Patient Perspectives on Low-Dose Computed Tomography for Lung Cancer Screening, New Mexico, 2014

Introduction National guidelines call for annual lung cancer screening for high-risk smokers using low-dose computed tomography (LDCT). The objective of our study was to characterize patient knowledge and attitudes about lung cancer screening, smoking cessation, and shared decision making by patient and health care provider. Methods We conducted semistructured qualitative interviews with patients with histories of heavy smoking who received care at a Federally Qualified Health Center (FQHC Clinic) and at a comprehensive cancer center-affiliated chest clinic (Chest Clinic) in Albuquerque, New Mexico. The interviews, conducted from February through September 2014, focused on perceptions about health screening, knowledge and attitudes about LDCT screening, and preferences regarding decision aids. We used a systematic iterative analytic process to identify preliminary and emergent themes and to create a coding structure. Results We reached thematic saturation after 22 interviews (10 at the FQHC Clinic, 12 at the Chest Clinic). Most patients were unaware of LDCT screening for lung cancer but were receptive to the test. Some smokers said they would consider quitting smoking if their screening result were positive. Concerns regarding screening were cost, radiation exposure, and transportation issues. To support decision making, most patients said they preferred one-on-one discussions with a provider. They also valued decision support tools (print materials, videos), but raised concerns about readability and Internet access. Conclusion Implementing lung cancer screening in sociodemographically diverse populations poses significant challenges. The value of tobacco cessation counseling cannot be overemphasized. Effective interventions for shared decision making to undergo lung cancer screening will need the active engagement of health care providers and will require the use of accessible decision aids designed for people with low health literacy.


Introduction
National guidelines call for annual lung cancer screening for highrisk smokers using low-dose computed tomography (LDCT). The objective of our study was to characterize patient knowledge and attitudes about lung cancer screening, smoking cessation, and shared decision making by patient and health care provider.

Methods
We conducted semistructured qualitative interviews with patients with histories of heavy smoking who received care at a Federally Qualified Health Center (FQHC Clinic) and at a comprehensive cancer center-affiliated chest clinic (Chest Clinic) in Albuquerque, New Mexico. The interviews, conducted from February through September 2014, focused on perceptions about health screening, knowledge and attitudes about LDCT screening, and preferences regarding decision aids. We used a systematic iterative analytic process to identify preliminary and emergent themes and to create a coding structure.

Introduction
Despite advances in treatment and management, lung cancer remains the most common cause of cancer-related disease and death in the United States (1). Most lung cancer deaths are attributed to smoking (2). Although recently declining smoking rates may reduce lung cancer incidence (1), former smokers remain at a high risk for the disease, and most lung cancers occur in former smokers (3).
Until recently, no effective screening method existed for detecting early-stage lung cancer (4). The multicenter National Lung Screening Trial (NLST) enrolled 53,454 high-risk smokers, defined as people aged 55 to 74 years with a 30 pack year smoking history, who were either current smokers or had quit within the previous 15 years (5). Lung cancer deaths among NLST participants randomly assigned to low-dose computed tomography (LDCT) were significantly reduced (16.0% [relative] (6) and 0.33% [absolute] (5) risk decrease) compared with those who had chest x-ray. However, concerns remain regarding the high false positive rates and complications following invasive diagnostic procedures (ie, percutaneous biopsy, bronchoscopy, or surgical procedure) (5).
Based on the NLST results, the United States Preventive Services Task Force (USPSTF) issued a Grade B recommendation supporting LDCT screening for high-risk adults from 55 to 80 years of age (7). The recommendation is important since the Affordable Care Act mandates coverage without copayment for preventive services that have a USPSTF recommendation of B or higher (8). Guided by the USPSTF recommendation, the Centers for Medicare and Medicaid Services (CMS) now provide annual coverage for lung cancer screening, with a requirement of a shared decisionmaking visit (ie, consultation between patient and health care provider), for beneficiaries meeting NLST eligibility criteria (9). The CMS directive also stipulates smoking cessation counseling and the use of decision aid(s) to discuss potential screening outcomes (9). The American Lung Association (10) and the American Cancer Society (11) also endorsed lung cancer screening for patients meeting the NLST eligibility criteria. Both organizations recommended both informed and shared decision making with clinicians and smoking cessation counseling (10,11).
Although there is a need to promote shared decision making for lung cancer screening (10,11), gaps exist in our understanding about how best to concisely and clearly present complex probability-based screening information (eg, lung cancer risk, false positives, consequences of diagnostic follow-up), while effectively integrating smoking cessation messages. Moreover, it is important to better understand the perspective of persons who are demographically different from those included in the NLST, the majority of whom were non-Hispanic white (91%) and male (59%) (5). To develop appropriate decision-support strategies, perspectives of people from diverse sociodemographic backgrounds need to be incorporated. New Mexico, as one of the 4 "majority-minority" states, features unique multiethnic and multicultural diversity with striking health and socioeconomic disparities. Lung cancer is a leading cause of cancer death among minority populations in New Mexico (12). We interviewed patients in clinics to characterize their knowledge and attitudes about LDCT lung cancer screening and smoking cessation and their views on supporting decision making for lung cancer screening. We previously reported on primary care providers' perspectives about implementing LDCT screening (13).

Study setting and sample
We conducted qualitative semistructured interviews with a diverse sample of patients who were receiving care at one of two settings in Albuquerque, New Mexico: a Federally Qualified Health Center (FQHC Clinic) or the University of New Mexico Comprehensive Cancer Center (UNMCCC) Multidisciplinary Chest Clinic (Chest Clinic). In New Mexico, the University of New Mexico in Albuquerque is the only academic medical center and is host to the only National Cancer Institute Comprehensive Cancer Center in the state. This poses access issues for rural patients. At the time of this study, neither the FQHC Clinic nor the Chest Clinic offered LDCT for lung cancer screening, nor were we aware of any comprehensive screening program in the state. The FQHC Clinic is part of the Research Involving Outpatient Settings Network (RIOS Net), the Practice-Based Research Network of New Mexico that predominantly serves Hispanic patients. By using a purposeful sampling approach, we identified, recruited, and interviewed adults aged 50 to 80 years with a history of heavy smoking who met NLST enrollment criteria (5). Eligible participants recruited from the FQHC Clinic self-reported not having lung cancer or nodules and not having undergone a lung computed tomography (CT) scan. Those recruited from the Chest Clinic were documented to have had abnormal results from a lung CT scan. Selecting a wide spectrum of at-risk patients enabled us to include perspectives of patients ranging from those considering screening to those who actually underwent the imaging and invasive diagnostic testing that would be part of a screening program. These perspectives were necessary to develop comprehensive decision support tools to inform patients about screening and surveillance testing and associated invasive diagnostic procedures. The University of New Mexico Health Science Center's Human Research Protections Office approved all aspects of the research protocol.

Data collection
We developed an interview guide that focused on the domains of knowledge and attitudes regarding LDCT lung cancer screening: perceived risk from diagnostic testing (including radiation exposure and complications from invasive diagnostic tests), views about the benefits and harms of lung cancer screening, and views about barriers and facilitators to smoking cessation or preventing relapse, particularly in the context of lung cancer screening. We also elicited views about the value, content, format, and timing of providing patient decision-making aids and preferences for discussing screening. Before the interview, we administered a survey that collected sociodemographic information. The survey and interview guide are available from the corresponding author.
Three members of the research team (A.L.S., A.M.M., C.M.G.) conducted the interviews from February through September 2014. Each digitally audio-recorded interview lasted from 45 to 60 minutes and was transcribed for analytic review. We conducted inperson interviews at the participant's choice of location. Each participant received a $50 merchandise card.

Data analysis
We used a systematic iterative process to identify preliminary and emergent themes of importance (14). Several research team members independently read sets of 2 or 3 transcripts to identify preliminary findings (S.I.M., A.L.S., A.M.M., C.M.G., R.R., R.M.H.), which facilitated confirmation of emergent themes and modifications to the interview guide. Through an iterative process of reviewing additional transcripts, we created an initial coding structure that was then revised until we reached consensus (15). We then imported all the transcripts into NVivo 10 (QSR International), a qualitative data analysis software program, for coding. After coding all transcripts, we queried the database by coding categories for a more refined level of interpretive analysis. We analyzed the quantitative survey data to produce a descriptive assessment of the study sample.

Sociodemographic characteristics of participants and their views on cancer screening
We interviewed 22 clinic patients, 12 from Chest Clinic and 10 from FQHC Clinic. Table 1 presents their sociodemographic characteristics. We categorized the qualitative interview findings into major themes, which were described by FQHC and Chest Clinic participants (Table 2). Participants in both clinics endorsed the general importance of screening for health problems, even if they personally did not undergo regular screening, because it "may reveal something unpleasant, but still it is good to know" to "catch things early." Some participants previously underwent cancer screening tests such as colonoscopy, mammography, or Pap smear; for one participant, this led to early cancer diagnosis and treatment. Not surprisingly, participants identified barriers to en-gaging in screening such as the cost of purchasing insurance, lack of transportation, and low health literacy (eg, "they use big words and I don't know what they mean"). They also generally noted that screening was not a normal behavior among their family and social networks.
Knowledge and receptivity to LDCT screening for lung cancer None of the participants interviewed at the FQHC Clinic had heard about using LDCT to screen for lung cancer. During the interview, we provided a brief explanation of LDCT screening, and most participants were interested. Participants who were either uncertain or conditional about whether they would undergo lung cancer screening expressed uncertainty about the value of screening (in terms of the benefits and necessity) and logistics of the screening procedure. Other participants were favorably inclined toward screening but were still concerned about costs and the accuracy of the test. Those strongly indicating that they would undergo screening were concerned for their health given the known consequences of long-term tobacco use.
Most Chest Clinic participants were unaware of LDCT screening. However, receptivity to receiving the test was generally high, especially if the test was recommended by a health care provider. Several participants reported being more inclined to undergo LD-CT screening (and other screening tests) because of a general sense of vulnerability that they attributed to their current situation (ie, being under surveillance for lung cancer or having a diagnosis of lung cancer) or their experience with other health problems.

Challenges to LDCT screening for lung cancer
In discussions with FQHC Clinic participants concerns about high costs, radiation exposure, and psychological distress (stress, anxiety) all emerged as considerations about whether to undergo LD-CT screening. Most participants were not dissuaded by the potential for false positives and the need for continuous followup and screening. We provided a brief explanation of LDCT screening but did not review specific details (eg, false positives, followup testing for positive scans). However, based on interviewer debriefing notes following the discussions, it was unclear whether all participants fully understood the meaning and implications of these terms (eg, false positives, followup testing for positive scans).
Chest Clinic participants identified several factors that could pose challenges to LDCT screening. They were particularly mindful about costs but said they would make sacrifices to obtain necessary care. As one participant noted, "I need to know if I have it or not . . . I'll deal with the money part of issues and aftercare when I get to that." Other challenges mentioned were the cost of treatment, transportation issues, distance to the screening site, being PREVENTING CHRONIC DISEASE www.cdc.gov/pcd/issues/2016/16_0093.htm • Centers for Disease Control and Prevention able to trust health care providers, taking tests that were inconclusive or invasive, and the need for annual screening. Participants were generally not concerned about radiation exposure, false positives, or the need for continued follow-up and screening.

Smoking cessation in the context of LDCT for lung cancer
We also explored views about how LDCT results -either positive or negative -might influence smoking behavior. FQHC Clinic participants indicated that undergoing screening by itself might not be a strong deterrent to smoking. However, several felt that a positive test result might be sufficient motivation for a quit attempt. Participants further noted that actually seeing a lung nodule on an image would serve as a more powerful message that something is wrong than just being told of a problem. Several indicated that even a normal test result would not be sufficiently reassuring and that they would still consider quitting smoking. However, participants expressing a fatalistic worldview might not be swayed into considering quitting. Chest Clinic participants expressed similar views about LDCT screening and smoking cessation. Of these 12 participants, 9 were no longer smokers and 5 had quit smoking because of a major scare (eg, receiving test results about a suspicious lung nodule, having a friend or family member with diagnosed lung cancer). Three Chest Clinic participants had quit smoking since learning about a positive finding on an x-ray or CT scan leading them to be seen at the Chest Clinic. However, several participants believed that a negative screening result might not be sufficient to persuade someone to quit smoking. Conversely, a few participants lamented that such a finding could have the opposite effect and serve as a green light to continue smoking.

Information needs and preferred communication methods
After describing LDCT screening to participants, we then asked them for input about the most important information to include in a decision-making aid. FQHC Clinic participants suggested pictures of a lung damaged by smoking, general information about lung cancer, benefits and harms of LDCT screening, cost and duration of screening, and potential future treatment and treatment side effects. Participants preferred learning about the screening test in a face-to-face encounter with their health care provider. Participants also suggested pamphlets or written materials; however, they raised concerns about poor literacy and inability to understand information. Others suggested using DVDs or CDs, which someone could view in a doctor's office and then seek clarification from the doctor. They felt that a Web-based decision aid might not be effective because of limited access to computers and the Internet, a concern with particular relevance in a low-income and rural state such as New Mexico.
Chest Clinic participants provided greater specificity regarding the types of information that they would like to see in a decision aid, including information about LDCT, risks and benefits of screening, false positives and negatives, and consequences of undergoing the test. These participants preferred to receive screening information through one-on-one discussions with a health care provider. Other suggested information sources included booklets (with pictures), videos, DVDs, and social media.

Discussion
This study presents findings from in-person qualitative interviews of high-risk patients, with or without lung cancer, regarding LD-CT screening for lung cancer, smoking cessation, and decisionmaking support. The views summarized in this article are some of the first from participants who were underrepresented in the NLST and from patients not considered eligible by the NLST criteria (ie, patients with a lung nodule or lung cancer). These interviews provided insights into the screening process and perceptions regarding smoking cessation within the context of screening. The findings offer insights into the relative value of focusing on smoking cessation counseling in the context of lung cancer screening, challenges to offering LDCT screening to sociodemographically diverse populations, and the content and structure of decision aids.
A consistent message from guidelines (9-11) is that patients be offered smoking cessation counseling along with lung cancer screening. Our findings provide evidence in support of these guidelines. Our findings suggest that smokers might attach personal benefits to lung cancer screening results, with negative findings implying "got good news" and "I'm not a problem smoker" and positive findings potentially motivating smokers to quit (16). Zeliadt et al (17), based on a qualitative study conducted at Veterans Health Administration sites, also reported that negative screening findings lowered participants' motivation for cessation. Meanwhile, the Chest Clinic patients who had quit smoking all cited the powerful effect of an abnormal finding, and many of the FQHC Clinic smokers thought that abnormal findings could motivate cessation. Indeed, the NLST reported that CT findings highly suspicious for cancer were strongly associated with subsequent smoking cessation (18). Our findings point to the importance of providing referrals for evidence-based smoking cessation interventions (16) and also integrating smoking cessation counseling when offering screening, discussing the results, and using these doctorpatient interactions as a teachable moment for cessation counseling (19); primary care providers in a parallel arm of this study endorsed this referral and counseling approach (13). Ultimately, achieving smoking cessation is likely to have a far greater public health benefit than screening alone.
Another recommendation from the lung cancer screening guidelines (9-11) is the use of decision aids and shared decision making between patient and physician. The lack of awareness about lung cancer screening coupled with high receptivity for the test among high-risk smokers underscores the need for decision aids. Before our study, however, there was no evidence to guide the content and structure of decision aids for lung cancer screening tailored for populations not recruited in the NLST. Our findings provide evidence for inclusion of key areas in decision aids, including a basic description of the LDCT screening test and how it is conducted, benefits and harms of screening, sequence of events following an abnormal screen, visual depiction of damage to the lungs from smoking, lungs with nodules or cancer, estimated cost of screening and follow-up procedures, and a clear statement about the benefits of smoking cessation regardless of screening results. Finally, our findings provide evidence for the need for shared decision making and the need for health care providers to actively engage eligible patients about smoking cessation and lung cancer screening through one-on-one interactions. The shared decision-making discussions should be supplemented with written and audiovisual (CD or DVD) aids to circumvent reading and health literacy constraints, and these discussions should be tailored to individual patient concerns and circumstances. A patient decision aid (20) is being evaluated in a Patient-Centered Outcomes Research Institute grant (21), and a Web-based (22) patient decision aid needs further study to determine its efficacy and generalizability.
Offering LDCT screening for lung cancer poses immense challenges. It is well-documented that participation in cancer screening is lower among racial/ethnic minorities than whites (23) and is positively associated with cancer-specific and screening-specific knowledge (24), proximity to screening centers (23), years of education (23), high income (23), residence in urban areas (25), social support and normative behaviors that promote preventive behaviors (26), and private insurance (23). These positive predictors for screening do not typify participants in our study who were predominantly Hispanic, unemployed, with low levels of education, less normative behavior for preventive care, and on government insurance (Medicaid). Furthermore, participants were unaware of the existence of the LDCT screening test and identified additional barriers to screening such as high costs, transportation barriers, need for annual screening, and psychological distress. Therefore, despite participants' indicating their receptivity to LDCT screening, sociodemographic and psychosocial barriers may preclude equitable delivery and uptake of lung cancer screening and may further perpetuate lung cancer disparities.
The health equity issues related to people who do not resemble those enrolled in the NLST may also extend to geographic areas. Guidelines are clear that screening should not be offered except through centers of excellence (7,11). Most NLST sites were university or academic cancer centers. Other hospital centers in New Mexico may not have sufficient resources and expertise to meet the criteria necessary to offer lung cancer screening. This could result in patchy national availability of lung cancer screening that could further exacerbate disparities. The number of institutions nationally having an active LDCT screening program are increasing (27); however, concerns exist regarding screening use, infrastructure and resources to handle screening demands, and compliance with recommended guidelines. Possible solutions could be to rely more on telemedicine to ensure high-quality image interpretations and to develop regional centers of excellence that could serve a broader population. Applying more stringent criteria for screening could also be beneficial by increasing the efficiency of screening. A post-hoc analysis of the NLST data showed that the 3 highest quintiles of risk accounted for 88% of the screening-prevented lung-cancer deaths (28).
Our study has strengths and limitations. The findings from this qualitative study are not representative or generalizable to other types of clinical settings and do not represent the unique belief systems and needs of sociodemographically diverse populations. The results reported here are consistent and contextually relevant perceptions of smoking cessation and lung cancer screening from 2 groups of high-risk patients and are necessary to guide future efforts toward implementing LDCT screening, especially in populations and communities unlike those featured in the NLST study. Although the National Cancer Institute excluded people with a history of lung cancer from the NLST, the American Association for Thoracic Surgery (29) and the National Comprehensive Cancer Network (30) recommend screening for lung cancer survivors. Participants in our study represent a challenging population to screen, especially in terms of access to care, health literacy, and insurance coverage. These participants and associated findings further underscore challenges in translating research into practice.
Implementing lung cancer screening programs among sociodemographically diverse populations will be challenging and could further perpetuate lung cancer disparities. Value of smoking cessation counseling for current smokers within the context of lung cancer screening cannot be overemphasized, especially in light of the evidence that positive screening findings might be viewed as a motivation to quit. Screening programs will need literacy-appro- General views on preventive care (screening)

PREVENTING CHRONIC DISEASE
•"[It's] especially like when you're maybe 35, 40 years old, that's when people start opening their eyes. I think that as you get older, people are more interested in, 'yeah, let's go check something out.'" "Well, I think that those are great. I get a colonoscopy every five years and a mammogram every year." • "I think more people would probably participate providing they can afford it. When you mention medical to anybody, the first thing they think of is, 'here goes the money.'" • " • "All the information I could get . . . and in plain words, which I could probably understand it, that would be the best. Information about the test and the consequences and the procedure itself time-wise. Most people do sit down a lot and watch movies or DVDs. I think they would just like a 15 minute explanation of the procedure, how it's done, why this is being done, what kind of machine is gonna be used. The doctor is the actual one that's gonna give you the results." • "I think there should be a lot more, instead of the smoking cessation like the 1-800-QUIT NOW. They showed the guy with the hole in the throat, the woman with the lung removed, things like that, and then after that could be an informative little message about if you [smoke] . . . this could be prevented with a CT scan now.
[A] family member should be involved or family member to come with them, 'cause the family member will most likely talk them into it. For me I want to know from the doctor. I think pamphlets in the doctor's office on the wall, like they have other pamphlets there, because the longer you wait, the more you get bored and the more you read." • "Just show me pictures of lungs. People on more medication. When I come into my appointments and I'm sitting down, that's how I read. Sometimes people won't come to doctors, because they just have a thing; for them I would say grocery stores, where the ads are for the Quick Quarter or the cars. Pamphlet there or something for smokers." • "All the risk and the cost, but mostly the risk. "Maybe on a DVD or something so you could watch, 'cause sometimes a lot of people don't like to read. Like I don't like to really read. The doctor, too, but sometimes the doctors have certain amount of time to talk to you. Sometimes some people don't have [time] for the Internet. They don't have computers, I don't have a computer." • "The correct information, the information that is stated on a piece of paper and the information that is stated from a professional. Information about the procedure, risks, benefits, future treatment, side effects. A video, you know like a video going through a facility like we're here and you pull out a video and say, 'look, this is what it consists of.' It could be a trained health professional." • "I know I've smoked for a long time. I don't really know what my lungs look like. If the people knew how bad their lungs were, maybe they would think about it a little more. Pretty much straight from my doctor, yeah I'd rather hear from him. Reading it and stuff really just doesn't hit you as like the doctor telling you. You might have a question that you want to ask, and with a video you won't be able to ask any questions."