Lung cancer is the leading cause of cancer-related deaths in both men and women in the United States [1,2] and is expected to account for approximately 21% of all cancer deaths in 2022 [2]. Most lung cancer occurrences are diagnosed at advanced stages, after the cancer has metastasized, directly contributing to the deadliness of the disease [2]. Lung cancer screening (LCS) with low-dose computed tomography (LDCT) can detect lung cancer in the early stages when curative and less-invasive treatment options exist [3,4]. LCS received a B recommendation from the United States Preventive Services Task Force (USPSTF) in late 2013, prompting private insurance companies to cover the cost of the service for high-risk individuals, defined as individuals aged 55 to 80 years, with a minimum tobacco smoking exposure history of 30 pack-years and currently smoking or have quit within the past 15 years [5]. The Centers for Medicare and Medicaid Services (CMS) subsequently approved reimbursement for LCS in 2015 for Medicare beneficiaries, although they recommended stopping screening at 77 years of age [6]. Recently, the LCS eligibility criteria were expanded by lowering the recommended starting age to 50 years and tobacco exposure to 20 pack-years [7,8]. The USPSTF implemented these recommendations in 2021, followed by Medicare in early 2022.

The efficacy evidence for implementing LCS in routine clinical practice is based on the results of 2 large randomized trials: the National Lung Screening Trial [3] and the Dutch-Belgian Lung Cancer Screening Trial [4]. The results from these trials showed that LCS can reduce lung cancer–specific mortality by at least 20% when performed annually. Importantly, both trials diagnosed more early-stage cancers on follow-up rounds of screening than on the first baseline LDCT scan [3,9], highlighting the importance of sustained engagement with LCS programs. Adherence to multiple rounds of screening exceeded 90% in both clinical trials, further demonstrating the importance of annual screening to maximize the effectiveness of LCS [3,9]. Unfortunately, the translation of yearly LCS participation has emerged as one of the core threats to fully achieving the population health benefits of early lung cancer detection, with real-world annual adherence rates as low as 22% [10]. There is an urgent need for individual- and system-level interventions that focus on increasing the annual adherence rates.

The American College of Chest Physicians recommends that LCS programs develop strategies to maximize compliance with annual screening, including mechanisms to communicate with screening participants [11]. Community and academic LCS programs have reported using reminder letters to address adherence [12-14], and the Community Preventive Services Task Force (established by the United States Department of Health and Human Services in 1996) strongly recommends client reminders for breast cancer screening with mammography and colorectal cancer screening with fecal occult blood test [15]. Research supporting this recommendation found that any type of reminder (live telephone call, automated telephone call, letters, and postcards) increased adherence rates over no reminders, and targeted and tailored materials were superior to generic materials for continued screening participation [16,17].

Although research on established cancer screening modalities serves as a starting point for understanding interventions for reminder preferences, modes, and frequency, the needs and context of the LCS-eligible population must be considered during intervention development. For optimal impact, the unique elements of the LCS and the eligible community must be used to inform efforts to build and sustain engagement with the LCS process. At the center of this milieu is the stigma surrounding cigarette smoking and lung cancer, which is deeply embedded in all levels of society and presents as a barrier for some LCS-eligible candidates to participate in screening [18]. In addition, individuals eligible for LCS often face additional health care disparities, as cigarette smoking is linked to indicators of low socioeconomic status such as rural residence, individuals with Medicaid or being uninsured, and racial and ethnic minority groups [19]. Furthermore, health messages that contain pictures are processed more efficiently [20], highlighting another integral component of LCS communication. Understanding the informational needs and imagery considerations of the LCS-eligible population is crucial for messaging in reminder intervention development.

This paper describes an iterative, mixed methods study that uses the principles of user-centered design and participatory research methods to develop a reminder message intervention specific to annual LCS. This research study will (1) determine how LCS participants process health information regarding screening to identify possible options for reminder message content targeting and tailoring, (2) pinpoint imagery options that engage LCS participants, and (3) use these data to develop adherence to reminder messages for individuals who participate in LCS. Accordingly, the aims of this study are as follows:

The goal of this research is to use a theory-informed, multiphase, and mixed methods approach with LCS experts and participants to develop a set of clear and engaging reminder messages to support LCS annual adherence. In a 3-step process, survey data (step 1) and themes for imagery informed by LCS participant photovoice and key informant mixed method interviews (step 2) will inform the development of a pool of candidate reminder messages (step 3) for usability evaluation and testing with LCS experts and participants. This research will be guided by methodological, theoretical, and conceptual frameworks and principles to direct the steps needed to design and evaluate reminder message content, provide guidance to measure relevant constructs about how LCS participants process health information, and inform message context and imagery through engagement with LCS participants.

This research study will be guided by the Step approach to Message Design and Testing (SatMDT), a four-step conceptual process—(1) preexisting individual characteristics, (2) message-related characteristics, (3) individual responses, and (4) message outcomes—that will guide the overall reminder message development, from content identification to outcome evaluation [21]. As shown in Figure 1, step 1 of the SatMDT corresponds to study aims 1 and 2, informing the selection of relevant message content identified from quantitative survey data and further guiding reminder message imagery selection through the photovoice and interview activity. Aim 3 operationalizes SatMDT steps 2 and 3 by first using content results from aim 1 and imagery results from aim 2 to develop 10 to 12 reminder messages with varying visual combinations for each reminder type (ie, letter, postcard, and SMS text messaging). Second, it conducts preliminary testing of the message content and imagery combinations with rapid, iterative feedback with LCS experts and participants. SatMDT step 4 is outside the scope of this study. However, future studies will evaluate the efficacy and effectiveness of reminder messages to improve LCS annual adherence rates in clinical trials and real-world settings and on implementation testing to measure the acceptability and feasibility of using the reminder system in busy clinical settings.

To find informative and appealing content for the reminder messages that can further be targeted and tailored to the intended recipients, the C-SHIP model [22] will be applied to identify, contextualize, and operationalize the content of the reminder messages for the aim 1 quantitative survey. The C-SHIP model draws from several theoretical frameworks within the fields of health, cognitive, clinical, social, and behavioral psychology to identify health information seeking and processing behaviors, with the main premise detailing that health decision-making processes are the result of an interaction between the individual and the environment [22]. Aim 1 survey data informed by C-SHIP will be collected to assess how LCS participants process health information aimed at health protective behavior to develop content for the reminder messages and pinpoint options for message targeting and tailoring.

Additional research methodology will be guided by user-centered design and participatory photovoice to identify relevant content and imagery through engagement with LCS-eligible individuals and assessing preliminary usability testing. User-centered design is an intervention design approach that encompasses information about the individuals that will be using the intervention in the development and testing processes [23]. The goal of user-centered design is to maximize the usability of the intervention among the population of interest in a real-world context [23]. User-centered design will be used to (1) design the reminder intervention based on the characteristics of LCS participants and (2) refine message content and imagery based on rapid testing and feedback from LCS experts and participants. Consideration of the needs and emotions of LCS participants is vital for effective adherence reminder messaging.

Photovoice is a participatory research method that invites participants to take pictures representing an area of research within their community, allowing areas of concern to be documented and contemplated [24]. Traditional photovoice is completed by providing participants with a camera to capture the visual representation of the concern or research question and is then discussed through critical dialog in focus groups [25]. In this research project, photovoice will be used to invite participants to identify images that represent LCS to them and to identify themes for reminder message imagery through mixed methods interviews with the photovoice participants to understand the selection, likes, and dislikes of each photo.

The purpose of aim 3 is 2-fold: first, to develop a group of reminder message candidates for 3 platforms (ie, letter, postcard, and SMS text messaging) with differing content and imagery combinations, and second, to conduct preliminary usability testing of the messages with LCS experts and participants. Evaluative testing will consist of rapid, iterative feedback cycles, with the refinement of messages occurring after each cycle of 3 experts or participants.

Experts will be prespecified by the research team before the survey invitation and will include pulmonologists, radiologists, primary care clinicians, advanced practice providers, medical oncologists, and implementation scientists with expertise in LCS. Experts who agree to complete the evaluative survey will be included in aim 3. LCS experts who do not agree to participate in the rating process will be excluded.

LCS experts will be selected a priori based on the knowledge and contacts of the research team. Experts will be contacted via email to introduce the study and ask about their willingness to participate. Every effort will be made to have a balanced panel of experts (ie, pulmonologists, radiologists, advanced practice coordinators, primary care providers, medical oncologists, and implementation specialists). Informed consent will be obtained electronically using REDCap.

Expert surveys will be closed-ended and administered on the REDCap electronic platform. Respondents will be presented with a mock-up of each message-type and a brief clinical vignette with targeting and tailoring characteristics (if applicable) and asked to rate each message on 6 categories adapted from the Centers for Disease Control and Prevention (CDC) Health Communication Playbook [51] and the National Institutes of Health (NIH) Health Literacy Online [52] guidelines. Message ratings will use a 7-point Likert scale, where 1 is “very poor” and 7 is “very good,” to rate the following categories specific to LCS annual adherence: (1) relevance of the message content, (2) readability of the message content, (3) accuracy of the message content, (4) intent of the message content (conveying screening benefit), (5) memorability of the visual appeal, and (6) overall rating of the message content and visual imagery combination (Multimedia Appendix 1 [30-43]). Experts will be asked to evaluate each message candidate, 10 to 12 for each message-type platform (letter, postcard, and SMS text messaging), or 30 to 36 messages in total.

Rapid, iterative feedback and refinement of messages will be completed in cycles comprised of 3 experts. It is expected that usability testing will consist of 3 to 4 iterative cycles, with a total of 9 to 12 LCS experts.

Data from the expert survey will be quantitative. Responses will be entered into a matrix in Microsoft Excel, and similar responses will be categorized and ranked in the order of frequency. Messages with lower ratings will be revised accordingly. Message candidates will be refined based on the feedback of each feedback cycle.

Message evaluation participants will be recruited solely from the UCHealth University of Colorado Hospital LCS program, with emphasis on participants who are close to their annual screening LDCT (Lung-RADS result of 1 or 2). This sampling of participants will be separate from the aims 1 and 2 samples. The focus for this aim will be on participants within 4 to 6 weeks of their recommended annual screening LDCT. Participants will be eligible for annual screening by CMS or USPSTF guidelines, English speaking, and due for their annual LDCT scan within 4 to 6 weeks. Participants who do not agree or consent to participate in the reminder evaluation interviews or are not eligible for annual screening (aged out of screening age, >15 years since quitting smoking, or competing comorbidities) will be excluded from this study.

Participants eligible for aim 3 will be due for their annual screening within 4-6 weeks. These participants will be identified in the database that is used to clinically manage the screening program at the University of Colorado Hospital. Recruitment strategies and informed consent procedures will mirror those used for aim 2.

Evaluation interviews will be conducted on the Zoom platform, with participants shown each reminder message content and imagery combination during the interview. The interview guide will include both open- and closed-ended questions and is based on prior research work in adherence for mental health appointments [53]. A total of 3 questions will focus on the understanding of the message content, 3 questions will ask about the message appeal (likes and dislikes), and 1 question will ask about how wording or content could be improved (Multimedia Appendix 1 [30-43]). Interview procedures and participant compensation will be parallel to those used in the aim 2 photovoice interviews. Participants will be asked to evaluate approximately 20 to 25 message candidates, 7 to 8 for each message-type platform (letter, postcard, and SMS text messaging). The interviews are expected to last approximately 60 minutes, and participants will receive US $60 as compensation on the completion of the interview.

Rapid, iterative feedback and refinement of messages will be completed in cycles comprised of 3 participants. It is expected that usability testing will consist of 3 to 4 iterative cycles, with a total of 9 to 12 LCS participants.

The participant evaluation will use qualitative and quantitative data collected from the same survey instrument and will therefore be completed using the same procedures as in the aim 2 analysis. Quantitative and qualitative data will be analyzed separately and then merged to make final inferences and recommendations for further reminder message refinement. It is expected that the data types will corroborate one another, with higher-rated messages and visuals containing positive and encouraging themes and lower-rated messages associated with themes about confusion or misunderstanding of the message. The matrix design will also offer the opportunity for the qualitative data to explain the quantitative ratings and highlight the specific aspects the participants did or did not like about high- or low-rated messages.

The determination of the final reminder messages for each reminder type and matched visual picture will be driven by 3 main criteria: understanding, appeal rating, and anticipated impact. Difficulties in understanding the core message or with specific words will be used to refine the wording. Final message determination will consider results from both the usability expert survey and participant interviews by comparing the concordance or discordance between overall message ratings. Study team consensus will determine the final message set. The entire message development process is illustrated in Figure 2. The study team expects to have a minimum of 6 messages per communication type ready for robust testing in busy clinical settings.

This study was approved by the Colorado Multiple Institutional Review Board (COMIRB #21-3881) with full waivers of informed consent and Health Insurance Portability and Accountability Act of 1996 authorization to use LCS clinical databases and electronic health records to identify potential participants. In addition, a waiver of written documents for informed consent was obtained to allow for the ease of the consent process in REDCap. Participants in all aims will agree to the contents of an informational consent form containing details about the study purpose, procedures, confidentiality, and use of data before the initiation of any study procedures.

This project proposes a novel approach to facilitate adherence to annual LCS through the development of reminder messages that embrace content and imagery representative of the target population directly in the design process. Developing effective strategies to increase LCS adherence is instrumental in achieving optimal LCS outcomes at individual and population health levels. Engaging participants in yearly screening will ultimately increase the number of early-stage lung cancers detected, leading to improved outcomes and quality of life. Interventions aimed at addressing this critical need must be designed with the unique LCS population and context in mind and eliminate stigmatizing and biased content and imagery.

This research will develop message content based on health information processing theory conceptualized with the C-SHIP model, which recognizes health protective behavior decisions are the result of an interaction between the individual and their environment [22]. C-SHIP provides a broad framework to describe the characteristics of LCS-eligible individuals and constructs that may influence decisions to adhere to annual screening and to explore options for message targeting and tailoring. This project also aims to pinpoint imagery that is visually appealing and engaging in the LCS-eligible population. Imagery is often considered a significant component of health messaging and is even more important for individuals with low health literacy [54]. Specific to LCS, previous qualitative research in individuals at high risk for lung cancer found that images with cigarettes and lungs are ineffective in informational campaigns [55], highlighting the importance of identifying appropriate imagery for LCS communication. Finally, the resulting reminder messages will be iteratively tested with both LCS experts and participants to refine the message accuracy, understandability, and value of the content and imagery combinations.

Although this approach will leverage the strengths of participatory research and user-centered design methods, there are limitations that must be acknowledged using the available settings and resources. First, we will not be able to assess social cognitive theory for all LCS populations that experience health disparities and could benefit from message targeting and tailoring (eg, lesbian, gay, bisexual, transgender, queer or questioning persons or those with mental and substance disorders). Second, Colorado’s population lags the US population with regard to African American representation but exceeds the national average of individuals with Hispanic ancestry [56]. These demographic differences may compromise the generalizability of the initial development and testing of the message combinations. Finally, the efficacy and generalizability of these messages to clinical and health care settings are unknown. Despite these limitations, the design choices are appropriate to stimulate collaborative development of interventions directed specifically at improving LCS annual adherence and to highlight important areas of the future to strengthen the healthy equity reach and future efficacy, effectiveness, and implementation of the intervention.