Abstract
Objective.
To determine the frequency, components of, and factors associated with shared decision making (SDM) discussions according to electronic health record (EHR) documentation among individuals receiving lung cancer screening (LCS).
Methods.
Prospective observational cohort study of individuals undergoing LCS between February 2015 and June 2020 at four LCS centers. The primary outcome was EHR-documented SDM, defined using Medicare-designated components. A multivariable logistic regression model was used to examine predictors of EHR-documented SDM. A secondary outcome was agreement of individual’s self-report of SDM and EHR documented SDM evaluated using Cohen’s kappa statistic.
Results.
Among screened individuals, 41.9% (243/580) had EHR-documented SDM, and 71.1% (295/415) had self-reported SDM. Decision aids were used in 55.6% (135/243) of EHR-documented SDM encounters, and 21.8% (53/243) of documented SDM encounters included all Medicare-designated components. SDM was documented more frequently in individuals with body mass index (BMI) ≥25 versus <25 (adjusted odds ratio (aOR)=1.63, 95% confidence interval (95%CI):1.05-2.52), and in currently versus formerly smoking individuals (aOR=1.53, 95%CI:1.02-2.32). Non-pulmonary referring clinicians were less likely to document SDM than pulmonary clinicians (internal medicine: aOR=0.32, 95%CI:0.18-0.53, family medicine: aOR=0.08, 95%CI:0.04-0.14, other specialties: aOR=0.08, 95%CI:0.03-0.21). In a subset of 415 individuals, there was little agreement between individual self-report of SDM and EHR-documented SDM (kappa=0.184), with variation in agreement based on referring clinician specialty.
Discussion.
While EHR-documented SDM occurred in less than half of individuals receiving LCS, self-reported SDM rates were higher, suggesting SDM may be under-documented in the EHR. In addition, EHR-documented SDM was more likely in individuals with higher BMI and those referred to LCS by pulmonary clinicians. These findings indicate areas for improvement in implementation and documentation of SDM.
MeSH Keywords: decision making, shared; mass screening; lung neoplasms; early detection of cancer
Summary Sentence
Variation in documentation of SDM among individuals screened for lung cancer underscores the need for interventions that facilitate high-quality SDM use and improved documentation.
Introduction
The National Lung Screening Trial demonstrated a 20% relative decrease in lung cancer mortality with lung cancer screening (LCS) with low-dose chest CT (LDCT) compared to chest radiography.1 In 2013, the US Preventive Services Task Force (USPSTF) issued a Grade B recommendation for annual LCS with LDCT in individuals between 55-80 years of age with ≥30 pack-year smoking history who either currently smoke or previously smoked (and quit within 15 years).2 The 2021 USPSTF recommendations expand eligibility to include individuals between 50-80 years of age with ≥20 pack-year smoking history.3 Annual LCS is a preventive health benefit under the Medicare program and covered by third-party insurers under the Patient Protection and Affordable Care Act.4,5
Unique to LCS enrollment is the Center for Medicare and Medicaid Services (CMS) required and reimbursed counseling encounter for shared decision-making (SDM).4,6-11 SDM is a fundamental tenet of LCS, intended to ensure that an individual’s preferences are incorporated into the LCS decision through a balanced discussion of an individual’s risk versus benefits.10 The SDM visit is a multicomponent patient-clinician encounter, which per Medicare coverage requirements, includes: use of a LCS decision-making aid, discussion of benefits and potential harms of LCS (i.e., follow up imaging/testing, false positives, radiation exposure), counseling on the importance of LCS adherence, discussion of comorbidities, including an individual’s ability or willingness to undergo treatment for lung cancer, and counseling on smoking cessation/maintenance.4,12
Despite the CMS requirement regarding SDM, limited evidence suggests that SDM is underutilized, inconsistently utilized, and of variable quality, with rates as low as 9%.6,9,11,13,14 Direct and robust evidence from LCS programs evaluating the components of and factors associated with SDM are lacking. Our primary purpose was to determine rates of SDM among screened individuals in academic and community settings via documentation in the electronic health record (EHR) and to evaluate individual and ordering clinician characteristics associated with EHR-documented SDM. We also evaluated documentation of the CMS-designated components of SDM and compared self-report of SDM with EHR-documented SDM.
Methods
Data Sources and Study Population
Data for this investigation were collected as part of a National Cancer Institute-funded registry which prospectively collects individual, radiology, and outcomes data on individuals undergoing LCS at participating locations.15 Specifically, the registry collects individual sociodemographic and lung cancer risk factor information, LCS exam and follow-up information (dates of exams, exam findings, Lung-RADS assessment, recommendation for follow-up), and outcomes data from abstracted EHR and radiology reports, EHR data exports, and linkage to state cancer registry data. This study was approved by the Institutional Review Board.
We included adults undergoing LDCT for LCS between February 2015 through June 2020 at four screening sites. This analysis was limited to the subset of registry participants who completed a one-page health history questionnaire (HHQ) that included questions on sociodemographic information (sex, race/ethnicity, highest level of educational attainment), smoking history, comorbid conditions, lung cancer risk factors, and a question about SDM. The SDM question asked: “At the visit where your lung cancer screening exam was ordered, did your doctor talk to you about the risks and benefits of screening?” with response options of “yes,” “no,” and “I don’t remember.” The first exam was selected for inclusion among individuals with more than one LCS exam. We excluded individuals with unknown smoking status or those who had never smoked (n=5).
Study team members performed chart abstraction of the following from the EHR: Documentation of a SDM conversation in the visit notes, type and specialty of the clinician ordering the LCS exam, and the CMS required components of the SDM conversation, including (1) use of decision aid, (2) mention of the benefits and harms of LCS including (a) follow-up diagnostic testing, (b) overdiagnosis, (c) false-positive rate, (d) radiation exposure, (3) the importance of adherence to annual screening, (4) impact of comorbidities, (5) ability or willingness to undergo diagnosis and treatment, (6) importance of maintaining or beginning cigarette abstinence and among those who currently smoked: (a) smoking cessation counseling provided, and (b) smoking cessation referral.4
Measures
Individuals were categorized as having SDM documented in the EHR if the abstracted visit note mentioned SDM or discussion of the risk and benefits occurred via individual-clinician conversation, or if the EHR documented that an educational LCS brochure was made available to the individual. Individuals were categorized as having self-reported SDM if they selected “yes” to the SDM HHQ and were categorized as not having SDM if they selected “no” or “I don’t remember.”
Individual sociodemographic characteristics included age (<65 and 65+ years) and race (White, Black, all other races/unknown). Participant location was categorized as rural/urban using the residential zip code and the 2013 Rural-Urban Continuum Codes, with Codes 1-3 classified as urban and codes 4-9 classified as rural.16 Other individual characteristics included: Body mass index (BMI) categorized as underweight/normal (BMI <25), overweight/obese (BMI ≥25); self-report of highest educational level (< high school degree, high school degree, some training after high school, some college, college graduate, or post-graduate/professional training); marital status (partnered, not partnered); EHR documented employment (employed, retired/on leave/disabled, or not employed); smoking status (current or former); EHR reported insurance type (Medicare, including those with an Advantage plan or dually eligible versus all other insurance types); self-report of personal history of cancer; self-report of first-degree family history of lung cancer (mother, father, brother, sister, or child); and self-report of chronic obstructive pulmonary disease (COPD; yes, no) or hypertension (yes, no). The ordering clinician type was classified as attending physician, physician in training, including fellows and residents, or advanced practice provider (Physician Assistant or Nurse Practitioner). The specialty of the ordering clinician was classified as Pulmonology, Internal Medicine or General Medicine, Family Medicine or Primary Care, or Other (including Gastroenterology, Geriatrics, Hematology/Oncology, Infectious Disease, Nephrology).
Statistical Analysis
Multivariable logistic regression was used to examine predictors of EHR-documented SDM, which were reported as adjusted odds ratios (aOR) with 95% confidence intervals (95% CIs). The adjusted model included sex, race, age, BMI, marital status, smoking status, insurance, COPD, ordering clinician type/specialty, and LCS site. Among individuals with EHR-documented SDM, we described the components of the SDM encounter. We did not adjust for multiple comparisons.
Among a subset of individuals who answered the HHQ item about SDM, we assessed agreement between the individual’s self-report of SDM and EHR documentation of SDM using Cohen’s kappa statistic. In addition, we determined if the agreement of self-reported SDM and EHR-documented SDM varied by the specialty and type of the clinician ordering the LCS. Analyses were conducted using SAS version 9.4 (SAS Institute Inc.), Cary, NC.
Results
Study Cohort and Demographics
We identified 580 individuals who received LCS during the study period and completed the HHQ (Table 1). This cohort of individuals was similar to the 3,768 individuals who did not complete a HHQ regarding age, sex, race, smoking status, COPD status, and BMI. The cohort was evenly split between males and females, with approximately 76% reporting white race, 20% reporting Black race, and 4% reporting other or unknown race. The median age was 65 years, and most individuals lived in urban areas. Almost 70% of individuals were overweight or obese. In terms of smoking status, 53% reported formerly smoking, and 47% reported currently smoking. Most individuals were referred for screening by an attending physician (66.7%) from an Internal Medicine or General Medicine specialty (36.6%).
Table 1:
Characteristics of Individuals Undergoing Lung Cancer Screening Overall and by Documentation of Shared Decision-Making (SDM) in the Electronic Health Record (EHR)
| Characteristics | All Individuals with Lung Cancer Screening |
Documentation of SDM in EHR |
No Documentation of SDM in EHR |
|---|---|---|---|
| N (% a) | N (% b) | N (% b) | |
| Total | 580 (100.0) | 243 (41.9) | 337 (58.1) |
| Sex | |||
| Male | 294 (50.7) | 107 (36.4) | 187 (63.6) |
| Female | 286 (49.3) | 136 (47.6) | 150 (52.4) |
| Race | |||
| White | 440 (75.9) | 178 (40.5) | 262 (59.5) |
| Black | 118 (20.3) | 57 (48.3) | 61 (51.7) |
| Other and Unknown | 22 (3.8) | 8 (36.4) | 14 (63.6) |
| Age (years) | |||
| Mean (SD) | 65.7 (6.4) | 66.0 (6.2) | 65.6 (6.5) |
| Median (IQR) | 65.0 (60.8-70.2) | 65.1 (60.9-70.3) | 64.9 (60.5-70.2) |
| <65 | 289 (49.8) | 120 (41.5) | 169 (58.5) |
| ≥65 | 292 (50.2) | 123 (42.3) | 169 (58.1) |
| Residence location | |||
| Rural | 98 (16.9) | 45 (45.9) | 53 (54.1) |
| Urban | 482 (83.1) | 198 (41.1) | 284 (58.9) |
| Body mass index | |||
| Underweight/Normal (<25) | 175 (30.2) | 62 (35.4) | 113 (64.6) |
| Overweight/Obese (≥25) | 401 (69.1) | 180 (44.9) | 221 (55.1) |
| Unknowna | 4 (0.7) | 1 (25.0) | 3 (75.0) |
| Education | |||
| Less than high school degree | 59 (10.2) | 22 (37.3) | 37 (62.7) |
| High school degree | 107 (18.4) | 49 (45.8) | 58 (54.2) |
| Some training after high school | 40 (6.9) | 13 (32.5) | 27 (67.5) |
| Some college | 102 (17.6) | 44 (43.1) | 58 (56.9) |
| College graduate | 68 (11.7) | 27 (39.7) | 41 (60.3) |
| Post-Graduate/Professional | 43 (7.4) | 14 (32.6) | 29 (67.4) |
| Unknown | 161 (27.8) | 74 (46.0) | 87 (54.0) |
| Marital status | |||
| Partnered | 274 (47.2) | 103 (37.6) | 171 (62.4) |
| Not partnered | 288 (49.7) | 131 (45.5) | 157 (54.5) |
| Unknown | 18 (3.1) | 9 (50.0) | 9 (50.0) |
| Employment | |||
| Employed | 98 (16.9) | 41 (41.8) | 57 (58.2) |
| Retired/Leave/Disabled | 356 (61.4) | 147 (41.3) | 209 (58.7) |
| Not employed | 95 (16.4) | 43 (45.3) | 52 (54.7) |
| Unknown | 31 (5.3) | 12 (38.7) | 19 (61.3) |
| Smoking status | |||
| Current | 273 (47.1) | 116 (42.5) | 157 (57.5) |
| Former | 307 (52.9) | 127 (41.4) | 180 (58.6) |
| Insurance type | |||
| Any Medicare | 375 (64.7) | 158 (42.1) | 217 (57.9) |
| Not Medicare/All other types | 195 (33.6) | 81 (41.5) | 114 (58.5) |
| Unknown | 10 (1.7) | 4 (40.0) | 6 (60.0) |
| Personal history of cancer | |||
| Yes | 101 (17.4) | 47 (46.5) | 54 (53.5) |
| No | 479 (82.6) | 196 (40.9) | 283 (59.1) |
| First-degree family history of cancer | |||
| Yes | 133 (22.9) | 56 (42.1) | 77 (57.9) |
| No | 447 (77.1) | 187 (41.8) | 260 (58.2) |
| Chronic obstructive pulmonary disease | |||
| Yes | 263 (45.3) | 128 (48.7) | 135 (51.3) |
| No | 317 (54.7) | 115 (36.3) | 202 (63.7) |
| Hypertension | |||
| Yes | 220 (37.9) | 86 (39.1) | 134 (60.9) |
| No | 360 (62.1) | 157 (43.6) | 203 (56.4) |
| Clinician type | |||
| Attending physician | 387 (66.7) | 168 (43.4) | 219 (56.6) |
| Physician-in-training | 144 (24.8) | 66 (45.8) | 78 (54.2) |
| Advance practice provider | 46 (7.9) | 9 (19.6) | 37 (80.4) |
| Unknown | 3 (0.5) | 0 (0) | 3 (100.0) |
| Clinician specialty | |||
| Pulmonology | 150 (25.9) | 107 (71.3) | 43 (28.7) |
| Internal Medicine or General Medicine | 212 (36.6) | 96 (45.3) | 116 (54.7) |
| Family Medicine or Primary Care | 172 (29.7) | 32 (18.6) | 140 (81.4) |
| Otherc | 46 (7.9) | 8 (17.4) | 38 (82.6) |
| Year of lung cancer screening exam | |||
| 2015/2016 | 190 (32.8) | 82 (43.2) | 108 (56.8) |
| 2017/2018 | 243 (41.9) | 103 (42.4) | 140 (57.6) |
| 2019/2020d | 147 (25.3) | 58 (39.5) | 89 (60.5) |
Column percent.
Row percent.
Other specialty includes: gastroenterology, geriatrics, hematology/oncology, infectious, nephrology.
2020 complete through June 15, 2020.
EHR Documentation of SDM
Among the 580 individuals who received LCS during the study period and completed the HHQ, 41.9% (243/580) had receipt of SDM documented in the EHR (Table 1). Multivariable analyses examining predictors of EHR-documented SDM, BMI, smoking status, and ordering clinician specialty were significantly associated with SDM documentation (Table 2). Specifically, individuals with BMI ≥25 were more likely to receive SDM compared to those with BMI <25 (aOR=1.63, 95% CI:0.1.05-2.52). In addition, compared to formerly smoking individuals, currently smoking individuals had higher frequency of EHR-documented (aOR=1.53, 95% CI:1.02-2.32). Additionally, compared with individuals who had LDCT ordered by clinicians in Pulmonology, individuals who had LDCT ordered by clinicians in Internal Medicine (aOR=0.32, 95%CI:0.18-0.53), Family Medicine (aOR=0.08, 95%CI: 0.04-0.14), or another non-pulmonary specialty (aOR=0.08, 05%CI:0.03-0.21) were less likely to have EHR-documented SDM.
Table 2:
Characteristics Associated with Electronic Health Record-Documented Shared Decision-Making
| Characteristic | Adjusted Odds Ratio (95% Confidence Interval) a |
|---|---|
| Sex: | |
| Male | Reference |
| Female | 1.48 (0.99-2.22) |
| Race: | |
| White | Reference |
| Black | 1.45 (0.89-2.37) |
| Other | 0.58 (0.18-1.82) |
| Age: | |
| <65 years | Reference |
| 65+ years | 1.31 (0.81-2.12) |
| BMI: | |
| Underweight/normal weight (<25) | Reference |
| Overweight/obese (25+) | 1.63 (1.05-2.52) |
| Marital status: | |
| Not partnered | Reference |
| Partnered | 0.71 (0.47-1.08) |
| Smoking status: | |
| Former | Reference |
| Current | 1.53 (1.02-2.32) |
| Insurance: | |
| All others | Reference |
| Medicare | 0.86 (0.51-1.42) |
| COPD status: | |
| No COPD | Reference |
| COPD | 0.98 (0.64-1.51) |
| Clinician type: | |
| Faculty Physician | Reference |
| Physician in Training | 0.75 (0.47-1.20) |
| Clinician type: | |
| Faculty Physician | Reference |
| Advance Practice Provider | 0.46 (0.19-1.14) |
| Clinician specialty: | |
| Pulmonology | Reference |
| Internal Medicine | 0.32 (0.18-0.53) |
| Family Medicine | 0.08 (0.04-0.14) |
| Other | 0.08 (0.03-0.21) |
The logistic regression model is adjusted for sex, race, age, body mass index (BMI), marital status, smoking status, insurance, chronic obstructive pulmonary disease (COPD), clinician type, clinician specialty, and screening site/facility.
Documentation of Components of SDM
Among the 243 individuals with EHR-documented SDM, there was variability in documentation of SDM components (Table 3).4 Over half of SDM encounters documented use of a decision aid (55.6%). The majority of currently smoking individuals received smoking cessation counseling and/or were referred to a smoking cessation program (75.0%). While 81.5% of SDM encounters had EHR documentation that the benefits and risks of LCS were discussed, documentation of the SDM components was variable, ranging from 44.0% of encounters discussing the individual’s willingness or ability to undergo additional testing and/or surgery to 58.8% of encounters discussing false-positive rates of LCS. Among the 243 SDM encounters, 21.8% had documentation of all CMS-required components, with documentation of all components occurring in 7.8% of individuals who were currently smoking and 34.6% of individuals who formerly smoked.
Table 3:
Frequency of CMS-Designated Counseling and SDM visit Components Among Individuals with Lung Cancer Screening and Documentation of SDM in the Electronic Health Record
| CMS-designated counseling and SDM visit components4 |
Among 243 individuals screened for lung cancer with SDM documented in EHR, were these components mentioned? |
|
|---|---|---|
| Yes | No | |
| N (%) | N (%) | |
| Decision aid used | 135 (55.6) | 108 (44.4) |
| Mention of benefits and harms of LCS | 198 (81.5) | 45 (18.5) |
| Follow-up diagnostic testing | 138 (56.8) | 105 (43.2) |
| Overdiagnosis | 129 (53.1) | 114 (46.9) |
| False-positive rate | 143 (58.8) | 100 (41.2) |
| Radiation exposure | 134 (55.1) | 109 (44.9) |
| Counseling on adherence to annual screening | 129 (53.1) | 114 (46.9) |
| Impact of comorbidities | 111 (45.7) | 132 (54.3) |
| Discussion of an individual’s willingness or ability to undergo diagnosis and treatment | 107 (44.0) | 136 (56.0) |
| Maintaining/beginning cigarette abstinence | 112 (46.1) | 131 (53.9) |
| Among individuals who smoked (n=116) | ||
| Smoking cessation counseling provided by LCS ordering clinician | 86 (74.1) | 30 (25.9) |
| Smoking cessation program referral | 34 (29.3) | 82 (70.7) |
| Smoking cessation counseling or referral to program | 87 (75.0) | 29 (25.0) |
| All of the CMS components (n=243) | 53 (21.8) | 190 (78.2) |
| Currently smoking at time of visit (n=116) | 9 (7.8) | 107 (92.2) |
| Formerly smoked at time of visit (n=127) | 44 (34.6) | 83 (65.3) |
SDM = shared decision-making; CMS = Center for Medicare and Medicaid Services; EHR = electronic health record; LCS = lung cancer screening.
Individual Self-Report of SDM
Among the subset of 415 individuals who answered the HHQ item about SDM, the majority (71.1%; 295/415) reported that their clinician discussed the benefits and harms of LCS (Table 4). Among the 295 encounters with individual self-report of SDM, 47.5% also had EHR-documented SDM. Of the 120 individuals without self-reported SDM, 23.3% had EHR documentation of SDM. There was little agreement between EHR-documented SDM occurrence and individual self-report of SDM (kappa=0.1836). The agreement between SDM in the EHR and self-report was higher in individuals referred for screening by pulmonology clinicians. Among those who self-reported SDM occurring, documentation in the EHR was 78.8% for those referred by pulmonology, 50.5% for those referred by internal medicine/general medicine, and 21.4% for those referred by family medicine/primary care.
Table 4:
Agreement of Self-Report of Shared Decision-Making (SDM) with Electronic Health Record (EHR)-Documented SDM overall, by clinician specialty
| Self-Report of SDM | EHR Documentation of SDM |
Kappa | ||
|---|---|---|---|---|
| Yes N (%) a |
No N (%) a |
|||
| All b | Yes (n=295) | 140 (47.5) | 155 (52.5) | 0.1836 |
| No (n=120) | 28 (23.3) | 92 (76.7) | ||
| Clinician Specialty | ||||
| Pulmonology | Yes (n=80) | 63 (78.8) | 17 (21.3) | 0.2906 |
| No (n=13) | 5 (38.5) | 8 (61.5) | ||
| Internal Medicine/General Medicine | Yes (n=105) | 53 (50.5) | 52 (49.5) | 0.1756 |
| No (n=56) | 17 (30.4) | 39 (69.6) | ||
| Family Medicine/Primary Care | Yes (n=84) | 18 (21.4) | 66 (78.6) | 0.0508 |
| No (n=36) | 5 (13.9) | 31 (86.1) | ||
Row percentages.
All includes pulmonology, internal medicine/general medicine, family medicine/primary care, and other; Other not shown due to small numbers.
Discussion
Among a population of individuals receiving LCS in academic and community settings, less than half had EHR documentation of SDM, while over 70% self-reported SDM. The higher self-reported SDM rates may suggest that SDM discussions are under-documented in the EHR. When SDM was documented, just over half of EHR notes indicated use of a decision aid, and about one-fifth of EHR notes included all CMS-required components.
While robust data regarding SDM frequency in clinical LCS is lacking, our results suggest higher rates of SDM than previously reported. For example, Goodwin et al. reported SDM occurring in 9% of Medicare beneficiaries receiving LCS in 2016.11 In our study, rates of EHR-documented SDM were 41.9%, and self-reported SDM rates were 71.1%. Despite the higher rates observed in our study, our results corroborate that of other studies implying that SDM for LCS is variably implemented and potentially discordant from the intent of the CMS and USPSTF guidelines.7,9,11,13
Adjusted analysis demonstrated that SDM documentation was more frequent among pulmonologists than other specialty clinicians, among individuals who were obese/overweight and those who were currently smoking. The former observation may be due to increased knowledge of LCS and mortality benefit, greater familiarity with lung nodule management, and early adoption of LCS guidelines by thoracic medical societies, including the American Thoracic Society, American College of Chest Physicians, and the Society of Thoracic Radiology.17,18 In contrast, the American Academy of Family Physicians only recently endorsed LCS in March 2021.19,20 The higher SDM rates observed among pulmonologists may also be related to their participation in centralized screening programs, which may facilitate SDM and also offer advantages in patient knowledge and screening adherence.6,21,22 Regarding the association between EHR-documented SDM and overweight/obese individuals, it is plausible that given the association of a high BMI with other comorbidities, clinicians considering LCS for these individuals are more likely to have SDM conversations. The finding that SDM was more frequent in currently smoking individuals may be driven by the high rates of smoking cessation counseling, a CMS-required component of SDM. This is supported by the observation that the majority of currently smoking individuals in our cohort (75.0%) received smoking cessation counseling and/or referral. It should also be noted that although SDM is required for Medicare coverage of LCS4, multivariate modeling demonstrated no difference in SDM between insurance types.
Our results suggest that the CMS-required components of SDM are variably documented in the EHR, with approximately 20% of SDM encounters documenting all components.4 The majority (81.5%) documented the occurrence of a discussion regarding LCS benefits and harm. While a lower proportion of SDM visits documented specific benefits/harms (53.1-58.8%), our results are somewhat counter to prior work, which suggests that clinicians may understate harms/risks when counseling patients.13,23,24 A qualitative investigation reported that discussing harms/risks associated with LCS, particularly overdiagnosis and false positives, was virtually nil.13 Our results suggest that such discussions occur with higher frequency. Decision-making aids, which were documented in over half of the SDM encounters, may be one strategy to improve SDM quality. Prior work by Reuland et al. evaluating a decision aid video reported improved patient knowledge of LCS risks and benefits.25 It is also notable that while currently smoking individuals were more likely to have EHR-documented SDM, a higher proportion of formerly smoking individuals had documentation of all CMS-required SDM components. One potential reason for this is that more of the SDM visit time is spent towards smoking cessation counseling for currently smoking individuals. Formerly smoking individuals may not require this same counseling, affording more time towards other SDM aspects.
There was little agreement between EHR documentation of SDM and individual self-report of SDM. Namely, most individuals (approximately 70%) reported that their clinician talked with them about LCS risks and benefits. This may imply that SDM occurs at a higher frequency than documented in the EHR. Given that direct assessment of individuals’ understanding of SDM was not performed, the relatively high self-report of SDM cannot necessarily be taken to mean that individuals felt informed about LCS. A recent study demonstrated that while the majority of screened individuals reported that a clinician discussed reasons to undergo LCS, self-reported knowledge regarding LCS was poor, and a modest number of individuals experienced decisional conflict around LCS.9
Given the central role of SDM in individual screening decisions, efforts are necessary to ensure that SDM is not only occurring but occurring in a manner that is effective and informative. In this regard, radiologists can play a key role in providing education.26,27 For example, Rosenkrantz, et al. reported that subspecialty radiologist-led education sessions to the lay public incorporating simple language and representative images increased participant’s knowledge and awareness regarding imaging-based screening tests, including LCS.28 Given the radiologist’s expertise in LDCT interpretation, lung neoplasm detection, and screen-related risks, such as radiation, radiologists have a significant opportunity to help provide education prior to LCS and help in strategies towards improving individuals’ understanding of LCS results.26,27
Our study has limitations. This study relied on EHR documentation of SDM and individuals' recall of SDM. It is plausible that our estimates of SDM may not accurately capture the true frequency with which SDM conversations occurred, as some clinicians may have conducted SDM but failed to document it in the EHR. In this regard, a checklist or template in the EHR for clinician documentation of SDM components may be one solution to guide and accurately capture comprehensive SDM discussions. Additionally, individuals may over or under self-report SDM. Our study cohort includes a subset of individuals who underwent LCS and completed a one-page HHQ. While individuals who did versus did not complete the HHQ were similar with regard to age, sex, race, smoking status, COPD status, and BMI there were some differences in the groups. Compared to those who did not complete the HHQ, those who did complete the HHQ were more likely to be from urban versus rural areas, were more likely to have undergone screening in 2015/2016 versus in 2019/2020, and were more likely to have been referred for screening by internal or general medicine versus by family medicine or primary care. Finally, all sites in this investigation were from one state in the Southeastern U.S., and practice patterns may differ across the U.S.
In summary, our results suggest that EHR-documented SDM in LCS practice is sub-optimal, although rates are higher than reported by prior work. The higher self-reported SDM rate observed may suggest clinicians are under-documenting the service. Although use of a decision-making aid, a CMS requirement, was observed in approximately 55% of SDM encounters, only one-fifth of SDM conversations had all of the CMS-required components. As SDM is required by CMS and recommended by the USPSTF, the varying use of SDM among individuals screened for lung cancer underscores the need for interventions that facilitate consistent and high-quality SDM use and documentation.
Take-Home Points.
EHR-documented SDM in clinical LCS practice is sub-optimal, although rates are higher than reported previously.
Self-reported SDM conversations were reported at higher rates suggesting clinicians may be under-documenting the service.
Although use of a decision-making aid, a CMS requirement, was observed in approximately 55% of SDM encounters, only one-fifth of SDM conversations documented all CMS-required components.
Pulmonary clinicians were more likely to document SDM occurrence in the EHR than clinicians in other specialties.
Funding
National Institutes of Health R01CA212014 and R01CA251686
Footnotes
Conflict of Interest
TT, DDD, PP, LL, LMH: The authors declare no conflict of interest.
MPR: In 2019, Advisory Board member, Biodesix and bioAffinity technologies, research consultant, Johnson & Johnson.
Data Statement
The authors declares that they had full access to all of the data in this study and the authors take complete responsibility for the integrity of the data and the accuracy of the data analysis.
References
- 1.The National Lung Screening Trial Research Team. Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening. N Engl J Med. 2011;365(5):395–409. doi: 10.1056/NEJMoa1102873 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Moyer VA. Screening for Lung Cancer: U.S. Preventive Services Task Force Recommendation Statement. Ann Intern Med. 2014;160(5):330–338. doi: 10.7326/M13-2771 [DOI] [PubMed] [Google Scholar]
- 3.US Preventive Services Task Force, Krist AH, Davidson KW, et al. Screening for Lung Cancer: US Preventive Services Task Force Recommendation Statement. JAMA. 2021;325(10):962. doi: 10.1001/jama.2021.1117 [DOI] [PubMed] [Google Scholar]
- 4.Centers for Medicare & Medicaid Services (CMS). Decision Memo for Screening for Lung Cancer with Low Dose Computed Tomography (LDCT) (CAG-00439N). Published online February 5, 2015. https://www.cms.gov/medicare-coverage-database/details/nca-decision-memo.aspx?NCAId=274 [Google Scholar]
- 5.Seiler N, Malcarney MB, Horton K, Dafflitto S. Coverage of Clinical Preventive Services Under the Affordable Care Act: From Law to Access. Public Health Rep. 2014;129(6):526–532. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Alishahi Tabriz A, Neslund-Dudas C, Turner K, Rivera MP, Reuland DS, Elston Lafata J. How Health-Care Organizations Implement Shared Decision-making When It Is Required for Reimbursement: The Case of Lung Cancer Screening. Chest. 2021;159(1):413–425. doi: 10.1016/j.chest.2020.07.078 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lowenstein LM, Deyter GMR, Nishi S, Wang T, Volk RJ. Shared decision-making conversations and smoking cessation interventions: critical components of low-dose CT lung cancer screening programs. Transl Lung Cancer Res. 2018;7(3):254–271. doi: 10.21037/tlcr.2018.05.10 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.CT Lung Cancer Screening Shared Decision Making Visit Requirements. American College of Radiology. Accessed April 5, 2021. https://www.acr.org/-/media/ACR/NOINDEX/Advocacy/CT-LCS-shared-decision-making_fillable.pdf [Google Scholar]
- 9.Nishi SPE, Lowenstein LM, Mendoza TR, et al. Shared Decision-Making for Lung Cancer Screening. Chest. Published online February 2021:S001236922100115X. doi: 10.1016/j.chest.2021.01.041 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hoffman RM, Reuland DS, Volk RJ. The Centers for Medicare & Medicaid Services Requirement for Shared Decision-making for Lung Cancer Screening. JAMA. 2021;325(10):933. doi: 10.1001/jama.2021.1817 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Goodwin JS, Nishi S, Zhou J, Kuo YF. Use of the Shared Decision-Making Visit for Lung Cancer Screening Among Medicare Enrollees. JAMA Intern Med. 2019;179(5):716–718. doi: 10.1001/jamainternmed.2018.6405 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.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: 10.1164/rccm.201508-1671ST [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Brenner AT, Malo TL, Margolis M, et al. Evaluating Shared Decision Making for Lung Cancer Screening. JAMA Intern Med. 2018;178(10):1311–1316. doi: 10.1001/jamainternmed.2018.3054 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Fukunaga MI, Halligan K, Kodela J, et al. Tools to Promote Shared Decision-Making in Lung Cancer Screening Using Low-Dose CT Scanning. Chest. 2020;158(6):2646–2657. doi: 10.1016/j.chest.2020.05.610 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.North Carolina Lung Screening Registry. North Carolina Lung Screening Registry. Accessed July 7, 2021. https://nclsr.unc.edu/ [Google Scholar]
- 16.USDA ERS - Rural-Urban Continuum Codes. Accessed June 7, 2021. https://www.ers.usda.gov/data-products/rural-urban-continuum-codes.aspx [Google Scholar]
- 17.Kazerooni EA, Austin JHM, Black WC, et al. ACR–STR Practice Parameter for the Performance and Reporting of Lung Cancer Screening Thoracic Computed Tomography (CT): 2014 (Resolution 4)*. J Thorac Imaging. 2014;29(5):310–316. doi: 10.1097/RTI.0000000000000097 [DOI] [PubMed] [Google Scholar]
- 18.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: 10.1164/rccm.201508-1671ST [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lung Cancer - Clinical Preventive Service Recommendations. Accessed June 17, 2021. https://www.aafp.org/family-physician/patient-care/clinical-recommendations/all-clinical-recommendations/lung-cancer.html [Google Scholar]
- 20.Armstrong C Lung Cancer Screening Recommendations from the ACCP. Am Fam Physician. 2018;98(11):688–689. [PubMed] [Google Scholar]
- 21.Mazzone PJ, Tenenbaum A, Seeley M, et al. Impact of a Lung Cancer Screening Counseling and Shared Decision-Making Visit. Chest. 2017;151(3):572–578. doi: 10.1016/j.chest.2016.10.027 [DOI] [PubMed] [Google Scholar]
- 22.Sakoda LC, Rivera MP, Zhang J, et al. Patterns and Factors Associated With Adherence to Lung Cancer Screening in Diverse Practice Settings. JAMA Netw Open. 2021;4(4):e218559–e218559. doi: 10.1001/jamanetworkopen.2021.8559 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hoffmann TC, Del Mar C. Clinicians’ Expectations of the Benefits and Harms of Treatments, Screening, and Tests: A Systematic Review. JAMA Intern Med. 2017;177(3):407–419. doi: 10.1001/jamainternmed.2016.8254 [DOI] [PubMed] [Google Scholar]
- 24.Clark SD, Reuland DS, Enyioha C, Jonas DE. Assessment of Lung Cancer Screening Program Websites. JAMA Intern Med. 2020;180(6):824. doi: 10.1001/jamainternmed.2020.0111 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Reuland DS, Cubillos L, Brenner AT, Harris RP, Minish B, Pignone MP. A pre-post study testing a lung cancer screening decision aid in primary care. BMC Med Inform Decis Mak. 2018;18(1):5. doi: 10.1186/s12911-018-0582-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Green DB, Pua BB, Crawford CB, et al. Screening for Lung Cancer: Communicating With Patients. AJR Am J Roentgenol. 2018;210(3):497–502. doi: 10.2214/AJR.17.18836 [DOI] [PubMed] [Google Scholar]
- 27.Snoeckx A, Franck C, Silva M, Prokop M, Schaefer-Prokop C, Revel MP. The radiologist’s role in lung cancer screening. Transl Lung Cancer Res. 2021;10(5):2356–2367. doi: 10.21037/tlcr-20-924 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Rosenkrantz AB, Mason D, Kunzler NM, Lee J. The Radiologist as Direct Public Educator: Impact of Sessions Demystifying Select Cancer Screening Imaging Examinations. J Am Coll Radiol JACR. 2016;13(2 Suppl):R38–42. doi: 10.1016/j.jacr.2015.12.012 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The authors declares that they had full access to all of the data in this study and the authors take complete responsibility for the integrity of the data and the accuracy of the data analysis.