Attitudes and Perceptions About Smoking Cessation in the Context of Lung Cancer Screening

Importance  Broad adoption of lung cancer screening may inadvertently lead to negative population health outcomes if it is perceived as a substitute for smoking cessation.

Objective  To understand views on smoking cessation from current smokers in the context of being offered lung cancer screening as a routine service in primary care.

Design, Setting, and Participants  As an ancillary study to the launch of a lung cancer screening program at 7 sites in the Veterans Health Administration, 45 in-depth semi-structured qualitative interviews about health beliefs related to smoking and lung cancer screening were administered from May 29 to September 22, 2014, by telephone to 37 current smokers offered lung cancer screening by their primary care physician. Analysis was conducted from June 15, 2014, to March 29, 2015.

Main Outcomes and Measures  Attitudes and perceptions about the importance of smoking cessation in the context of lung cancer screening.

Results  Lung cancer screening prompted most current smokers to reflect for the first time on what smoking means for their current and future health. However, 17 of 35 (49%) participants described mechanisms whereby screening lowered their motivation for cessation, including the perception that undergoing an imaging test yields the same health benefits as smoking cessation. Other misperceptions include the belief that everyone who participates in screening will benefit; the belief that screening and being able to return for additional screening offers protection from lung cancer; the perception by some individuals that findings from screenings have saved their lives by catching their cancer early when indeterminate findings are identified that can be monitored rather than immediately treated; and a reinforced belief in some individuals that a cancer-free screening test result indicates that they are among the lucky ones who will avoid the harms of smoking.

Conclusions and Relevance  In this qualitative, lung cancer screening prompted many current smokers to reflect on their health and may serve as a potential opportunity to engage patients in discussions about smoking cessation. However, several concerning pathways were identified in which screening, when offered as part of routine care and described as having proven efficacy, may negatively influence smoking cessation. Health care professionals should be aware that the opportunity for early detection of lung cancer may be interpreted as a way of avoiding the harms of smoking. To promote cessation, discussions should focus on the emotional response to screening rather than clinical details (eg, nodule size) and address misperceptions about the value of early detection so that screening does not lower motivation to quit smoking.

Health care professionals and health systems are beginning to implement lung cancer screening following the publication of the National Lung Screening Trial (NLST) and recommendations by organizations, including the US Preventive Services Task Force, that patients with a history of heavy smoking be offered annual lung cancer screening with low-dose computed tomographic (LDCT) chest imaging.^1-4 Guidelines have emphasized, and Medicare requires, carefully counseling patients to ensure that screening is not perceived as a substitute for prevention (smoking cessation).⁵

Prior reports on smoking cessation rates following lung cancer screening have been mixed.^6-11 In the ongoing Dutch-Belgian screening trial,¹² participation in screening was associated with a lower quit rate of 14.5% at 2 years among current smokers, compared with 19.1% (P < .05) in the group that did not receive screening. When these rates were adjusted for the lower follow-up response rate by the control group, the difference was no longer significant (P = .35).⁶ The NLST⁸ reported that participation in LDCT screening had little effect on cessation, with the quit rate at 3 years of 23.8% in those undergoing LDCT screening and 23.2% in those undergoing chest radiography (P = .38). Little is known about how patients will respond to lung cancer screening as it moves from being offered as an exploratory intervention in the context of a clinical trial to a service with proven efficacy offered as part of routine care. Enthusiasm for cancer screening is widespread in the United States; more than 90% of adults believe cancer screening is almost always a good idea.¹³ The large National Survey of Medical Decisions (DECISIONS study)¹⁴ of breast, colorectal, and prostate cancer screening observed that patients substantially overestimate the benefit of cancer screening and have misperceptions about how much screening will reduce their chances of being harmed by cancer. Discussions about the risks and benefits of other cancer screening tests have proved to be challenging, with many patients not fully understanding the limitations of screening.¹⁵ Distortions of logic, specious rationalizations, and cognitive dissonance related to health outcomes and tobacco use is common among smokers, especially heavily addicted,¹⁶ long-term smokers who are the target group for screening.^17,18

The Veterans Health Administration (VHA) Lung Cancer Screening Clinical Demonstration Project is one of the first efforts to implement screening in routine practice using a system-wide clinical reminder, similar to prompts for breast and colorectal cancer screening.¹⁹ Details about the Demonstration Project, including the educational materials developed for this effort, are available from the VHA National Center for Health Promotion and Disease Prevention.²⁰ The objective of our interview study was to learn from patients who were offered screening how the availability of screening influenced their motivations regarding smoking cessation.

Patients in this study were participating in the VHA’s Lung Cancer Screening Clinical Demonstration Project,¹⁹ which is a pilot study implementing a primary care–based clinical reminder in a sample of the VHA’s 151 medical centers. The lung cancer screening project is being implemented with the objective of identifying the processes and resources needed for broad dissemination of a lung cancer screening program across the VHA. For this study, from May 29 to September 22, 2014, we approached potential participants within a few days of being offered screening using the VHA’s medical record system.²⁰ Our goal was to conduct 2 interviews: 1 shortly after being offered screening and a second interview after the participant had received the results of the screening test, although owing to how quickly participants underwent their screening tests, most were interviewed only once, after they had received their results. Potential participants were identified consecutively at 7 sites, oversampling women and nonwhite individuals. Up to 30 invitations were sent per week. Participants were offered a $50 cash incentive for participation in each interview. Analysis was conducted from June 15, 2014, to March 29, 2015. All study procedures were reviewed and approved by the VA Puget Sound Health Care System Institutional Review Board and the University of Washington Institutional Review Board. All participants returned signed written informed consent forms by mail before the interviews were conducted.

In-depth semi-structured telephone interviews were used to explore veterans’ experiences with being offered lung cancer screening and receiving the results of the tests. The interview guides were developed to elicit attitudes and beliefs toward smoking and discussions with health care professionals regarding smoking and screening findings. Interviews were digitally recorded and summary transcriptions and notes were completed by the interviewer (C.S.).

Interviews were analyzed using simultaneous deductive and inductive content analysis.²¹ Inductive content analysis consists of open, unstructured coding and allows for the identification of previously unidentified or unexpected themes.^22,23 Deductive content analysis is similar to testing prespecified hypotheses in that the raw data are reviewed for evidence of support of a priori categories. A priori codes included constructs from known behavior change and smoking cessation theories that we hypothesized could be influenced by lung cancer screening, including consciousness-raising, emotional response, and self reevaluation.^9,24,25 Codes were iteratively developed to describe distinct aspects of participants’ experience. Coding was conducted by 2 members of the research team (G.S. and C.S.) with AtlasTi software (Scientific Software Development GmbH). All members of the research team iteratively reviewed transcripts and early codes that emerged from the first 17 interviews, continuing until saturation.²⁶ Transcripts were reviewed again to identify exemplar quotes and to assess groundedness—the degree to which findings are empirically faithful to the data.

Forty-five interviews were conducted with 37 patients.¹⁹ We approached 186 patients who were offered screening, for an overall participation rate of 19.9%. There were no significant differences in characteristics of veterans who participated in interviews vs those who declined in terms of pack-year smoking (P = .93), average age (P = .22), and educational attainment (P = .72) (Table).

Eight patients were interviewed twice, first after being offered screening but before undergoing it, and again after they had been notified of their results. Twenty-two patients were interviewed once after undergoing screening and being notified of their results. Seven patients were interviewed once after being offered screening but had not been screened at the time of the interview, with 4 of these patients declining screening within the study period. Three participants self-reported during their interview that they had quit smoking for at least 30 days since they were first offered screening.

We identified 3 major themes related to how lung cancer screening may influence attitudes and motivations around smoking cessation (Box 1): screening universally stimulated multiple complex emotions in the participants about smoking; screening is perceived as an external agent of change in contrast to the internal challenge of cessation; and screening reinforced the perception for some smokers that lung cancer is their main concern and dissociated smoking from other cardiovascular and coronary disease risks. In addition to these main themes, we identified 5 groups of misperceptions reported by patients offered lung cancer screening.

Box Section Ref ID

The interviews started by asking participants about the discussion they had with their health care professional in regard to lung cancer screening, and almost all participants began by describing their long history of smoking and failed attempts to quit. Participants described the offer of lung cancer screening as stimulating a period of self-reflection. Most participants described the availability of screening very positively; however, some participants described being anxious for many days thinking about smoking and lung cancer during the entire screening process, from the time of being offered the test, scheduling it, and waiting for the results.

The emotional arousal induced by screening was strongly influential to some participants. Notably, 3 participants reported having quit smoking for at least 30 days because of screening. One participant indicated that he quit after being offered screening because the conversation caused him to think differently about his health and smoking. One participant who reported quitting described how suspicious findings provided motivation for quitting: “I feel I am not going to smoke any more especially now, knowing they [2 nodules] will grow.” Other participants described being motivated to consider quitting by results of their tests, although most ultimately commented that now was not a good time to try quitting, said they could wait to see if the findings on their LDCT scan progressed, or provided other reasons why they were not likely to take immediate action to quit.

Nearly all participants provided unsolicited descriptions early in the interviews of prior experiences trying to quit smoking. The exasperated and hopeless tone about quitting contrasted strongly with the language used to describe the ease and effortlessness of screening. These participants described relief at being able to do something good for their health that their physician recommended. They described lung cancer screening as much simpler than other cancer screening tests. One participant described the test as “no fuss, no muss.” Although no participants directly stated that they saw screening as a substitute for cessation, most were extremely enthusiastic about how simple the screening process was while earlier in the interview they had spent a significant amount of time recounting the futility of trying to quit smoking.

Participants were asked during the interviews to think generally about their future health and what screening or their screening results (if available) meant for their future health. Most participants focused on talking about their future risk of lung cancer, with only 2 of 37 participants mentioning any other diseases attributed to smoking. This focus stood out, as many participants discussed their existing health problems, including chronic obstructive pulmonary disease, peripheral artery disease, prior myocardial infarction, or other conditions, for which smoking is a risk factor. There was little concern about how continuing to smoke would affect existing or future conditions. In probes on this topic, most participants were only concerned about how smoking contributed to lung cancer.

We identified 5 groups of themes that emerged related to how patients described misperceptions about lung cancer screening and their smoking behaviors (Box 2).

Box Section Ref ID

Many study participants expressed that offering screening was a highly valuable service because everyone who undergoes screening will receive a benefit from it. Although patients were provided with education about the limited absolute benefit of screening,²⁰ nearly all participants mentioned the belief that everyone who is screened will benefit in some way. Several patients used future-looking phrases to describe how they felt protected from lung cancer just knowing that a screening program is available.

We identified a similar theme focusing on beliefs about the individual benefit current smokers described receiving from screening. Many participants wanted to undergo screening to see “how much damage” they had done to their lungs, a theme that arose both in interviews prior to knowing the LDCT results as well as interviews after results were known. When probed about quitting smoking, participants indicated that they knew they needed to quit smoking, but often the interviewees would change the topic, focusing on the value of “catching [cancer] early.” Several participants brought up the idea that screening would provide a more precise estimate of their individual risk of developing cancer.

One unexpected theme that emerged from the interviews was the frequently expressed belief that detection of a nodule meant that screening was working for the participant. Rather than being alarmed by identification of a nodule or suspicious findings requiring monitoring with future imaging,^28,29 several participants expressed the belief that identification of the nodule meant their cancer had been found so early that it was currently harmless, and that monitoring would tell them when they needed to start treating it. Several participants indicated that they were expecting bad news; so when they were told the findings were not urgent and did not require immediate action, they expressed relief. In interviews conducted after patients had received the results of their scan, some participants reported feeling they should be more motivated to quit. However, when asked in more depth to describe those feelings, several patients described a lack of urgency for quitting linked with plans to follow their findings with additional imaging.

A minor theme that was mentioned by a few participants was the belief that screening could help them avoid devastating chemotherapy and other treatments experienced by relatives who received a diagnosis of cancer. These participants described beliefs that screening by itself could cure cancer if the cancer was found early enough. Descriptions of family members going through cancer treatment were very detailed, with several participants expressing interest in screening to avoid needing similar treatment.

Several participants voluntarily relayed stories about relatives or individuals they knew who smoked heavily all their lives and lived to be very old or died of something other than lung cancer. These stories were often prompted by probes about why they were interested in screening or what they thought their screening results meant for their health. Although participants did not explicitly state that their cancer-free results were proof that they were among these lucky individuals, this theme was tightly woven into their narratives about what the screening results meant to them. Interestingly, even participants who were identified with nodule findings described feeling that their smoking had not yet harmed them because they were not told they have a cancer diagnosis.

These interviews summarize some of the first experiences of current smokers’ reactions to lung cancer screening as part of routine care. They also highlight several important points that health care professionals and health systems should consider in the efforts to broadly implement lung cancer screening. The topic of lung cancer triggers many emotions among current smokers, and nearly every participant described misperceptions about smoking that were exacerbated by screening. The finding that 3 of 37 participants reported they had quit smoking for at least 30 days is encouraging, suggesting that screening can lead to a change in behavior.

Our finding that current smokers attach exaggerated personal benefits to lung cancer screening results has not been noted, to our knowledge, in studies among screening trial participants. Park and colleagues¹⁰ conducted qualitative interviews with current smokers from the NLST, highlighting a lack of emotional effect of screening results, and did not report that screening appeared to instill trial participants with a feeling of protection from the harms of smoking. An important difference is that Park and colleagues interviewed participants 1 year or more after undergoing screening; thus, recall bias may be one reason for the difference in the emotional reactions they observed compared with reactions seen in our participants. In addition, participants in the NLST were more educated, voluntarily enrolled in a trial, and their expectations may be different compared with those of patients offered a service by their primary care physician that now has demonstrated efficacy.

Our results highlight evidence from other areas of risk communication that find patients are highly affective in how they process medical risk information, with behaviors driven more by emotions than by factual information.³⁰ How best to use screening to motivate smokers to quit rather than perpetuating and increasing misperceptions about smoking is not known. We recommend that health care professionals deemphasize clinical aspects of screening results (eg, nodule size) and focus on emotional reactions smokers have to undergoing screening. Rather than confront such cognitive biases directly, which is likely to be ineffective or even strengthen such biases,^31,32 we propose that health care professionals use the opportunity of screening to engage smokers who are reluctant to quit in a broader discussion around personal risks of smoking (including, but not limited to, lung cancer), benefits of quitting, and barriers to cessation.²⁴ It is not surprising that current smokers hold misperceptions about their susceptibility to the harms of tobacco. However, we must acknowledge that these beliefs can be reinforced and potentially exacerbated by screening. Our finding that cancer-free screening test results were consistently perceived by current smokers as support for the belief they have avoided the harms of smoking may indicate that it will be challenging to dissuade patients of such misperceptions.

The lack of concern when a nodule or other abnormal finding was detected by screening contrasts with reports from patients with incidental nodules detected on diagnostic imaging, where patients have reported being surprised and distressed.^29,33,34 This response may be owing to patients entering into the screening process looking for (and sometimes even expecting) abnormalities, and are relieved that they are not told they have cancer. Expectations are clearly important and have implications for using screening results as a teachable moment for cessation counseling.^7-11

This study has both strengths and limitations. The qualitative nature of the study provides rich descriptions of study participants and their potential motivations regarding cessation but does not allow us to quantify how screening will affect quit rates or generalize to broad populations. Our participation rate was 19.9%, which is appropriate for a qualitative study³⁵; however, those who participated may have been more positive toward screening. Although we did not observe differences between participants and those who declined in demographic variables and other factors including pack-years, there may be differences in other factors such as trust in health care professionals. Study themes were present across all demographic groups; however, despite purposive sampling of women and nonwhite patients, these groups remained underrepresented, precluding subgroup analyses. The number of participants identified with abnormal screening results in our sample (18 of 30 [60%]) was higher than in the NLST, which only reported positive nodule findings. Community studies have reported that between 16% and 73% of patients will have some type of abnormal finding on initial screening LDCTs.^36-39

Offering lung cancer screening to current smokers triggers many strong emotions. Health care professionals should be aware that many patients will overestimate the value of screening and that patients will work to interpret the findings in a way that reinforces cognitive biases about their smoking behavior. We need to develop messages to address misperceptions, especially because many long-term smokers are undergoing screening looking for evidence that smoking has not harmed them or are hoping screening will protect them from future cancer risks. We should also carefully monitor smoking cessation rates as screening is implemented, because these cognitive biases may not be able to be eliminated and lower cessation rates may be an unavoidable harm of screening.

Accepted for Publication: May 20, 2015.

Corresponding Author: Steven B. Zeliadt, PhD, MPH, Center of Innovation for Veteran-Centered and Value-Driven Care, VA Puget Sound Health Care System, 1660 S Columbian Way, Seattle, WA 98108 (steven.zeliadt@va.gov).

Published Online: July 27, 2015. doi:10.1001/jamainternmed.2015.3558.

Author Contributions: Dr Zeliadt had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Zeliadt, Heffner, Sayre, Au.

Acquisition, analysis, or interpretation of data: Zeliadt, Heffner, Sayre, Klein, Simons, Williams, Reinke.

Drafting of the manuscript: Zeliadt, Sayre, Simons, Au.

Critical revision of the manuscript for important intellectual content: Zeliadt, Heffner, Sayre, Klein, Simons, Williams, Reinke, Au.

Statistical analysis: Zeliadt, Reinke.

Obtained funding: Zeliadt, Sayre, Williams, Au.

Administrative, technical, or material support: Zeliadt, Heffner, Simons, Williams.

Study supervision: Zeliadt, Williams.

Conflict of Interest Disclosures: Dr Au reported receiving grant funding from Gilead Sciences, travel and honorarium from Boehringer Ingelheim, and honorarium from the Society of Hospital Medicine for topics related to chronic obstructive pulmonary disease quality of care. No other disclosures were reported.

Funding/Support: This study was supported by grant DP9001911-05S2 from the Centers for Disease Control and Prevention’s Health Promotion and Disease Prevention Research Centers Program through a Special Interest Project (SIP13-038: Low Dose CT Lung Cancer Screening: Informed Decision Making and Smoking Cessation). This material is the result of work supported by resources from VA Puget Sound Health Care System.

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the US government. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

Previous Presentation: This study was presented in part at the 36th Annual Society for Medical Decision Making Meeting; October 21, 2014; Miami, Florida.

Additional Contributions: We acknowledge the input of clinical coordinators and providers involved in the Veterans Health Administration’s (VHA) Lung Cancer Screening Demonstration Project who provided guidance to the research team in understanding field activities and interpreting the findings reported by participants. Linda S. Kinsinger, MD, MPH, and Martha L. Larson, RN, BSN, MS, VHA National Center for Health Promotion and Disease Prevention, are overseeing the VHA Demonstration Project. Kim Hamlett-Berry, PhD, and Dana E. Christofferson, PhD, Department of Veterans Affairs Office of Public Health and members of the VHA’s Smoking and Tobacco Use Cessation Technical Advisory Group, provided guidance on preliminary findings. Thomas B. Richards, MD, and Mary C. White, ScD, MPH, Epidemiology and Applied Research Branch, Centers for Disease Control and Prevention, interpreted preliminary study data. We also thank anonymous reviewers for their thoughtful comments on an early version of the manuscript. None of the contributors were compensated for their contributions.