Optimizing patient-centered interventions using digital health for repeat lung cancer screening in Washington state

Abstract

Lung cancer is the leading cause of cancer mortality in the United States. Annual screening with low-dose computed tomography (LDCT) reduces lung cancer mortality by 20%. However, annual adherence to lung cancer screening (LCS) remains low, ranging from 22-55% nationally. The Larch Trial tested two multilevel digital health interventions embedded in the EHR patient portal to improve LCS adherence at Kaiser Permanente Washington (KPWA) from 2022-2024. Nested within the Larch trial, this dissertation sought to understand mechanisms of action and gaps in reach of multilevel digital health interventions using both quantitative and qualitative data. We identified factors at the patient-, clinician- and system-level associated with access to the EHR patient portal and engagement with intervention portal messaging (Aim 1). We qualitatively explored perceptions, attitudes and needs related to digital health and technology in LCS-eligible adults (Aim 2). Finally, we conducted an exploratory analysis to understand the relationship between knowledge and tobacco-related stigma with LCS adherence (Aim 3). In an integrated health system in Washington state, we found that current tobacco use, reduced length of enrollment in KPWA and lower income measured at the census-level were associated with no patient portal access. The same factors, along with older age, demonstrated similar association with no engagement with patient portal messaging (Aim 1). In interviews with LCS-eligible adults (aged 51-75 years), we found that this population expressed self-efficacy and comfort using technology, specifically leveraging the patient portal for health management. They also expressed pervasive concerns with data privacy and security that impact their engagement with digital health tools even when delivered from the health care provider and insurance company (Aim 2). Lastly, while we did not observe a detectable effect for tobacco-related stigma, we found that knowledge about when to return for annual LCS is associated with increased likelihood of LCS adherence (Aim 3). These studies provide valuable information on structural and behavioral factors that may influence patient access to the EHR patient portal, engagement with digital health, and adherence to repeat LCS. Together, these findings provide valuable insight on multilevel factors that shape lung cancer screening behaviors, providing actionable areas of intervention for more effective and equitable cancer prevention.