Assessing the Effectiveness of Interactive Robot-Assisted Virtual Health Coaching for Health Literacy and Disease Knowledge of Patients with Chronic Kidney Disease: Quasiexperimental Study

Abstract

Background: Chronic kidney disease (CKD) imposes a significant global health and economic burden, impacting millions globally. Despite its high prevalence, public awareness and understanding of CKD remain limited, leading to delayed diagnosis and suboptimal management. Traditional patient education methods, such as 1-on-1 verbal instruction or printed brochures, are often insufficient, especially considering the shortage of nursing staff. Technology-assisted education presents a promising and standardized solution, emphasizing the need for innovative and scalable approaches to improve CKD-specific knowledge and health literacy.

Objective: This study aimed to develop and evaluate the effectiveness of an innovative 12-unit virtual health coaching program delivered through interactive robots that is intended to enhance disease knowledge and health literacy among patients with CKD.

Methods: A quasiexperimental design was used, and 60 participants were evenly assigned to experimental and comparison groups. However, due to attrition, 14 participants in the experimental group and 16 participants in the comparison group completed the study. The intervention involved a 12-unit program, with each unit lasting approximately 20 minutes to 30 minutes and delivered across 3 to 4 learning sessions, and participants completed 3 to 4 units per session. The program addressed key aspects of CKD-specific health literacy including functional, communicative, and critical literacy and CKD-specific knowledge including basic knowledge, prevention, lifestyle, dietary intake, and medication. Data were collected through validated pre and postintervention questionnaires. All 30 participants completed the program and subsequent evaluations, with outcome measures assessing changes in CKD-specific knowledge and health literacy.

Results: Postintervention analysis using generalized estimating equations, adjusted for age, revealed that the experimental group (n=14) had significantly greater improvements in health literacy (coefficient=2.51, Wald χ²₁=5.89; P=.02) and disease knowledge (coefficient=1.66, Wald χ²₁=11.75; P=.001) than the comparison group (n=16). Postintervention t tests revealed significant improvements in CKD-specific health literacy and disease knowledge (P<.001) between the experimental and comparison groups. Additional analyses identified significant group × time interactions, indicating improvements in communicative literacy (P=.01) and critical literacy (P=.02), while no significant changes were observed in functional literacy. Regarding disease knowledge, the experimental group demonstrated a significant improvement in medication (P<.001), whereas changes in basic knowledge, prevention, lifestyle, and dietary intake were not significant.

Conclusions: This study demonstrated that interactive robot-assisted eHealth coaching effectively enhanced CKD-specific disease knowledge and health literacy. Despite the challenges posed by the COVID-19 pandemic, which constrained sample sizes, the findings indicate that this program is a promising patient education tool in clinical nephrology. Future research should involve larger sample sizes to enhance generalizability and examine additional factors influencing effectiveness.

Keywords: chronic kidney disease; disease knowledge; eHealth; health coaching; health education; health literacy; interactive robot.

Figure 1

Flowchart of the study. CKD: chronic kidney disease.

Figure 2

Interface example of the interactive robot-assisted virtual health coach program for chronic kidney disease (CKD). RN: registered nurse.

Figure 3

Comparison of pre and posttest (A) CKD-specific health literacy and (B) CKD-specific disease knowledge, reported as the mean (95% CI) scores, with comparisons of pre and postintervention scores between the groups.