An alternating-intervention pilot trial on the impact of an informational handout on patient-reported outcomes and follow-up after lung cancer screening

Introduction Lung cancer screening (LCS) can reduce lung cancer mortality; however, poor understanding of results may impact patient experience and follow-up. We sought to determine whether an informational handout accompanying LCS results can improve patient-reported outcomes and adherence to follow-up. Study design This was a prospective alternating intervention pilot trial of a handout to accompany LCS results delivery. Setting/Participants Patients undergoing LCS in a multisite program over a 6-month period received a mailing containing either: 1) a standardized form letter of LCS results (control) or 2) the LCS results letter and the handout (intervention). Intervention A two-sided informational handout on commonly asked questions after LCS created through iterative mixed-methods evaluation with both LCS patients and providers. Outcome measures The primary outcomes of 1)patient understanding of LCS results, 2)correct identification of next steps in screening, and 3)patient distress were measured through survey. Adherence to recommended follow-up after LCS was determined through chart review. Outcomes were compared between the intervention and control group using generalized estimating equations. Results 389 patients were eligible and enrolled with survey responses from 230 participants (59% response rate). We found no differences in understanding of results, identification of next steps in follow-up or distress but did find higher levels of knowledge and understanding on questions assessing individual components of LCS in the intervention group. Follow-up adherence was overall similar between the two arms, though was higher in the intervention group among those with positive findings (p = 0.007). Conclusions There were no differences in self-reported outcomes between the groups or overall follow-up adherence. Those receiving the intervention did report greater understanding and knowledge of key LCS components, and those with positive results had a higher rate of follow-up. This may represent a feasible component of a multi-level intervention to address knowledge and follow-up for LCS. Trial registration ClinicalTrials.gov NCT05265897.

A.6.Significance Summary: These studies illustrate the need to develop optimal ways to communicate LCS results, and to understand barriers and motivators to adherence to LCS in diverse populations.Evidence for how best to communicate results to both patients and providers is surprisingly limited, particularly as LCS is initiated in the community with the potential to impact millions of individuals.Our proposal is responsive to published ALA6 and American Thoracic Society (ATS) research priorities62.68 to advance implementation of LCS with attention to improving patient-oriented outcomes of screening.Our proposal will yield directly relevant findings that can be rapidly translated to improve clinical care of patients at risk for lung cancer.

B. INNOVATION:
This project seeks to inform and shift current clinical practice by being among the first to: • Engage patients actively in the process of LCS by having them view and give feedback on the LDCT report; this strategy is relevant and pro-active, as patient access to full radiology reports is growing.21• Evaluate patient-preferred methods to improve communication of LCS results and incorporate provider feedback;69 results will inform a framework that can become part of an ALA standardized toolkit for LCS.• Assess patient acceptability, understanding and distress associated with a patient-oriented LCS report.
• Assess barriers and facilitators to adherence to LCS.
• Incorporate patient populations that include a high proportion of minority race/ethnicity and low-income patients who have a high smoking prevalence but have been poorly represented in other studies.

C. APPROACH C.1. Investigative team:
We are a multidisciplinary team with extensive experience in qualitative and quantitative methods, and in the rigorous conduct of clinical research studies.We have expertise in pulmonary medicine and LCS (PI Crothers, co-I Kross, co-I Triplette); primary care (co-I Cole); dissemination and implementation science (co-I Cole); patient communication and psychological outcomes (Kross); developing and adapting health communication materials (collaborator Ko); and integrating smoking cessation with screening and standardization of radiology reporting (collaborator Zeliadt).A Steering Committee composed of Drs.Elmore (primary care, cancer screening) and Pipavath (thoracic radiology) along with Zeliadt and Ko will meet quarterly to provide additional input from key perspectives and assist with formative evaluations on implementation as we aim to develop this project into a future multi-center trial informed by these results.

C.2. Preliminary work:
C.2.a.Patients' attitudes regarding LCS and decision aids: A survey and focus group study.9In the 1st year (2014-15) of our initial ALA LCD Award we determined, in a low-income, racially diverse population, participants' 1) experience, preferences and reactions to web-based and paper decision aids, and 2) understanding of LCS harms and benefits.We completed 6 focus groups with 45 current or former smokers at Harborview Medical Center (HMC), an urban county hospital.Participants had a mean age of 61; 42% were non-white; 27% had not completed high school; and 50% had an annual income 5,000.Based on pre-and post-surveys, participants' understanding of LCS increased after the groups, particularly of harms including false-positives and extra testing.Five major themes emerged from qualitative analyses: participants 1) were not aware of the purpose of LCS; 2) wanted to know about the benefits and harms; 3) felt physicians need to communicate more effectively; 4) found decision aids helpful and influential for decision-making regarding LCS; and 5) wanted the discussion to be personalized and tailored.These results highlight the diversity of our population, our ability to recruit these participants, and our methodologic expertise, productivity and success of our initial ALA project that we will carry forward in this renewal as we focus on steps after the LDCT exam.C.2.b. Assessing knowledge, barriers, and facilitators to LCS among primary care and pulmonary providers.10To inform implementation of LCS into practice, in the 2nd year (2015-16) of our ALA LCD Award we administered a web-based survey to providers in primary care and pulmonary clinics in the University of Washington (UW) hospital system to assess beliefs, knowledge, current practices, barriers, and facilitators to LCS.Of 196 respondents, 65% were attending physicians and 71% were primary care providers.While the majority (78%) felt that LCS was effective for preventing death from lung cancer, provider knowledge to estimate lung cancer risk and follow-up of nodules was poor.Overall, 50% reported they "don't know" or were "unsure" whether Lung-RADS is important in follow-up of nodules, and 68% needed more information on follow-up recommendations.Providers also identified other barriers (including adherence) at the patient, provider and system levels.These data highlight deficiencies particularly surrounding management of nodules.These data also further underscore the gaps in communication with patients if providers themselves are not clear on recommendations after LDCT.To complement ongoing provider education initiatives, in this renewal we focus on understanding the needs and preferences of patients to inform communication of LCS results.C.2.c.Adherence to follow-up recommendations after LDCT screening in a real-world setting is low.All patients who undergo LCS in the UW system are tracked using Primordial, a licensed software program to manage LCS registries (www.PrimordialDesigns.com).We analyzed our current LCS database to determine patient adherence with follow-up; 438 patients who underwent initial screening from 1/2012-6/2016 were included.Adherence to recommended first follow-up was substantially lower than in the NLST.Overall followup after initial LDCT was 58% and varied by Lung-RADS score 1-4 (44%, 67%, 75%, 92% respectively).Past barriers to adherence that are likely improved currently include: variability in radiologist recommendations before publication of Lung-RADS; lack of insurance coverage prior to guideline and CMS recommendations;7.8 and lack of a robust system to automatically track patients.However, UW began to use Primordial prospectively in 9/2016, and hired a nurse coordinator for the program in 1/2017.Thus, while we anticipate that adherence to follow-up in Aim 1 will be higher as we analyze results of more recent LCS, adherence will likely still be less than 95% seen in the NLST.The work proposed in this project will allow us to investigate factors that can interfere with adherence in real clinical settings, a critical component to successful LCS.C.2.d.Primary care patient use of web-portals in diverse primary care practices and access to radiology reports.Dr. Cole and colleagues conducted a survey of more than 800 primary care patients across five primary care clinics to assess patient reported use of web-portals.7°Approximately 39% of patients reported awareness of web-portals for patient-provider communication, and of those 67% reported regular use of webportals.Dr. Elmore's team led the largest cross-sectional study of patient access to radiology reports to date, involving -130,000 patients in a major health system and found high patient interest in reading radiology reports online, with 51% of all patients with web-portal access viewing available radiology reports.21This high level of interest among patients is corroborated by earlier, smaller studies regarding patient preferences.Among surveyed patients using web-portals, 79% of patients preferred new portal-based methods of receiving results over historical methods, including directly from the referring physician.2°Similarly, focus groups noted that most patients were dissatisfied with traditional reporting of radiology results, citing delays leading to undue anxiety and distress and little detail when radiologic findings are relayed verbally by a referring physician!' gpiFollowmupai assess patient-specific concerns and preferences to development of individualized, patient-centered approaches to communication, including use of the actual LDCT report.We will test the individualized report and assess acceptability, understanding of and distress with LCS results in Aim 3.With input from the Steering Committee, we will evaluate the outer and inner setting, individual health professionals, and the process of implementation in this pilot trial to prepare us for a larger, multicenter trial to improve implementation of LCS.
C.4.a.Study setting for all Aims: Participants who had LCS will be recruited from primary care and chest clinics affiliated with the University of Washington Medical Center (UWMC), Harborview Medical Center (HMC), and Seattle Cancer Care Alliance (SCCA), [collectively referred to as "UW system].We have extensive experience recruiting vulnerable patients for similar studies, including our prior ALA project.9.6°' 73 ' 74 We anticipate recruiting -50°/0 of our participants from HMC, the county hospital, where Drs.Crothers, Cole and Kross are providers.Sites are further described in "Resources," and human subjects in "Assurances."C.4.b. Patient participants for all Aims: Inclusion criteria: 1) Has undergone LCS, and 2) Able to give informed consent.Exclusion criteria: 1) Cognitive or language limitations (i.e., expressive language limitations, non-English speaking); or 2) Known malignancy other than non-melanoma skin cancer.
Aim 1. Determine patient-level factors associated with poor adherence to follow-up (repeat CT, other testing) in patients who have previously undergone LCS, using survey methods and EHR data.C.5.a.Rationale: Survey-based study design allows collection of self-reported information from patients to determine patient-level characteristics associated with barriers and facilitators to adherence to follow-up.We will identify LCS patients using Primordial software and collect survey and EHR data to assess whether poor patient-reported understanding or other factors such as smoking, low SES, minority race, or other comorbidities are associated with poor adherence.Additional data will be retrieved from the EHR including actual radiology reports for LDCT results, recommended follow-up, and provider ordered testing.By recruiting patients who have already had LCS, we can efficiently obtain data for analyses in this 2-year award.
Focus in ALA LCD Renewal screening in diverse populations C.5.b. Overview of study design and procedures: This is a cross-sectional study.Patients identified from the Primordial LCS registry will be mailed an informed consent and survey to complete.They can choose to either: 1) Complete the paper survey and return it via mail using an enclosed stamped envelope or; 2) Complete the survey via secure on-line access using an identical web-based survey.We will use standard methods to enhance participation rates including a $10 gift card incentive, reminder/thank-you postcards sent 2 weeks after the initial contact, and second packets with surveys sent to non-responders after 4 weeks.79

C.5.c. Data Collection: Participants will complete surveys 1-2 months after their planned follow-up per the
Lung-RADS score from their index LDCT (e.g.12 months after Lung-RADS 1 or 2; 6 months for Lung-RADS 3

C.5.e. Analysis:
To determine whether patient adherence (yes/no) to follow-up study (as ordered by the provider) after LDCT differs by patient characteristics and understanding of LDCT results, chi-squared tests will be used to compare categorical variables (such as sex, race/ethnicity, insurance, income, smoking status), ttests for continuous approximately normally distributed variables (such as age) and Wilcoxon rank-sum tests for non-normally distributed variables (such as pack-years of smoking).Understanding of LDCT (e.g.nodule yes/no) and follow up plan will be dichotomized as good vs. poor.We will evaluate patient's self-assessments of being informed as well as understanding of findings defined by comparing patient-report of results to radiologist report in the EHR.Patient characteristics associated at p<0.1 will be assessed in multivariable logistic regression models, adjusting for potential covariates and confounders to determine independent associations.Collinearity will be assessed using Variance Inflation Factors.Potential effect modification (e.g.education, quality of communication with provider) will be assessed by interaction terms and stratification.C.5.f.Sample size: We will survey patients who have been screened in the 6-15 months prior to the start of the award, and will continue to administer surveys over the first 18 months.In the past 2 years, referrals have grown substantially with 505 patients initiating LCS in 2016 and 2017.We anticipate mailings to -500 patients, a 40-50% response rate,42' 43 a range in the primary exposure of 30-60% (poor understanding of LCS results42. 43)and a 20-40% range in non-adherence from our preliminary data.Conservatively estimating 200 respondents, we have 80% power to detect an odds ratio of 2.25.This sample size is comparable or larger than studies of incidentally detected nodules that found significant associations.11,42,43 Aim 2. Explore patient understanding and preferences regarding the methods and information that providers should include in order to communicate the results of LCS in a focus-group based study.C.6.a.Rationale: There are limited published studies (primarily on incidental nodules".49) that consider a patient's: 1) understanding of LCS results, and 2) preferences for methods to communicate LCS results.A qualitative approach is particularly appropriate for identifying and describing beliefs and issues that have not been previously examined, and avoids assumptions that could prove to be incorrect if we were to only administer surveys in which items are necessarily constrained and feedback is limited.Patients will form the initial groups, but final groups will include primary care providers to engage in dialogue and obtain feedback.69C.6.b. Overview of study design and procedures: This will be a qualitative study using focus group discussions.A purposive sample of patients who have undergone LCS in the last year will be recruited, making every effort to include patients of diverse race/ethnicity, sex, income, and spanning the eligible age range for LCS.We will pay particular attention to recruiting women as well as men.We will use a combined approach as in our prior study for recruitment,9 and we will screen using Primordial.Primary care providers in the UW system will be recruited via email or in person to partake in final groups with patients structured as a "circle within a circle" per prior work.69After explaining the study, informed signed consent will be obtained from those who will participate.Modest incentives will be offered (e.g.$25 grocery store card, bus voucher, refreshments).C.6.c.Sample Size: Based on prior experiences, we anticipate that we will reach thematic saturation, the point at which no new information is gained,69 by the conclusion of 6-7 focus groups, each with 7-10 participants.The last 2-3 focus groups will include both patients and primary care providers together.C.6.d.Data Collection: Focus Groups will be led by the moderator (G Keppel) and note-taker (S Shahrir), and will follow a guide with structured but open-ended questions (see Appendix); they will follow the participants' lead and allow relevant topics not specified to be addressed.We will first pilot the guide9 and a template of the patient-oriented report, based on Lung-RADS and prior studies,49.49 63 with individual patients.Then, focus groups will start with patients individually viewing de-identified actual LDCT radiology reports, examples of patient letters generated by Primordial for Lung-RADS categories, and completing brief surveys before a group discussion.Surveys will include demographics, education, health literacy, understanding and distress associated with each LDCT report and Primordial letter.As a group, we will assess knowledge and understanding of the materials.We will also discuss other Lung-RADS "Category S" findings such as emphysema; these findings have largely been over-looked but can cause distress and have actionable health consequences.6'81 We will ask participants how they would want to be informed of results,48' 49' 86 review our patient-oriented report for LCS results, and discuss what other patient-individualized information should be included beyond radiology findings, such as individual risk of lung cancer given findings and the "What is a Nodule" handout (Table 2).89With Dr. Zeliadt's expertise,99 we will present smoking cessation messages to identify ways that avoid stigma and encourage adherence.91.92 Participants will be asked to discuss the format and content they prefer, why, and suggestions for improvement.We will also elicit barriers and motivators to adherence.We will iteratively refine the report from group to group.The last 2-3 groups will include patients and providers.69Groups will last -70-90 minutes, will be audio-taped with two digital recorders, de-identified and transcribed verbatim by a professional service.

C.6.e. Analysis:
To analyze the focus group data, we will use a modified grounded theory approach93 to describe observations and to construct theory based on the described observations.94.95 Steps include: 1) open coding; 2) axial coding; 3) process coding; and 4) selective coding.Findings are used to create a framework in which to understand the factors, context and relationships that enhance or impede understanding and adherence; they will also be used to understand the features of LCS reports that participants identify as most useful, accessible and understandable.Coding will be iterative, with transcripts analyzed after each group so that suggestions or concerns raised by one group can be used to refine the report; this will be further vetted in subsequent groups.The analytic team will include Drs.Crothers, Kross, Cole, and the two interviewers.Two focus group transcripts will be reviewed by the whole analytic team who will develop the coding manual; pairs of analysts will then review transcripts as they are completed with the goal of achieving consensus on the use of codes and contributing to the refinement of the individualized patient-oriented report of LCS results.Aim 3. Compare patient understanding, distress and intention to adhere to follow-up after receipt of a patient-oriented report of LCS results vs. usual care notification in a pilot randomized controlled trial.C.7.a.Rationale: We will assess patient understanding, acceptability, and distress with receipt of a patientoriented report of LCS findings, developed based on Aims 1 and 2, in comparison to usual care.Given the twoyear duration of this award, we will assess patient intention to adhere to one-year routine follow-up but actual adherence to shorter-term follow-up.This trial will provide preliminary data in support of a larger clinical trial.

C.7.b. Overview of design:
This will be a randomized, controlled pilot study of patients undergoing LCS in the UW system.Patients randomized to the intervention will receive a patient-oriented report based on individual characteristics and results that will replace usual care notification; we will compare this to usual care notification.All participants will receive either notification by mail; surveys will be mailed one week later.

C.7.c. Randomization, intervention and data collection:
Participants will be randomized by stratified permuted block randomization with varying block size to receipt of the patient-oriented report of LCS results or usual care notification after LDCT.While there is currently no standard for notification of LCS in the literature, our program has adopted a consistent approach, which includes mailing letters to patients using the templates from Primordial for each Lung-RADS category.Given high uptake of Primordial at other centers across the US, this standard likely reflects usual care across many LCS programs.Providers are expected to telephone patients with LungRADS 3-4 results, but not necessarily LungRADS 1-2.Because those who are initiating LCS may differ from those returning, participants in each arm will be stratified by whether they are undergoing initial vs. follow-up LCS, as well as by clinical site at UW.All patients who have online access to their own EHR may continue to access their data.Providers will be blinded to treatment group.Surveys and consents will be Table 2. Discussion points on LCS results for focus groups Lung-RADS results: "1.Negative," to "4.Suspicious"5b Lung-RADS Category S or "Other" findings (emphysema, coronary calcification) l "What is a Nodule?" handout' s' Preferences for including and communicating individual lung cancer risk84-66 Smoking cessation messages mailed to patients one week after their LCS report to obtain timely feedback on understanding and reactions to materials received.We will provide paper and secure internet-based options for the survey as in Aim 1.

C.7.d. Outcomes, predictors and covariates:
Primary outcomes: 1) Correct understanding of LDCT (nodules, and category S findings), and needed follow-up (evaluated by concordance between radiologist recommendation and patient statements); 2) Distress, measured using the IES scale:76.77and 3) Acceptability to patients (e.g."The information was useful"; "The report explained things in terms I could understand.").Secondary outcomes will include: 4) Reported intention and actual adherence to follow-up; 5) Motivation to quit smoking or to remain smoke-free.Other predictors and covariates will be obtained from patients as in Table 1.

C.7.e. Analysis:
We will generate descriptive statistics (e.g.mean, standard deviation [SD], median, percent).We will consider differences between groups, adjusting for strata, in generalized linear models for continuous normally distributed variables, in non-parametric rank-based test procedures, and in logistic regression models or stratified contingency tables for binary or categorical variables.We can explore if outcomes differ by site and patient characteristics, including race, SES, and patient-rated quality of provider communication.While we present power for outcomes, a main purpose of a pilot is to test study loqistics96 and gain critical insights into recruitment, randomization, data collection forms, participation rates and barriers/facilitators to implementation.

C.7.f. Sample Size:
We will obtain IRB permission to randomize all patients undergoing LCS at UW to this quality improvement intervention (projected -434 LDCT in 2018, -1/2 with initial LCS).Assuming 35-40% response rate to surveys, this yields an adequate sample for a pilot study.Using a mean distress (IES) score of 12.9 (SD 15.1)," a minimal important difference (MID) of 7 (SD/2),97 and a balanced design, we have adequate power (80%) with 74 participants in each arm to detect a difference of points with alpha of.05.If using categorical data for the outcomes of distress or of poor understanding of results, 80 participants in each group is required, assuming 45% of controls have mild or greater distress (IES8)44 or poor understandin compared to 20-25% of the intervention group, using Pearson's chi-square test for statistical independence.

C.8. Overall Limitations and Alternative Strategies:
A critical piece of our work is to ensure that results are relevant across a broad range of patient backgrounds.Though limited to one geographic region, we do not foresee difficulty in achieving our target sample size and are assured of having racially, ethnically, and socioeconomically diverse participants as in our prior work.9A potential concern is low response rate from under-served populations; however, we have a large number of patients and a successful track record of recruitment.If we have poor response to surveys, we will obtain IRB permission to call potential participants, and could complete surveys over the telephone with research staff (blinded to study arm in Aim 3).If focus group enrollment is low, we can also recruit participants for one-on-one interviews.
Finally, although we have limited our sample to English speakers, this is the primary language for most of our patients, including at HMC.

C.9. Strengths, Deliverables and Future
Directions: We will provide the ALA with data to augment their toolkit with concrete and creative recommendations to improve the implementation of LCS.We will build upon what we have learned in our first ALA LCD to bring a full perspective to improve communication with patients, beginning with a decision to undergo LCS before LDCT is performed to after the exam is completed.Results will be disseminated in the form of presentations at conferences and manuscripts.Our work will establish the basis for future trials, including testing implementation of an individualized LCS report to patients as well as providers.Finally, this ALA LCD would allow Dr. Crothers to continue the mentoring begun during her initial LCD Award, including of Dr. Triplette (now faculty) and Dr. Perrin Romain, who will continue to participate in the scientific investigations proposed.With this renewal, the ALA will also be facilitating and supporting the career development of junior investigators interested in improving lung cancer outcomes.

Introduction-purpose and ground rules
My name is and I will be leading this session to talk with you about lung cancer screening.
Thank you so much for being here today.We have received a grant from the American Lung Association to find out good ways that doctors can use in talking with their patients about the results of lung cancer screening.You are all here today because have recently undergone a low-dose CT for lung cancer screening, and we are interested in your opinions on how doctors can best communicate the results of these CT tests to their patients.

INDIVIDUAL INTRODUCTIONS
Before we get started, has everyone signed the consent to be audiotaped?This tape is used to summarize our discussion.No one will listen to it but members of the research team.
A few ground rules: We are not looking for you to all agree; in fact, we don't expect you to agree with each other.The screening test for lung cancer is an imaging study called a low-dose chest CT scan.You all recently went through one of these tests.We want to learn ways to communicate better with patients about how to talk about the results so that they are easy to understand.We expect there will be a range of experiences and we expect to learn from the variety --and tailor our program to address what patients worry about and what they need to know.
If any questions come up about your own recommendations, please discuss your questions with your doctor or healthcare provider --Please do not make any screening or treatment decision based on our group discussion.
Just a few more rules: If you tend to talk a lot, I hope you won't mind if I ask you to stop so that we can hear from everyone.If you are shy, I hope you'll speak up and participate.
We encourage you to share your opinions with the group but please avoid side conversations while others are speaking.We are taping this group so we can hear from everyone, and when there are side conversations it interferes with the clarity of the taping.
A reminder: please turn off your cell phone.
Are there any questions before we get started?

1.
Everyone here has had a lung cancer screening CT exam.What is the ONE thing you remember most about your CT exam?

2.
Most of you have received the results of your LDCT.What was it like waiting to hear the results?What is the ONE thing you remember most about getting the results of your CT? What was it like hearing the results?Are you happy with your decision to be screened?Why, or why not?

6.
When your doctor went over the results of your LDCT with you, what parts were confusing to you?How were those parts explained to you?What parts were distressing to you?What about their communication didn't work well?What could your doctor have done differently to increase your understanding of the results?What could your doctor have done to make you less worried?

7.
Pass out "What is a nodule" information sheet.I'm going to show you some of the issues to consider about the possible results of lung cancer screening using pamphlets that your doctor could show you.We want you to tell us your thoughts about whether this helps you understand what your possible test results would be.[Provide each participant with a copy of the pamphlets] Based on our discussion thus far and seeing this pamphlet, do you think you understand the possible results of a lung cancer screening test?
a. What could the results of a LDCT be? b.What would the results mean to the patient?c.Did the pamphlet help you?d.How do you think the pamphlet could be improved?e. Do you want to know how frequently each type of result can occur, as illustrated in the picture, or just what the possible results could be? f.Would you want your doctor's office to use a pamphlet like this when explaining your results?What other methods of explaining could be used?

8.
After hearing more about the results of lung cancer screening, have you decided to undergo a repeat LDCT in the future when recommended by your doctor?-Why/why not?Thank you so much for participating.This has been a really useful experience and you have provided a lot of very valuable information.Do you have any final comments?
IRO RECD OCT 17 2018   Community-based Outcomes and Adherence in Lung cancer Screening (COALS) November 1, 2017

Introduction
The National Lung Screening Trial (NLST), published in 2011, demonstrated a 20% lung cancer mortality risk reduction from annual low-dose computed tomography (LDCT) compared to chest radiography in a defined cohort of high-risk smokers.'On the basis of this trial, the United States Preventive Services Task Force (USPSTF) now recommends annual lung cancer screening (LCS) with LDCT for patients who are age 55-80, are either current smokers or former smokers who quit 15 years prior, and who have a 30 pack-year smoking history. 2 However, the literature of outcomes from and adherence to screening in usual clinical settings is limited.Both these factors may affect the risks, benefits and effectiveness of screening.The overall goal of this study is to examine the patient population and outcomes of those undergoing screening at a diverse tertiary care medical center with wide community referral catchment area, examine issues of adherence and patient-related factors associated with non-adherence, and build an ongoing cohort for LCS research.

Relevant Literature and Data
Previous studies of lung cancer screening are limited, given the novelty of the preventive health service.The largest population to undergo screening is the NLST cohort; however, this study population is significantly healthier, contained fewer active smokers, and had higher adherence (95%) due to the trial setting than patients eligible and undergoing screening in the United States: 1 '3.4The Veterans Affairs (VA) lung cancer screening demonstration project is the largest study of screening in a real-world setting.This study included 8 participating VA medical centers and two years of screening data, and demonstrated vastly different outcomes from NLST, including a higher prevalence of false-positive findings than would be expected.'However, this study may have limited validity outside of the VA setting as the majority of patients were men and many had a higher burden of comorbid illness.
Presently, the implementation of screening is suboptimal, and screening rates remain low.' Several studies, including our own prior work, have focused on assessing patient-, provider-and system-level factors to understand (and ultimately improve), utilization and referra1.5.7-11 The majority of these studies have found that the current state of patient and provider knowledge is likely limiting referral of eligible patients, and that there are key barriers and facilitators that can be addressed to improve access and ease of referral.On the other hand, there are few studies of adherence in patients referred to screening.Given both the high prevalence of abnormal screening studies and the requirement for a minimum of annual screening, high adherence is likely a necessary component to achieve benefits from screening.Adherence may be particularly important for effective smoking cessation coupled with screening, as smoking cessation rates may improve with repeated reinforcement.'Further evaluation of screening would benefit from an assessment of screening participants, outcomes and key factors related to adherence in representative clinical settings.'We currently have 716 patients enrolled and tracked within our screening program, and the program is growing and expanding to multiple sites across the UW/SCCA sysiffendlikikri ,AMk% represents an opportunity to build a research database of our patients, and Fol5576117' DEC 1 3 2013 Document Released Date attractive database for future collaboration across systems to build a multicenter prospective cohort.

Compliance Statement
This study will be conducted with Fred Hutchinson Cancer Research Center (FHCRC) policies and procedures and all applicable Federal and state laws and regulations including 45 CFR 46 and the HIPPA Privacy Rule.Any episode of noncompliance will be documented.Collecting, recording, and reporting of data will be accurate and will ensure the privacy, health and welfare of research subjects during and after the study.

Aims
Aim 1. Examine the demographic features of patients and referring clinics for lung cancer screening throughout the UW system.Aim 2. Examine key outcomes from lung cancer screening in the UW system compared to outcomes from the NLST and the VA demonstration project.Aim 3. Examine follow-up after initial screening.Compare adherence to follow-up during period before and after tracking, by Lung-RADS category, by smoking status and by patient demographic features.Aim 4. Develop an ongoing research database of patients undergoing LCS within the UW system for future investigation.

General Schema of Study Design
The study will include: 1) a retrospective cohort study of electronic medical records for patients who have undergone LCS within the UW system (Aims 1-3); and 2) development of an ongoing prospectively enrolling cohort of subjects undergoing LCS and follow-up from LCS within the UW system.

Date Range of Study
Subjects will be patients who underwent LCS at any UW site from 2012-present (duration of LCS program).These patients are currently identified in the SCCA Primordial tracking software.Ongoing enrollment will occur via review of the Primordial tracking software for new examinations every 3 months.

Total Number of Study Sites/Total Number of Subjects Projected
The study will be conducted at the FHCRC but will include patients undergoing screening throughout the UW system.Patients are referred for screening through a variety of providers including but not limited to primary care providers and pulmonologists.They undergo CT screening at 4 sites within the system: 2 UW Neighborhood clinic (UWNC) sites, the Seattle Cancer Care Alliance (SCCA) and the UW Medical Center (UWMC) with plans to expand to

Figure 1 .
Figure 1.Framework: Improving implementation of lung cancer

Aim 1 Month 1 - 2 : 2 Months 1 - 5 : 3 Months 6 - 8 :
Finalize surveys & obtain IRB approvals for all Aims; use Primordial to identify patients; Months 2-18: administer surveys to patients who have previously undergone LCS; Conduct interim analyses of data obtained by month 8 to inform Aim 3, with final analyses of all data conducted in months 18-21.Anticipated manuscripts: 1) Patient understanding and distress associated with LCS results in a real-world setting; 2) Patient characteristics associated with poor adherence to LCS Aim Recruit participants and conduct focus groups; Month 6-7: transcription, coding, and analysis of qualitative interviews.Anticipated manuscripts: 1) Patient preferences for reporting and understanding of LCS results Aim Finalize LCS reports & surveys; Months 9-21: Randomize patients, collect surveys; Months 21-24: clean & analyze data.Anticipated manuscripts: 1) Patient-oriented outcomes associated with LCS report; 2) Implementation assessment g42, 43

3 .
REVIEW OF MOCK RADIOLOGY REPORT BASED ON LUNG-RADS: While we were waiting for this focus group discussion to begin, we gave you each time to view actual radiology reports from different LDCT scans.Let's walk through this example -not one of your own reportstogether, and I'd like to hear from you how well you understand the various pieces of the report.a.What were the results of this scan, and what do the results mean for the patient?b.What parts of the report make sense to you, and please explain to the group what those pieces mean.c.What parts are confusing to you, and please explain to the group what you find confusing.d.Is it a good idea for patients to see the radiology report?e.Is it a good idea for patients to see the pictures from the radiology test itself?When your doctor went over the results of your LDCT with you, what parts of the discussion did you understand well?How were those parts explained to you?What was it that you liked about your doctor's communication?Did your doctor talk to you about your risk of lung cancer?Did you want to know what your risk of lung cancer was based on your LDCT?