may be of particular interest to thoracic surgeons who may prefer to receive referrals for and to follow up screen-detected SPN themselves or in the setting of a multidisciplinary cancer screening center. We recently conducted a survey of PCPs in Los Angeles County designed to understand current practice patterns in the utilization of LDCT and to identify barriers to lung cancer screening.13 The aim of this current study was to examine the perceptions and practices of PCP in the detection and management of SPN and to ascertain whether they affect the use of lung cancer screening.
METHODS
Participants
We randomly identified a cohort of 1840 PCPs using the American Medical Association’s Physician Masterfile. Eligible participants were self-identified as family physicians, general internists, and general practitioners who were less than 75 years of age and were listed as practicing in Los Angeles County with complete demographic and available contact information. Of the initial cohort, 456 PCPs were excluded from the study because of one of the following reasons: (1) return of the introductory letter or subsequent questionnaire in the mail; (2) inability to contact the physician by telephone because of lack of current contact information; (3) physician relocation outside Los Angeles County; (4) was found to be unlicensed, sanctioned, retired, deceased, or was identified as a subspecialist. The final cohort of 1384 eligible PCPs was surveyed between January and October 2015.
Survey Methods
Introductory letters were initially mailed in batches to all eligible PCPs inviting them to participate in the survey and offering a chance to win a prize from a drawing valued at $450 (Apple iPad or compact refrigerator). This was followed by the first mailing of the main survey questionnaire, designed using TeleForm (OpenText Corporation, Waterloo, Ontario), to the listed addresses of all eligible participants in similar batches. Physicians were given the option to complete the paper surveys and to return them by self-addressed envelopes, by fax, or electronically through a link to the same questionnaire, designed using DatStat (DatStat, Incorporated. Seattle, WA), that could be submitted online. Each survey candidate then received a follow-up telephone call approximately 2-3 weeks after the initial survey mailing as an encouragement and a reminder to participate. This process was repeated up to 2 times with participants who failed to return the survey within 2 months after the initial survey mailing. Follow-up and reminders were sent by mail, fax, or email when available. At the completion of the study, 20 survey respondents were randomly selected to receive a prize of their choice.
The present study was reviewed and approved by the City of Hope Institutional Review Board (#14228). Surveys included an information sheet on the study and an experimental subject’s bill of rights.
Questionnaire
The PCP survey contained 54 questions organized into 4 sections (Appendix 1) and was derived in part from a prior survey on lung cancer screening conducted by Klabunde et al from 2006 to 2007.14 Details of each section have been previously described.13 Appendix 2 provides definitions for items used in the survey. Sections pertaining to this current study included part B, which assessed how frequently PCP ordered LDCT for lung cancer screening in asymptomatic patients over the previous 12 months. Additionally, this section included questions pertaining to the proportion of high-risk patients referred for LDCT or to a lung cancer screening program. Section C6 ascertained physician confidence in the management of SPN, preference in self-managing SPN, and referral patterns of patients with SPN to a pulmonologist, a surgeon, or an oncologist. The final survey section collected demographic data. Answer choices were organized using 3- to 5-point Likert scales. In cases of 3-point Likert scales, responses “somewhat agree” and “strongly agree” were combined as “agree,” whereas “somewhat disagree” and “strongly disagree” were combined as “disagree.”
Data Collection and Statistical Analysis
All paper questionnaires by mail and fax were scanned and responses verified through Teleform. Data collected through Teleform were merged with data collected from DatStat online questionnaires to create a comprehensive dataset for analysis. Missing responses comprised of less than 3% of total responses and were excluded from most analyses. Descriptive statistics were used to present provider and practice characteristics. These were then compared between survey responders and nonresponders as well as family practitioners and internists using chi-squared test with the level of significance set at P < 0.05. We similarly compared attitudes and referral patterns between family practitioners and internists on managing patients with pulmonary nodules. Logistic regression was used to identify predictors of confidence and preference for the self-management of SPN and to determine whether attitudes and management practices pertaining to SPN were associated with the use of LDCT, referral to lung cancer screening centers, or initiation of discussion about lung cancer screening. All statistical analyses were performed on SAS (version 9.3, Cary, North Carolina).
RESULTS
There were 250 survey respondents (18.1%). As previously described, there were no significant demographic differences between survey respondents and nonrespondents when comparing age, sex, and specialty distribution.13 Two-thirds of the respondents were male and the majority had been in practice for more than 10 years (87%). At least half of respondents were in private practice and 60% were part of a single-specialty small practice (<5 physicians). The respondents were 45% family practitioners, 54% internists, and 1% general practitioners. Family practitioners and internists shared similar demographic and practice characteristics with the exception that the latter were more frequently in a larger, multispecialty practice and saw a lower volume of patients weekly than family practitioners (Table 1).
Table 1.
Physician and Practice Characteristics of Survey Respondents (N = 250)
| All Respondents N = 250 (%) |
Family Practitioners N = 112 (45) |
Internists N = 136 (54) |
P | ||
|---|---|---|---|---|---|
| Physician characteristics | |||||
| Age | <40 | 15 (6) | 5 (4) | 10 (7) | 0.77 |
| 40-49 | 72 (29) | 30 (27) | 42 (31) | ||
| 50-59 | 85 (34) | 40 (36) | 44 (32) | ||
| 60-69 | 62 (25) | 30 (27) | 31 (23) | ||
| 70 + | 16 (6) | 7 (6) | 9 (7) | ||
| Sex | Male | 161 (64) | 68 (61) | 91 (67) | 0.31 |
| Female | 89 (36) | 44 (39) | 45 (33) | ||
| Years in practice | 0-4 | 3 (1) | 0 (0) | 3 (2) | 0.17 |
| 5-10 | 27 (11) | 11 (10) | 16 (12) | ||
| 11-20 | 86 (34) | 34 (30) | 52 (38) | ||
| 21-30 | 73 (29) | 40 (36) | 33 (24) | ||
| >30 | 60 (24) | 26 (23) | 32 (24) | ||
| Not answered | 1 (0) | 1 (1) | 0 (0) | ||
| Practice characteristics | |||||
| Practice type | Full or part owner of practice | 125 (50) | 53 (47) | 70 (51) | 0.31 |
| Employee of physician-owned practice | 18 (7) | 10 (9) | 8 (6) | ||
| Employee of a large medical group or health-care system | 43 (17) | 18 (16) | 25 (18) | ||
| Employee of staff or group model HMO | 25 (10) | 16 (14) | 9 (7) | ||
| Employee of university hospital or clinic | 25 (10) | 9 (8) | 16 (12) | ||
| Employee of a non–university affiliated hospital or clinic | 13 (5) | 5 (4) | 8 (6) | ||
| Not answered | 1 (0) | 1 (1) | 0 (0) | ||
| Practice size | 1 | 64 (26) | 29 (26) | 34 (25) | 0.016 |
| 2-5 | 86 (34) | 45 (40) | 40 (29) | ||
| 6-15 | 43 (17) | 15 (13) | 28 (21) | ||
| 16-49 | 34 (14) | 19 (17) | 15 (11) | ||
| 50-99 | 10 (4) | 1 (1) | 9 (7) | ||
| 100 + | 12 (5) | 2 (2) | 10 (7) | ||
| Not answered | 1 (0) | 1 (1) | 0 (0) | ||
| Practice specialty mix | Single specialty | 151 (60) | 73 (65) | 77 (57) | 0.046 |
| Multispecialty | 95 (38) | 35 (31) | 59 (43) | ||
| Other | 1 (0) | 1 (1) | 0 (0) | ||
| Not answered | 3 (1) | 3 (3) | 0 (0) | ||
| Type of health record system used | Paper charts | 34 (14) | 20 (18) | 13 (10) | 0.30 |
| Partial EHR | 12 (5) | 6 (5) | 6 (4) | ||
| Transitioning from paper to full EHR | 40 (16) | 14 (13) | 25 (18) | ||
| Full EHR | 161 (64) | 71 (63) | 90 (66) | ||
| Not answered | 3 (1) | 1 (1) | 2 (1) | ||
| Percent of uninsured patients | 0-5 | 167 (67) | 79 (71) | 87 (64) | 0.58 |
| 6-25 | 55 (22) | 21 (19) | 33 (24) | ||
| 26-50 | 8 (3) | 4 (4) | 4 (3) | ||
| 51-75 | 2 (1) | 0 (0) | 2 (1) | ||
| 76-100 | 4 (2) | 1 (1) | 3 (2) | ||
| Not answered | 14 (6) | 7 (6) | 7 (5) | ||
| Percent of patients insured by Medi-Cal | 0-5 | 115 (46) | 54 (48) | 60 (44) | 0.55 |
| 6-25 | 70 (28) | 31 (28) | 39 (29) | ||
| 26-50 | 26 (10) | 9 (8) | 17 (13) | ||
| 51-75 | 21 (8) | 11 (10) | 9 (7) | ||
| 76-100 | 7 (3) | 4 (4) | 3 (2) | ||
| Not answered | 11 (4) | 3 (3) | 8 (6) | ||
| Patient volume during typical week | ≤25 | 23 (9) | 7 (6) | 16 (12) | 0.036 |
| 26-50 | 36 (14) | 10 (9) | 26 (19) | ||
| 51-75 | 63 (25) | 27 (24) | 36 (26) | ||
| 76-100 | 69 (28) | 37 (33) | 32 (24) | ||
| 101-125 | 32 (13) | 17 (15) | 15 (11) | ||
| 126+ | 20 (8) | 12 (11) | 6 (4) | ||
| Not answered | 7 (3) | 2 (2) | 5 (4) | ||
EHR, electronic health record. Medi-Cal is the health insurance for low-income persons including families, seniors, those with disability, and pregnant women in California.
According to Table 2, 55.5% of PCPs would somewhat or strongly agree to personally managing a lung nodule detected on imaging (median response: somewhat agree). However, when asked regarding preference to self-managing, the evaluation and follow-up of said nodule, only 21.3% strongly or somewhat agreed (median response: somewhat disagree). In fact, 75.4% agreed to most commonly referring a patient with a pulmonary nodule or mass to a pulmonologist for evaluation. Additionally, 28.9% would commonly refer the same patient to a surgeon and 24.2% would refer to an oncologist (Fig. 1).
Table 2.
Primary Care Physicians’ Perceptions and Practices in Pulmonary Nodule Management
| N | Median Physicia n Response (IQR)* |
Physicians Who Strongly or Somewhat Agree (%) |
|
|---|---|---|---|
| When a lung nodule is detected on CXR or CT, I feel confident managing the workup. | 247 | 2 (2-4) | 55.5% (n = 137) |
| I prefer to manage lung nodule evaluation and follow-up myself rather than to have a specialist do so. | 249 | 4 (3-5) | 21.3% (n = 53) |
| I most commonly refer a patient with a pulmonary nodule or mass to a pulmonologist for evaluation. | 248 | 2 (1-2) | 75.4% (n = 187) |
| I most commonly refer a patient with a pulmonary nodule or mass to a surgeon for evaluation. | 249 | 4 (2-4) | 28.9% (n = 72) |
| I most commonly refer a patient with a pulmonary nodule or mass to an oncologist for evaluation. | 248 | 4 (3-5) | 24.2% (n = 60) |
CXR, chest radiography; IQR, interquartile range.
Results do not include “not answered” or missing responses.
Median physician response according to Likert scale: 1 = strongly agree, 2 = somewhat agree, 3 = neither agree nor disagree, 4 = somewhat disagree, and 5 = strongly disagree.
Figure 1.
Primary care physicians’ confidence and preferences in the management of pulmonary nodules. “Agree” includes all somewhat agree and strongly agree responses. “Disagree” includes all somewhat disagree and strongly disagree responses.
When comparing responses from family practitioners and internists in Figure 2, the latter more frequently expressed confidence in managing the workup of imaging-detected SPN (46% vs 63%, P = 0.041). However, the bulk of both types of practitioners disagreed to self-managing the evaluation and follow-up of SPN (65% vs 59%, P = 0.41; Table 3), preferring instead to refer to a pulmonologist (77% vs 74%, P = 0.58), a surgeon (26 vs 31%, P = 0.54), or an oncologist (22% vs 25%, P = 0.68). Univariate analysis displayed in Table 4 demonstrated that PCPs aged 50-59 years were more likely to feel confident managing SPN than those 40 years old or younger (odds ratio [OR] 3.36, 95% confidence interval [CI] 1.05-10.73). This trend was not observed in other age groups likely because of the small sample sizes. PCPs with an internal medicine specialty were 2-fold more likely to feel confident in managing SPN workup compared to those with family medicine specialty (OR 1.95, 95% CI 1.09-3.48). Sex, years in practice, practice type, using of electronic medical records, percentage uninsured, and patient volume were not predictive of PCP confidence in managing pulmonary nodules.
Figure 2.
PCPs’ self-reported confidence in managing solitary pulmonary nodules by specialty. “Agree” includes all somewhat agree and strongly agree responses. “Disagree” includes all somewhat disagree and strongly disagree responses.
Table 3.
Primary Care Physician Perceptions and Practices in Pulmonary Nodule Management by Specialty
| Family Practitioners N = 112 (45%) |
Internists N = 136 (54%) |
P Value | ||
|---|---|---|---|---|
| When a lung nodule is detected on CXR or CT, I feel confident managing the workup. | Agree | 51 (46) | 82 (63) | 0.041 |
| Disagree | 40 (36) | 33 (25) | ||
| Neutral | 19 (17) | 16 (12) | ||
| I prefer to manage the lung nodule evaluation and the follow-up myself rather than to have a specialist do so. | Agree | 18 (16) | 31 (23) | 0.41 |
| Disagree | 71 (65) | 79 (59) | ||
| Neutral | 21 (19) | 23 (17) | ||
| I most commonly refer a patient with a pulmonary nodule or mass to a pulmonologist for evaluation. | Agree | 84 (77) | 99 (74) | 0.58 |
| Disagree | 19 (17) | 22 (17) | ||
| Neutral | 6 (6) | 12 (9) | ||
| I most commonly refer a patient with a pulmonary nodule or mass to a surgeon for evaluation. | Agree | 29 (26) | 41 (31) | 0.54 |
| Disagree | 62 (56) | 75 (56) | ||
| Neutral | 19 (17) | 17 (13) | ||
| I most commonly refer a patient with a pulmonary nodule or mass to an oncologist for evaluation. | Agree | 24 (22) | 33 (25) | 0.68 |
| Disagree | 73 (67) | 82 (62) | ||
| Neutral | 12 (11) | 18 (14) |
Results do not include “not answered” or missing responses.
“Agree” includes all somewhat agree and strongly agree responses.
“Disagree” includes all somewhat disagree and strongly disagree responses.
Table 4.
Factors Associated With Primary Care Physicians’ Confidence and Preferences in Pulmonary Nodule Management
| Variable | Confidence Managing Pulmonary Nodules |
Prefers to Self-manage |
||
|---|---|---|---|---|
| N (%) | OR (95% CI) | N (%) | OR (95% CI) | |
| Age | ||||
| <40 | 15 (7.2) | 1.00 (reference) | 13 (6.5) | 1.00 (reference) |
| 40-49 | 61 (29.3) | 2.03 (0.65-6.34) | 55 (27.4) | 0.70 (0.18-2.64) |
| 50-59 | 63 (30.3) | 3.36 (1.05-10.73)* | 68 (33.8) | 1.08 (0.30-3.88) |
| 60-69 | 56 (26.9) | 1.77 (0.56-5.56) | 50 (24.9) | 0.49 (0.12-1.97) |
| 70 + | 13 (6.3) | 1.83 (0.40-8.27) | 15 (7.5) | 0.56 (0.10-3.17) |
| Sex | ||||
| Male | 135 (64.9) | 1.00 (reference) | 123 (61.2) | 1.00 (reference) |
| Female | 73 (35.1) | 0.88 (0.49-1.59) | 78 (38.8) | 0.34 (0.16-0.71)* |
| Years in practice | ||||
| 0-4 | 3 (1.4) | Undetermined | 3 (1.5) | 5.69 (0.48-68.13) |
| 5-10 | 26 (12.5) | 0.42 (0.16-1.11) | 22 (10.9) | 1.07 (0.34-3.31) |
| 11-20 | 69 (33.2) | 0.89 (0.41-1.96) | 68 (33.8) | 0.88 (0.38-2.04) |
| 21-30 | 58 (27.9) | 0.63 (0.28-1.41) | 57 (28.4) | 0.93 (0.39-2.22) |
| >30 | 51 (24.5) | 1.00 (reference) | 50 (24.9) | 1.00 (reference) |
| Specialty | ||||
| Family practice | 91 (43.8) | 1.00 (reference) | 89 (44.3) | 1.00 (reference) |
| Internal medicine | 115 (55.3) | 1.95 (1.09-3.48)* | 110 (54.7) | 1.55 (0.80-3.00) |
| Primary practice arrangement | ||||
| Full or part owner of practice | 104 (50.0) | 1.00 (reference) | 103 (51.2) | 1.00 (reference) |
| Employee of physician-owned practice | 12 (5.8) | 0.35 (0.10-1.17) | 14 (7.0) | 2.47 (0.78-7.82) |
| Employee of large medical group or health-care system | 38 (18.3) | 1.57 (0.67-3.67) | 31 (15.4) | 1.81 (0.76-4.30) |
| Employee of staff or group model HMO | 21 (10.1) | 0.79 (0.30-2.08) | 18 (9.0) | 1.65 (0.56-4.85) |
| Employee of university hospital or clinic | 21 (10.1) | 0.53 (0.21-1.38) | 22 (10.9) | 0.52 (0.14-1.91) |
| Employee of non–university affiliated hospital or clinic | 11 (5.3) | 0.58 (0.17-2.05) | 12 (6.0) | 0.30 (0.04-2.44) |
| Practice specialty mix | ||||
| Single specialty | 124 (59.6) | 1.00 (reference) | 124 (61.7) | 1.00 (reference) |
| Multi-specialty | 81 (38.9) | 1.55 (0.85-2.82) | 73 (36.3) | 1.01 (0.52-1.96) |
| Practice size | ||||
| 1 | 57 (27.4) | 1.00 (reference) | 52 (25.9) | 1.00 (reference) |
| 2-5 | 64 (30.8) | 0.69 (0.33-1.44) | 73 (36.3) | 1.41 (0.61-3.26) |
| 6-15 | 37 (17.8) | 1.35 (0.54-3.36) | 29 (14.4) | 1.68 (0.60-4.70) |
| 16-49 | 29 (13.9) | 0.82 (0.32-2.07) | 27 (13.4) | 1.06 (0.35-3.28) |
| 50-99 | 10 (4.8) | 2.00 (0.39-10.36) | 8 (4.0) | 0.53 (0.06-4.80) |
| 100 + | 10 (4.8) | 0.75 (0.19-2.98) | 11 (5.5) | 2.13 (0.53-8.61) |
| Type of health record system used | ||||
| Paper charts | 30 (14.4) | 1.00 (reference) | 28 (13.9) | 1.00 (reference) |
| Partial EHR | 10 (4.8) | 0.75 (0.17-3.28) | 10 (5.0) | 4.00 (0.77-20.82) |
| Transitioning from paper to full EHR | 29 (13.9) | 0.71 (0.25-2.04) | 36 (17.9) | 0.75 (0.17-3.31) |
| Full EHR | 136 (65.4) | 1.01 (0.44-2.34) | 124 (61.7) | 2.75 (0.89-8.47) |
| Percent uninsured patients | ||||
| 0-5 | 135 (64.9) | 1.00 (reference) | 128 (63.7) | 1.00 (reference) |
| 6-25 | 49 (23.6) | 0.71 (0.37-1.39) | 49 (24.4) | 0.71 (0.32-1.57) |
| 26-50 | 6 (2.9) | Undetermined | 6 (3.0) | 0.55 (0.06-4.90) |
| 76-100 | 4 (1.9) | 1.60 (0.16-15.83) | 3 (1.5) | 1.38 (0.12-15.74) |
| Percent of Medi-Cal insured patients | ||||
| 0-5 | 97 (46.6) | 1.00 (reference) | 89 (44.3) | 1.00 (reference) |
| 6-25 | 56 (26.9) | 1.61 (0.78-3.32) | 54 (26.9) | 1.08 (0.51-2.27) |
| 26-50 | 21 (10.1) | 0.44 (0.17-1.15) | 24 (11.9) | 0.37 (0.10-1.34) |
| 51-75 | 19 (9.1) | 1.28 (0.45-3.66) | 20 (10.0) | 0.45 (0.12-1.68) |
| 76-100 | 6 (2.9) | 2.95 (0.33-26.26) | 6 (3.0) | 0.51 (0.06-4.60) |
| Patient volume (weekly) | ||||
| ≤25 | 18 (8.7) | 1.00 (reference) | 20 (10.0) | 1.00 (reference) |
| 26-50 | 30 (14.4) | 1.31 (0.40-4.22) | 32 (15.9) | 1.59 (0.36-7.01) |
| 51-75 | 56 (26.9) | 1.80 (0.62-5.27) | 50 (24.9) | 2.20 (0.56-8.71) |
| 76-100 | 60 (28.8) | 2.00 (0.69-5.82) | 55 (27.4) | 1.93 (0.49-7.61) |
| 101-125 | 24 (11.5) | 2.43 (0.68-8.70) | 25 (12.4) | 1.42 (0.29-6.81) |
| 126+ | 15 (7.2) | 4.00 (0.84-19.16) | 13 (6.5) | 6.61 (1.28-34.1)* |
| Feels confident in managing workup when pulmonary nodule is detected | ||||
| Agree | – | – | 104 (51.7) | 1.00 (reference) |
| Disagree | – | – | 70 (34.8) | 0.14 (0.06-0.36)* |
| Neutral | – | – | 25 (12.4) | 0.29 (0.09-0.91)* |
P < 0.05.
“Agree” includes all somewhat agree and strongly agree responses.
“Disagree” includes all somewhat disagree and strongly disagree responses.
Undetermined–value unable to be calculated due to the small sample size.
Similar univariate analysis evaluating predictors of PCP preference for self-management of SPN demonstrated that PCPs who were women reported preference to self-managing imaging-detected pulmonary nodules one-third as frequently as PCPs who were men (OR 0.34, 95% CI 0.16-0.71). Additionally the few PCPs who saw the highest volume of patients (>126 per week) were 6 times more likely to prefer to self-manage SPN (OR 6.61, 95% CI 1.28-34.14) compared with PCPs seeing fewer patients per week. As expected, those who reported disagreeing or were neutral to feeling confident in managing the workup of SPN were less likely to self-manage its evaluation and follow-up (OR 0.14, 95% CI 0.06-0.36, and OR 0.29, 95% CI 0.09-0.91, respectively).
In Table 5, PCPs who disagreed to preferring self-management of SPN referred a patient for lung cancer screening half as often as those who agreed (OR 0.52, 95% CI 0.27-1.00, P = 0.052). PCP confidence in managing the workup of SPN did not predict referral to a lung cancer screening center, and neither confidence nor preference to self-manage SPN was predictive of initiating a discussion regarding the benefits and harms of lung ordering LDCT for lung cancer screening.
Table 5.
Primary Care Physician Perceptions as Predictors for Lung Cancer Screening Practices
| Survey Item and Response Options | Ordered LDCT for Lung Cancer Screening |
Referred Patient to Lung Cancer Screening Center for Screening |
Initiated Discussion About Benefits and Risks of Lung Cancer Screening With Patient |
|||||||
|---|---|---|---|---|---|---|---|---|---|---|
| N (%) | OR (95% CI) | P | N (%) | OR (95% CI) | P | N (%) | OR (95% CI) | P | ||
| When a lung nodule is detected on CXR or CT, I feel confident managing the workup. | Agree | 133 (56.1) | (Ref) | – | 132 (56.2) | (Ref) | – | 131 (55.7) | (Ref) | – |
| Disagree | 71 (30.0) | 1.04 (0.58-1.85) | 0.90 | 69 (29.4) | 0.95 (0.39-2.35) | 0.91 | 71 (30.2) | 0.88 (0.46-1.71) | 0.72 | |
| Neutral | 33 (13.9) | 1.22 (0.57-2.64) | 0.61 | 34 (14.5) | 1.55 (0.56-4.33) | 0.40 | 33 (14.0) | 0.65 (0.28-1.48) | 0.30 | |
| I prefer to self-manage lung nodule evaluation and follow-up rather than to refer to a specialist. | Agree | 51 (21.3) | (Ref) | – | 50 (21.1) | (Ref) | – | 51 (21.5) | (Ref) | – |
| Disagree | 145 (60.7) | 0.52 (0.27-1.00) | 0.052 | 143 (60.3) | 0.66 (0.26-1.66) | 0.38 | 143 (60.3) | 0.72 (0.34-1.50) | 0.37 | |
| Neutral | 43 (18.0) | 1.67 (0.70-3.99) | 0.25 | 44 (18.6) | 0.83 (0.26-2.61) | 0.75 | 43 (18.1) | 1.35 (0.49-3.67) | 0.56 | |
Ref, reference.
“Agree” includes all somewhat agree and strongly agree responses.
“Disagree” includes all somewhat disagree and strongly disagree responses.
DISCUSSION
We surveyed PCPs in a large population county and found that just over half of PCP respondents reported feeling confident managing SPN detected on imaging. However, confidence in the management of SPN did not translate into practice as only 21% of PCPs reported preferring to self-manage the workup and follow-up of SPN. Women were less likely to prefer self-managing SPNs, although the reasons for this are unclear. Three out of four PCPs would refer a patient with SPN to a pulmonologist and approximately one in four would refer to a surgeon or an oncologist. Confidence in managing SPN was not associated with ordering LDCT or referring to a lung cancer screening center. PCPs who disagreed to preferring self-management of pulmonary nodules may be ordering fewer LDCTs for lung cancer screening. The present study is important because it evaluates the perceptions and practices of PCPs when confronted with incidental lung nodules. As lung cancer screening becomes increasingly more common, a large number of SPNs will be detected, which will need to be monitored, either by PCP or specialists, including thoracic surgeons and pulmonologists.
Structured reporting in lung cancer screening, most commonly Lung Imaging Reporting and Data System, typically includes recommendations for further evaluation of any detected abnormalities, short-interval follow-up imaging, or continued annual screening. Indeed, the use of Breast Imaging Reporting and Data System (BI-RADS) and the Bethesda System have greatly standardized breast cancer and cervical cancer screening results.15,16 Although structured reporting in the setting of LDCT typically includes recommendations for evaluation of detected SPN, positive findings must still be followed up and communicated to patients by practitioners knowledgeable in the evaluation, treatment, and follow-up of lung nodules and lung cancer. Understanding the knowledge, confidence, and willingness of PCPs to follow up lung nodules has important implications in the workforce needed for the successful implementation of lung cancer screening guidelines.
We also sought to determine whether confidence in evaluating SPN might be a barrier to the adoption of lung cancer screening recommendations. Despite the guidelines for lung cancer screening by the United States Preventive Services Task Force, National Comprehensive Cancer Network, and Centers for Medicare Services, we previously reported that only 52% of the PCPs surveyed ordered LDCT in the last year for lung cancer screening and even fewer (12%) referred patients to a lung cancer screening center.13 PCPs have a vital role in encouraging screening and preventative care, so it is important to identify the factors that limit PCPs from participating and facilitating screening for lung cancer. A similar survey study of PCPs at an academic institution reported that only 12% of respondents ordered LDCT for lung cancer screening.17 Another study surveying PCPs attending 2 large continued medical education events cited that 56% of respondents planned to refer patients for lung cancer screening, but only 10% had a formal lung cancer screening program in their practice.18 wenty percent of French general practitioners surveyed recommended lung cancer screening in daily practice, but 94% used inappropriate screening tools, mainly chest radiography.19
A qualitative study that interviewed 15 PCPs caring for patients with SPN reported that most PCPs did not have adequate resources with which to counsel patients about SPN, and most PCPs did not include patients when making plans for follow-up.20 Such knowledge and practice gaps would limit the coverage of lung cancer screening by Medicare, which mandates counseling and shared decision making between physicians and patients. These gaps may be remedied by educational or outreach endeavors from specialists, including thoracic surgeons from multidisciplinary screening centers who may prefer to follow up these patients. Our data suggest that confidence and preference in the self-management of SPN did not significantly correlate with the use of LDCT for the screening or the initiation of discussion about the benefits and risks of lung cancer screening. Consequently, confidence in managing SPN does not seem to be a barrier to ordering LDCT for screening. The reasons why PCPs who feel confident in self-managing SPN but choose not to are unclear and warrant further investigation.
Several important limitations exist in our study, the first of which is the low response rate to our survey, although there was no difference found in measureable demographic and practice characteristics between responders and nonresponders. This suggests that responders and nonresponders are similar. The low response rate may predispose to selection bias and our small cohort in some cases limited the ability to detect statistical significance, for instance, the potential that age affected confidence in the management of SPN. As a survey study, our data relied on the accuracy of PCPs’ self-reported attitudes and practices on the subject of lung cancer screening, which, without objective validation, is subjected to recall bias. Additionally, although Los Angeles County is a large and diverse region, our study may also have regional bias and the results may not be applicable to other parts of the country.
Despite the limitations, our study offers insight into the attitudes and practices of PCPs faced with SPN found on screening LDCT. Approximately half of PCPs did not feel confident managing these nodules, and half of those who did feel confident in working up and following up SPN chose not to, preferring, instead, to seek expert consultation. We identified that a minority of PCPs refer SPN to thoracic surgeons. However, many thoracic surgeons may prefer to manage and follow up screening-detected SPNs along with pulmonologists and oncologists. PCP education may improve the referral rates of SPN to thoracic surgeons. Multidisciplinary teams composed of thoracic surgeons, pulmonologists, and radiologists that are associated with a lung cancer screening program may be best equipped to manage SPN detected on LDCT. Promotion of screening centers of excellence may further facilitate referrals. Further study on best practices in the management of SPN, taking into account the health-care workforce, quality, and costs is warranted as lung cancer screening becomes more common.
Central Message.
Although most primary care physicians feel comfortable working up pulmonary nodules, the majority prefer to refer management and follow-up to specialists, but infrequently to thoracic surgeons.
Perspective Statement.
Solitary pulmonary nodules (SPNs) found during lung cancer screening require appropriate workup and follow-up. The role of primary care physicians in the management of SPN is unclear. In a survey of these providers, half felt confident in self-managing SPN, but most preferred specialist referral. As more SPNs are detected on imaging, it is important to identify the workforce that should be managing them.
Acknowledgment
We acknowledge the generous support of the Baum Family Foundation in support of this research.
Research reported in this publication is supported by the Lung Cancer Research Foundation, and the National Cancer Institute of the National Institutes of Health under award numbers NIH 5K12CA001727-20 (Raz) and P30CA33572 through the use of the City of Hope Survey Research Core. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Appendix
APPENDIX 1
Primary Care Physicians Survey
APPENDIX 2
Definitions
Family practitioner—physician who completed a postgraduate residency training in family medicine
Internist—physician who completed a postgraduate residency training in internal medicine
General practitioner—physician who completed medical school and 1-2 years of residency
Practice type—the type of practice that the primary care provider works for.
Full or part owner of practice
Employee of physician-owned practice
Employee of large medical group or health-care system
Employee of staff or group model or health maintenance organization (HMO)
Employee of university hospital or clinic
Employee of non–university affiliated hospital or clinic
Practice size—number of providers in the practice
Practice specialty mix—whether the provider practiced in a single-specialty, or multispecialty setting.
Type of electronic health record system used—whether provider used full or partial electronic medical record, or paper records in their practice.
Percent (%) of patients uninsured—the proportion of PCP practice treating uninsured patients.
Percent (%) of patients insured by Medi-Cal—the proportion of PCP practice treating low-income patients, including seniors, persons with disabilities, foster children, pregnant women.
Patient volume during typical week—how many patients are seen weekly in the patient’s practice.
Footnotes
Dr. Raz is a consultant for Cireca, LLC. The remaining authors have no commercial interests to disclose.
REFERENCES
- 1.National Lung Screening Trial Research T, Aberle DR, Adams AM, et al. : Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med 365:395–409, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Moyer VA, Force USPST: Screening for lung cancer: U.S. Preventive Services Task Force Recommendation Statement. Ann Intern Med 160:330–338, 2014 [DOI] [PubMed] [Google Scholar]
- 3.Wood DE, Eapen GA, Ettinger DS, et al. : Lung cancer screening. J Natl Compr Canc Netw 10:240–265, 2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Wender R, Fontham ET, Barrera E Jr, et al. : American cancer society lung cancer screening guidelines. CA Cancer J Clin 63:107–117, 2013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.American Lung Association. Providing Guidance on Lung Cancer Screening. The American Lung Association Interim Report on Lung Cancer Screening. 2015. Available at: http://www.lung.org/assets/documents/lung-cancer/lung-cancer-screening-report.pdf. Accessed June 13, 2016 [Google Scholar]
- 6.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 144:33–38, 2012 [DOI] [PubMed] [Google Scholar]
- 7.Jensen T, Chin J, Ashby L, et al. : Final National Coverage Determination on Screening for Lung Cancer with Low Dose Computed Tomography (LDCT). Baltimore, MD, The Centers for Medicare & Medicaid Services (CMS), 2015. [Google Scholar]
- 8.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 21:1049–1059, 2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Bach PB, Mirkin JN, Oliver TK, et al. : Benefits and harms of CT screening for lung cancer: A systematic review. JAMA 307:2418–2429, 2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Yip R, Henschke CI, Yankelevitz DF, et al. : CT screening for lung cancer: Alternative definitions of positive test result based on the national lung screening trial and international early lung cancer action program databases. Radiology 273:591–596, 2014 [DOI] [PubMed] [Google Scholar]
- 11.McKee BJ, Regis SM, McKee AB, et al. : Performance of ACR lung-RADS in a clinical CT lung screening program. J Am Coll Radiol 12:273–276, 2015 [DOI] [PubMed] [Google Scholar]
- 12.Emery JD, Shaw K, Williams B, et al. : The role of primary care in early detection and follow-up of cancer. Nat Rev Clin Oncol 11:38–48, 2014 [DOI] [PubMed] [Google Scholar]
- 13.Raz DJ, Wu GX, Consunji M, et al. : Perceptions and utilization of lung cancer screening among primary care physicians. J Thorac Oncol 11:1856–1862, 2016 [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 10:102–110, 2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Liberman L, Menell JH. Breast imaging reporting and data system (Bi-Rads). Radiol Clin North Am 40:409–430, v, 2002 [DOI] [PubMed] [Google Scholar]
- 16.Solomon D, Davey D, Kurman R, et al. : The 2001 Bethesda System: Terminology for reporting results of cervical cytology. JAMA 287:2114–2119, 2002 [DOI] [PubMed] [Google Scholar]
- 17.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 24:664–670, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Volk RJ, Foxhall LE. Readiness of primary care clinicians to implement lung cancer screening programs. Prev Med Rep 2:717–719, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Couraud S, Girard N, Erpeldinger S, et al. : Physicians’ knowledge and practice of lung cancer screening: A cross-sectional survey comparing general practitioners, thoracic oncologists, and pulmonologists in France. Clin Lung Cancer 14:574–580, 2013 [DOI] [PubMed] [Google Scholar]
- 20.Golden SE, Wiener RS, Sullivan D, et al. : Primary care providers and a system problem: A qualitative study of clinicians caring for patients with incidental pulmonary nodules. Chest 148:1422–1429, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]