A total of 173 participants responded to the survey (85 online and 88 by paper). All returned surveys were included in the analysis, including those that were partially completed. Participant characteristics are presented in Table 2 and included screening-eligible adults from WV with variation in demographic characteristics, SES, and prior LCS experience. SES indicators were examined exploratorily to contextualize clinical conversation experiences and screening-related outcomes in a rural population with known screening disparities.
Table 2 Participant characteristics (n = 173)
Analyses were conducted using all available responses for each item, so sample sizes vary across variables due to item-level missing data and because questions were only applicable to participants who reported having a conversation with a healthcare provider. Tables 3 and 4 present analyses stratified by screening status, with sample sizes reflecting participants with valid responses for screening status (n = 170).
Table 3 Shared decision-making conversation elements overall and by lung cancer screening status. Lower mean values indicate greater occurrence of shared decision-making elements (1 = Yes, 2 = No)Table 4 Knowledge discussed and post-decision perceptions by lung cancer screening status. Lower mean values indicate greater knowledge discussion or stronger endorsement of decision-related perceptions (1 = Yes/Very, 2 = No/Less)
Conversation elements
The prevalence of specific conversation elements and their associations with LCS completion are presented in Table 3. Nearly half of participants reported speaking with a healthcare provider about LCS, and among those who reported a screening conversation, most indicated that the discussion was initiated by the provider. Discussions of reasons to get screened were common, whereas discussions of potential harms, use of decision support tools, and provision of educational materials were reported less frequently.
LCS completion was significantly associated with multiple conversation elements. Participants who completed LCS were more likely to report having spoken with a healthcare provider about screening (p < .001), provider initiation of the discussion (p < .001), discussion of reasons to get screened (p = .003), elicitation of patient screening preferences (p = .012), receipt of a clear screening recommendation (p = .039), and receipt of educational materials (p = .003). No significant differences were observed for discussions of reasons not to get screened or use of decision support tools. Across significant indicators, participants who completed LCS consistently reported greater exposure to facilitative conversation elements.
Exploratory analyses indicated no differences in conversation elements by educational attainment. Participants with higher household incomes were more likely to report having spoken with a healthcare provider about LCS (p = .039), with no other income-based differences observed.
Multivariable predictors of LCS completion
A binomial logistic regression model examined the independent associations between selected conversation elements and LCS completion. The model included provider initiation of the screening conversation, elicitation of patient screening preferences, provision of a clear screening recommendation, and receipt of educational materials. The model was statistically significant (χ²(4) = 26.07, p < .001) and correctly classified 69.6% of cases (sensitivity 81.6%, specificity 55.4%).
Provider initiation of the screening conversation (OR = 4.07, 95% CI [1.86, 8.88], p < .001) and receipt of educational materials (OR = 2.63, 95% CI [1.02, 6.78], p = .045) were independently associated with higher odds of LCS completion. Elicitation of patient preferences (OR = 1.21, 95% CI [0.46, 3.16], p = .700) and providing a clear recommendation (OR = 0.80, 95% CI [0.31, 2.10], p = .652) were not significant in the multivariable model.
Information sources
Ratings of information sources, collapsed as important versus unimportant, are shown in Fig. 1. Healthcare providers were most frequently rated as important for LCS decision-making, whereas social media was least frequently rated as important. Family members were commonly identified as important sources, while friends and internet-based sources were more variable. Participants who completed LCS were significantly more likely to rate healthcare providers as an important information source compared with those who did not complete screening (p = .023). No other information sources differed by screening status.
Fig. 1
Ratings of information sources on lung cancer screening decisions
Exploratory analyses indicated no differences by educational attainment. Participants with lower household incomes were more likely to rate social media (p = .019) and family members (p = .050) as important information sources of screening information, suggesting greater reliance on informal or non-clinical channels.
Content discussed during LCS conversations
LCS content discussed during screening conversations is presented in Table 4. Discussions of screening frequency and potential diagnosis or treatment were more commonly reported than discussions of potential screening harms. Participants who completed LCS were significantly more likely to report conversations that included discussion of screening frequency (p < .001), potential harms (p = .003), and potential diagnosis or treatment (p = .010) compared with those who did not complete screening.
Exploratory analyses indicated no differences in content discussed by educational attainment. Participants with lower household incomes were less likely to report discussions that included potential diagnosis or treatment compared with those reporting higher household incomes (p = .047).
Decision attributes influencing LCS decisions
Participants’ ratings of attributes influencing LCS decisions, collapsed as important versus unimportant, are illustrated in Fig. 2. Clinical effectiveness of screening and lung cancer treatment, along with provider recommendation, were the most strongly endorsed decision attributes. Practical considerations, including screening convenience, facility location, and cost or insurance coverage, were also commonly rated as important, whereas family member or friend recommendation was less influential overall. LCS completion differed significantly for only one decision attribute. Participants who completed LCS were more likely to rate the perceived success of lung cancer treatment as an important factor in their screening decision (p = .050). No other decision attributes were significantly associated with screening completion.
Fig. 2
Ratings of lung cancer screening decision attributes
Exploratory analyses indicated no differences in decision attribute ratings by educational attainment. Participants with lower household incomes were more likely to rate screening convenience (p = .020) and family or friend recommendation (p = .050) as important decision attributes, highlighting the role of logistical feasibility and social context in screening decisions among economically constrained populations.
Post-conversation perceptions
Participants’ post-conversation perceptions of being informed and confident following LCS clinical conversations are summarized in Table 4. Overall, most participants reported feeling informed (72.5%) and confident (77.4%) after discussions about LCS; however, more than one in five participants reported lower levels of feeling informed or confident. Nearly all participants (91.8%) indicated that feeling informed and confident was important in their LCS decision-making.
Independent samples t-tests indicated that participants who completed LCS were significantly more likely to report feeling informed (t(157) = -3.96, p < .001) and confident (t(156) = -2.73, p = .007) following clinical conversations compared with those who did not complete screening. In contrast, the importance placed on feeling informed and confident did not differ by screening status (t(143) = -0.04, p = .967).
Exploratory analyses examining SES factors indicated no significant differences in post-conversation perceptions by household income. Educational attainment, however, was associated with post-conversation perceptions. Participants with a high school diploma or less were more likely to report feeling informed (t(156) = -2.76, p = .006) and confident (t(155) = -2.66, p = .009) following clinical discussions compared with those with higher levels of formal education. These participants also placed greater importance on feeling informed and confident in their screening decisions (t(143) = -2.06, p = .020).
