The Feasibility and Equity of Text Messaging to Determine Patient Eligibility for Lung Cancer Screening

ABSTRACT

Objectives: Text messaging could be effective for determining patient eligibility for lung cancer screening (LCS). We explored people’s willingness to share their tobacco use history via text message among diverse groups.

Study Design: Cross-sectional survey.

Methods: In 2020, we conducted a cross-sectional survey asking respondents about cellular phone usage, smoking habits, sociodemographic characteristics, and the likelihood of responding to a text message from their health care provider’s office about tobacco use. We used χ² and analysis of variance tests for comparisons.

Results: Among 745 respondents, 90% used text messaging casually. Overall, 54% never smoked, 33% currently smoked, and 13% previously smoked. Six percent were LCS eligible, and 20% used both cigarettes and e-cigarettes (dual users). Current smokers were significantly younger, less likely to be female, and more likely to use text messaging. LCS-eligible respondents were older and less likely to have a high income. Dual users were younger, less likely to report female gender and live in rural areas, and more likely to have a college education and high income. Most respondents (83%) indicated they were likely to respond to text message inquiries regarding smoking status. Middle-aged respondents (mean age, 37 years) were significantly more willing to report smoking status than younger or older respondents (91% vs 84% and 84%, respectively). Respondents with no college education (83% vs 88%) or with a low income vs a middle or high income (81% vs 86% and 88%, respectively) were significantly less willing to report smoking status via text messages.

Conclusions: Text messaging showed promise for evaluating smoking history and for simplifying the process of identifying LCS-eligible individuals. However, achieving equity in identifying eligibility for LCS requires the implementation of multimodal strategies.

Am J Manag Care. 2024;30(9):In Press

Takeaway Points

• Given that 90% of respondents used text messaging and 83% indicated they were likely to report smoking status via a text message, this approach shows promise for identifying eligibility for lung cancer screening (LCS).
• Middle-aged respondents with a college education and medium or high income were more likely to report their smoking status via text messaging.
• A text message program alone might not comprehensively identify an LCS-eligible population. Those overlooked are likely the individuals who could benefit the most from LCS.
• A multimodal approach is crucial for the equitable implementation of programs aimed at detecting eligibility for LCS.

Current lung cancer screening (LCS) models^1,2 emphasize identifying eligible patients as a component of program success.^3-5 Because critical information such as patient smoking history is frequently incomplete or inaccurate within electronic health records (EHRs), identifying and informing eligible patients of LCS guidelines pose a challenge for health care systems.⁶

Growing access to cellular phones and text messaging increases opportunities to reach people via low-cost communication channels. Previous research data show that 50% to 60% of patients are willing to receive text messages from their primary care providers, including information about cancer screening tests.⁷ Further, this channel of communication is a known cost-effective way to reach people to boost cancer-preventive behavior.^8,9 Nonetheless, to our knowledge, text message interventions have not been studied in the context of LCS.⁹

When implementing digital interventions, it is crucial to proceed with caution and proactively assess acceptability. This is particularly important for individuals adversely affected by the digital divide,^10,11 which disproportionately impacts those from marginalized groups, including those who report older age, Black race, less total education, rural residence, or low income.¹² To better inform LCS implementation strategies, we considered whether text messaging could assist with identifying an LCS-eligible population, summarizing findings by sociodemographic characteristics associated with the digital divide.

METHODS

In December 2020, we conducted a cross-sectional anonymous online survey supported by Qualtrics. To recruit a sample representative of the US population, Qualtrics staff randomly sampled participants from their panels, which included individuals who voluntarily signed up to partake in surveys and receive cash or points in return. Participants had to be 18 years and older and reside in the US. Extending the study beyond the age of those who are eligible for LCS allowed us to explore and compare the opinions of both those who are eligible and not eligible for screening to ensure proper tailoring of future interventions. The survey items reported here were part of a longer survey that took 15 to 20 minutes to complete.

Respondents were informed about the study procedures, privacy, and confidentiality terms at the survey’s outset. The study was approved with a waiver of informed consent by the institutional review board at the University of North Carolina at Chapel Hill. Respondents reported their access to cellular phones, the frequency of their text messaging, and their likelihood of responding to a text message from their doctor’s office about their tobacco smoking status, which is critical information for determining LCS eligibility. Responses were obtained using a 3-point Likert scale (very likely, somewhat likely, or not at all likely). The questions were developed de novo by our team, which included experts in LCS and health care communications. Respondents also reported whether they had smoked 100 cigarettes in their lifetime, years since they had quit smoking, years of smoking, and the average number of cigarettes smoked per day. Survey questions are available in the eAppendix (available at ajmc.com).

Respondents were categorized as never having smoked if they reported smoking less than 100 cigarettes in their lifetime; individuals with affirmative responses to smoking 100 cigarettes in their lifetime were categorized as formerly smoked or currently smoke. LCS eligibility was determined among those who currently smoke or previously smoked.¹³ LCS-eligible respondents were aged 50 to 80 years, had a smoking history equivalent to 20 pack-years, and currently smoked or had quit smoking within the last 15 years. We also assessed respondents’ e-cigarette use, and based on their responses, we identified dual users of cigarettes and e-cigarettes.

We conducted χ² tests of independence¹⁴ and analysis of variance tests as appropriate to compare willingness to respond to text messages to ascertain eligibility for LCS and willingness to disclose smoking status by sociodemographic characteristics. χ² tests were conducted only when group size was 5 or more participants. We conducted the following comparisons: (1) those who currently smoke vs all other respondents, (2) dual users vs all other respondents, and (3) LCS-eligible respondents vs all other respondents. These populations were chosen because they are most likely to benefit from the targeted interventions for LCS and tobacco cessation. Younger individuals (< 50 years) not yet eligible for LCS were also included in this study because they could benefit from smoking cessation efforts. We removed cases with missing values from each analysis independently and reported the number of participants in each analysis.

RESULTS

Sample Characteristics

The survey was completed by 745 respondents. The mean (SD) age of the participants was 44.4 (18.0) years. Approximately 59% (n = 436) of the respondents identified as female, 38% (n = 286) as male, and 3% (n = 22) as another gender category. Reporting race, respondents were allowed to choose more than one option. Overall, 78% (n = 584) reported being White, 12% (n = 90) Black, 6% (n = 42) Asian, 4% (n = 31) American Indian, 1% (n = 8) Native Hawaiian, and 2% (n = 18) other; 14% (n = 101) identified as Hispanic. In total, 43% (n = 319) of respondents had a bachelor’s degree. More than a third of respondents (34%; n = 250) resided in urban areas, 46% (n = 342) in suburban settings, and 20% (n = 151) in rural areas. Income distribution showed that 30% (n = 224) fell within the low-income tertile (< $25,000 annually), 23% (n = 170) within the middle-income tertile ($25,000-$50,000), and 47% (n = 349) within the high-income tertile (> $50,000). The detailed observational statistics are reported in Table 1.

Over half the respondents (54%; n = 397) used text messaging daily, 36% (n = 271) used text messaging sometimes, 7% (n = 52) never used text messaging, and 3% (n = 24) did not have a cellular phone. Regarding smoking habits, respondents could fall into multiple categories. More than half the respondents (54%; n = 403) never smoked cigarettes, 33% (n = 245) currently smoked cigarettes, 13% (n = 96) reported past cigarette use, 11% (n = 82) reported vaping only, and 20% (n = 148) used both cigarettes and e-cigarettes. Approximately 6% (n = 48) were eligible for LCS. Among those who currently or formerly smoked (n = 341), 36 participants reported having been screened for lung cancer, 16 of whom were eligible for LCS screening. Additionally, among those who currently or formerly smoked, 26% (n = 88) had never heard about LCS.

The sociodemographic characteristics of the sample differed by smoking status, LCS eligibility status, and e-cigarette use (Table 2). Respondents who currently smoked compared with those who had never or formerly smoked were younger (mean age, 42 vs 46 years, respectively), less likely to be female (47% vs 65%), and more likely to use text messaging (93% vs 88%). Respondents who were eligible for LCS compared with those not eligible were older (mean age, 64 vs 43 years, respectively) and less likely to have a high income (27% vs 48%), including those who had never smoked or who currently or formerly smoked. Dual users of cigarettes and e-cigarettes were younger (aged 38 vs 46 years), less likely to report female gender (30% vs 66%), and less likely to live in rural areas (9% vs 23%) than others, including those who did not use cigarettes or used only e-cigarettes. Dual users were also more likely to have a college education (60% vs 39%) and a high income (62% vs 44%) than those who only smoked cigarettes or only vaped and those who neither vaped nor smoked.

Willingness to Disclose Smoking Status

Overall, 83% (n = 615) of respondents reported they would “likely” or “very likely” answer a text from their doctor’s office inquiring about their smoking status, 14% (n = 105) reported they were “not at all likely” to do so, and the remaining 2% (n = 25) did not answer the question. Willingness to report smoking status did not vary significantly by respondents’ smoking status: 84% (n = 323) who had never smoked, 88% (n = 214) who currently smoked, 84% (n = 78) who formerly smoked, 83% (n = 40) of LCS-eligible respondents, and 88% (n = 70) of those who used e-cigarettes only were willing to report smoking status via text message. Dual users were more willing to report their smoking status than all other participants (91% vs 84%; χ²₁ = 3.95; P = .047).

We found statistically significant differences in willingness to report smoking status via text message response by sociodemographic characteristics (Figure). A tertile of respondents with a mean age of 37 years were more likely to report their smoking status via text (91%; n = 150) than a tertile of younger patients with a mean age of 24 years (84%; n = 161) or a tertile of older patients with a mean age of 59 years (84%; n = 298; χ²₂ = 7.39; P = .03). Respondents were less likely to report smoking status via text messaging if they reported Hispanic ethnicity (80% vs 86%; χ²₁ = 4.09; P = .04), had no college education (83% vs 88%; χ²₁ = 11.42; P < .01), or had low income (81%) vs middle income (86%) and high income (88%) (χ²₂ = 9.85; P < .01). There were no significant differences by gender, race, and residential rurality.

DISCUSSION

Using a sample of adults from online panels, we explored the feasibility of text messages from providers’ offices to ascertain smoking status for LCS eligibility. More than 80% of respondents were willing to report their smoking status via text message. As such, text messages from providers’ offices could be a practical way to augment patient health records with information about tobacco use. This method could help overcome a barrier to LCS implementation and might further facilitate targeted smoking cessation efforts, particularly among subgroups like young men who smoke and vape. A more detailed analysis found critical differences in respondents’ willingness to communicate their smoking status via text message by sociodemographic characteristics, highlighting the potential for such practices alone to further augment present LCS inequities.

Marginalized populations—including older respondents, respondents of Hispanic origin, respondents without a college degree, and those with low incomes—were less willing to report their smoking status via text messaging. Potential reasons for these findings, at least in part, might be explained by a lower comfort level with technology and the absence of reliable cellular phones or data connections in marginalized communities.^15,16 For instance, it was shown that individuals with relatively low incomes are more likely to have disconnected cellular phones or data limitations.¹⁷ Furthermore, older patients in previous research reported having discomfort with technology and sensory or other loss associated with the aging process, which may hinder effective communication via digital devices.¹⁸

Notably, LCS-eligible respondents were more likely to be older and report a low income, and both of these factors were associated with a reduced willingness to respond to a text message inquiry from their doctor’s office regarding their smoking status. Although the study sample size prevented us from cross-tabulating results by current smoking status and respondents’ willingness to report their smoking status by sociodemographic characteristics, our results point to potential concerns that using a text message program alone might not comprehensively identify an LCS-eligible population. Moreover, those overlooked by such a program are likely to be individuals who could benefit most from LCS.

These observations are consistent with previous work demonstrating that individuals with low income and no college education preferred telephone communications with their doctor’s office and were less likely to choose communication via digital channels.⁷ Another study showed that patients from marginalized populations were more inclined to use their cellular phones for conversations rather than text messaging.¹⁹ Based on the previous findings^7,19,20 and our results, we recommend using text messaging to establish eligibility for LCS but emphasize the importance of integrating additional communication channels, such as telephone calls, mail, or in-person interactions, particularly when targeting older and marginalized populations. Further research should explore the optimal combinations of communication channels to maximize successful outreach to diverse populations for equitably establishing LCS eligibility.

The feasibility demonstrated in our study regarding the use of text messaging to assess smoking status suggests potential avenues for further exploration. First, researchers should focus on identifying optimal pathways for integrating text messages and patient EHR data. Second, the demonstrated willingness of participants to communicate via text messages in our study may warrant extending similar approaches in other contexts where details typically not contained within the EHR are needed to identify patient eligibility for service. While leveraging the affordability and acceptability of text messaging by patients, such programs should also consider known sociodemographic variations in patient communication preferences and employ multiple channels to ensure they reach marginalized populations who would otherwise be missed.

Limitations

This study relies on self-reported measures, which opens it to various biases. For example, due to the stigma associated with smoking, respondents might underreport current tobacco use.²¹ As such, our findings may reflect an overestimation of the ability to ascertain LCS eligibility via text messaging. Also, the study was conducted via an online survey, which presumably included those with a reliable internet connection and access to an electronic device for connectivity, as well as people with better technical literacy. Despite these limitations, we were able to observe significant variations in willingness to use text messaging among different populations. Additionally, the prevalence of smartphone access and use of text messaging reported among our sample is generally consistent with that reported by other studies that used nondigital means to assess the prevalence of access to smartphones.⁷

CONCLUSION

Although our results demonstrated that reaching patients via text messaging could be effective in identifying an LCS-eligible population, this approach should be complemented with telephone conversations or other modes of communication to ensure that marginalized populations who are eligible for LCS screening are adequately engaged in LCS-eligibility evaluation. 

Acknowledgments

ChatGPT 3.5 was used to assist with manuscript editing.

Author Affiliations: University of North Carolina at Chapel Hill Lineberger Comprehensive Cancer Center (JEL), Chapel Hill, NC; Cancer Prevention Precision Control Institute, Center for Discovery and Innovation (IF, LCB), Nutley, NJ; Henry Ford Health (CMND), Detroit, MI.

Source of Funding: This research was supported by funding from Lineberger Comprehensive Cancer Center.

Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (IF, CMND, JEL); acquisition of data (JEL); analysis and interpretation of data (IF, LCB, CMND, JEL); drafting of the manuscript (IF, LCB, JEL); critical revision of the manuscript for important intellectual content (IF, LCB, CMND, JEL); statistical analysis (IF); administrative, technical, or logistic support (JEL); and supervision (JEL).

Address Correspondence to: Ilona Fridman, PhD, University of North Carolina at Chapel Hill Lineberger Comprehensive Cancer Center, 450 West Dr, Chapel Hill, NC 27599. Email: Ilona_fridman@med.unc.edu.

REFERENCES

1. Lam S, Tammemagi M. Contemporary issues in the implementation of lung cancer screening. Eur Respir Rev. 2021;30(161):200288. doi:10.1183/16000617.0288-2020

2. Tabriz AA, Neslund-Dudas C, Turner K, Rivera MP, Reuland DS, Lafata JE. How health-care organizations implement shared decision-making when it is required for reimbursement: the case of lung cancer screening. Chest. 2021;159(1):413-425. doi:10.1016/j.chest.2020.07.078

3. Kee D, Wisnivesky J, Kale MS. Lung cancer screening uptake: analysis of BRFSS 2018. J Gen Intern Med. 2021;36(9):2897-2899. doi:10.1007/s11606-020-06236-9

4. Narayan AK, Gupta Y, Little BP, Shepard JO, Flores EJ. Lung cancer screening eligibility and use with low-dose computed tomography: results from the 2018 Behavioral Risk Factor Surveillance System cross-sectional survey. Cancer. 2021;127(5):748-756. doi:10.1002/cncr.33322

5. Yong PC, Sigel K, Rehmani S, Wisnivesky J, Kale MS. Lung cancer screening uptake in the United States. Chest. 2020;157(1):236-238. doi:10.1016/j.chest.2019.08.2176

6. Kukhareva PV, Caverly TJ, Li H, et al. Inaccuracies in electronic health records smoking data and a potential approach to address resulting underestimation in determining lung cancer screening eligibility. J Am Med Inform Assoc. 2022;29(5):779-788. doi:10.1093/jamia/ocac020

7. Fridman I, Smalls A, Fleming P, Lafata JE. Preferences for electronic modes of communication among older primary care patients: cross-sectional survey. JMIR Form Res. 2023:7:e40709. doi:10.2196/40709

8. Guerriero C, Cairns J, Roberts I, Rodgers A, Whittaker R, Free C. The cost-effectiveness of smoking cessation support delivered by mobile phone text messaging: txt2stop. Eur J Health Econ. 2013;14(5):789-797. doi:10.1007/s10198-012-0424-5

9. Uy C, Lopez J, Trinh-Shevrin C, Kwon SC. Sherman SE, Liang PS. Text messaging interventions on cancer screening rates: a systematic review. J Med Internet Res. 2017;19(8):e296. doi:10.2196/jmir.7893

10. López L, Green AR, Tan-McGrory A, King R, Betancourt JR. Bridging the digital divide in health care: the role of health information technology in addressing racial and ethnic disparities. Jt Comm J Qual Patient Saf. 2011;37(10):437-445. doi:10.1016/s1553-7250(11)37055-9

11. Cho Y, Yang R, Gong Y, Jiang Y. Use of electronic communication with clinicians among cancer survivors: Health Information National Trend Survey in 2019 and 2020. Telemed J E Health. 2023;29(6):866-874. doi:10.1089/tmj.2022.0203

12. Lythreatis S, Singh SK, El-Kassar AN. The digital divide: a review and future research agenda. Technol Forecast Soc Change. 2022;175:121359. doi:10.1016/j.techfore.2021.121359

13. Krist AH, Davidson KW, Mangione CM, et al; US Preventive Services Task Force. Screening for lung cancer: US Preventive Services Task Force recommendation statement. JAMA. 2021;325(10):962-970. doi:10.1001/jama.2021.1117

14. McHugh ML. The chi-square test of independence. Biochem Med (Zagreb). 2013;23(2):143-149. doi:10.11613/bm.2013.018

15. Reuland CJ, Godage SK, Wu L, et al. Information and communication technology access and use among low-income Latino immigrant parents. Matern Child Health J. 2021;25(12):1807-1813. doi:10.1007/s10995-021-03265-6

16. Estrela M, Semedo G, Roque F, Ferreira PL, Herdeiro MT. Sociodemographic determinants of digital health literacy: a systematic review and meta-analysis. Int J Med Inform. 2023;177:105124. doi:10.1016/j.ijmedinf.2023.105124

17. Gonzales AL. Health benefits and barriers to cell phone use in low-income urban U.S. neighborhoods: indications of technology maintenance. Mob Media Commun. 2014;2(3):233-248. doi:10.1177/2050157914530297

18. Mubarak F, Suomi R. Elderly forgotten? digital exclusion in the information age and the rising grey digital divide. Inquiry. 2022;59:469580221096272. doi:10.1177/00469580221096272

19. Kumar D, Hemmige V, Kallen MA, Giordano TP, Arya M. Mobile phones may not bridge the digital divide: a look at mobile phone literacy in an underserved patient population. Cureus. 2019;11(2):e4104. doi:10.7759/cureus.4104

20. Leigh JW, Gerber BS, Gans CP, Kansal MM, Kitsiou S. Smartphone ownership and interest in mobile health technologies for self-care among patients with chronic heart failure: cross-sectional survey study. JMIR Cardio. 2022;6(1):e31982. doi:10.2196/31982

21. Stuber J, Galea S, Link BG. Smoking and the emergence of a stigmatized social status. Soc Sci Med. 2008;67(3):420-430. doi:10.1016/j.socscimed.2008.03.010