Background/Purpose: The COVID-19 pandemic accelerated telemedicine adoption, yet readiness and digital literacy vary widely across patient populations. Understanding these differences is crucial to designing equitable and effective telehealth systems for chronic disease management.

Methods: Latent class analysis (LCA) was performed on 138 RA patients enrolled at National University Hospital, Singapore using 18 items from a Telemedicine Readiness Survey [1] and the eHealth Literacy Scale [2]. Age, sex, race, education, employment, and marital status were not part of the clustering but were compared. Cluster solutions were evaluated based on model fit indices and interpretability.

Results: We identified four distinct clusters ranging from those who were digitally disengaged and resistant to telehealth to those who were confident and ready adopters of remote care (Figure 1).Cluster 1 (C1) (n=26, 16.3%) was digitally disengaged. Patients in this cluster demonstrated the lowest levels of telemedicine engagement and digital confidence. 95.8% reported lacking confidence in accessing or using online health information. Half of the participants found telemedicine too complicated.Cluster 2 (C2) (n=38, 23.8%) had moderate digital literacy and occasional telemedicine use, and remained ambivalent due to concerns about communication quality during teleconsultations. Approximately 65.8% expressed difficulty or dissatisfaction with communicating effectively during virtual consultations.Cluster 3 (C3) (n=38, 23.8%) was open to remote care but demonstrated low digital self-efficacy. 38.0% reported active use of telemedicine but 97.3% lacked confidence in navigating online health information effectively. This cluster reflected a mismatch between willingness to engage with telemedicine and the digital skills required to do so.Cluster 4 (C4) (n=36, 22.5%) demonstrated high telemedicine literacy, minimal difficulty using digital tools, and strong preference for remote care when appropriate. Notably, 94.4% agreed that time and cost savings positively influenced their decision to use telemedicine, and 97.2% preferred remote consultations over delays in care. This cluster exemplified a confident, proactive user group capable of navigating digital healthcare environments.Demographic comparisons revealed differences between clusters. C1, the oldest group (mean age: 68.0±10.4 years), had the highest proportion of participants with only primary or no formal education (50.0%). In contrast, C4, the youngest group (mean age: 51.5±15.7 years) had the highest proportion of participants with tertiary education (74.1%). Additionally, English proficiency differed across clusters, ranging from 69.2% in C1 to 94.4% in C4 (p=0.038).

Conclusion: Four distinct patient clusters were identified, reflecting a spectrum of telemedicine readiness. Cluster membership was primarily shaped by patients’ digital confidence and eHealth literacy. Our findings support targeted efforts to boost digital literacy and confidence, especially among older adults and those with less formal education.[1] Gurupur V et al. Health Informatics Journal. 2016;23(3):181-196[2] Norman CD, Skinner HA. J Med Internet Res. 2006 Nov 14;8(4):e27

Figure 1. Canonical discriminant analysis of the 4 clusters and their group centroids. The plot shows the clear discrimination of the 4 clusters based on the discrimination functions 1 and 2.

*Between-cluster differences were assessed using the one-way ANOVA for continuous variables and the chi-square test for categorical variables.

*Between-cluster differences were assessed using chi-square test for categorical variables.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/identifying-patterns-of-telemedicine-readiness-and-digital-health-literacy-in-rheumatoid-arthritis-patients-a-latent-class-analysis/