Associations between survey completion mode and sociodemographic factors among individuals eligible for lung cancer screening

TL;DR

This study found that how people prefer to complete surveys for lung cancer screening research is linked to their age, race, and language, suggesting that offering only online options might exclude some groups.

Contribution

The study identifies sociodemographic factors influencing survey mode preference in lung cancer screening research.

Findings

Mail was the most common survey mode, especially among non-White participants and those at OUHSC.

Older age and Spanish language preference were linked to phone survey completion.

Offering only online surveys may reduce the diversity of participants in LCS research.

Abstract

To enhance representation in LCS research, we examined associations between participant characteristics and their preferred mode of survey completion among individuals eligible for LCS.

Between February 2023 and April 2024, prospective participants were identified using electronic healthcare records from Mass General Brigham and the University of Oklahoma Health Sciences (OUHSC). We offered three modes of survey completion: online, by mail, or by phone. Eligible participants were 50–80 years old, had smoked within the past 15 years, and spoke English or Spanish. We used multinomial logistic regression to estimate relative risk ratios (RRR).

Outreach to 2,822 individuals resulted in a sample of 315 participants. The mean age was 61.7 years (SD = 10.9). Most respondents were women (63.0%) and identified as White (63.3%), 29.8 % were Hispanic. The most common survey completion mode was mail (37.1%), followed by online (35.9%) and phone (27.0%). Characteristics associated with completion by mail were study site (RRR = 6.86, 95%CI:3.10–15.14), and race (RRR = 3.63, 95%CI:1.53–8.61); with respondents at OUHSC or who did not identify as White being more likely to choose mail over online modality. Characteristics associated with phone completion, included older age (RRR = 1.11, 95% CI: 1.03–1.20), Spanish language preference (RRR = 9.28, 95%CI:2.38–36.09), and with local government or community insurance (RRR = 9.91, 95% CI:1.92–51.3).

The current trend toward online surveys may not fully account for individual preferences for LCS research engagement, and could limit the representativeness in LCS studies if offline alternatives are not offered.

Figures