Digital Health Platform Blood Pressure Management Study

Overview

This pilot study evaluated the effectiveness of an AI health education assistant compared to case manager support in hypertension management. Key outcomes included changes in systolic blood pressure (SBP) and diastolic blood pressure (DBP), compliance with health behaviors, 722 goal achievement (monitoring, measurements, lifestyle improvements), and Technology Acceptance Model (TAM) metrics, including perceived usefulness (PU), perceived ease of use (PEU), and behavioral intention (BI). This study aims to examine whether the AI assistant group will achieve greater reductions in systolic and diastolic blood pressure compared to the case manager group. Additionally, it will evaluate whether the AI assistant group will demonstrate higher engagement in 722 goals and greater perceived ease of use. The results of this study are expected to provide further evidence supporting the potential of AI-driven interventions.

Background: This pilot study aimed to evaluate the effectiveness of an AI health education assistant compared with case manager support in managing hypertension. The primary focus was on changes in systolic blood pressure (SBP) and diastolic blood pressure (DBP), compliance with health behaviors, engagement in 722 goal achievement (monitoring, measurements, and lifestyle improvements), and Technology Acceptance Model (TAM) metrics, including perceived usefulness (PU), perceived ease of use (PEU), and behavioral intention (BI). This study aims to examine whether the AI assistant group will achieve greater reductions in systolic and diastolic blood pressure compared to the case manager group. Additionally, it will evaluate whether the AI assistant group will demonstrate higher engagement in 722 goals and greater perceived ease of use. Methods: This two-arm, randomized controlled pilot study enrolled participants aged 18-75 years with hypertension. Participants were allocated to either the AI health education assistant group or the case manager group. The AI group received app-based interventions, including medication reminders, behavioral nudges, educational quizzes, and abnormal blood pressure alerts powered by AI-driven monitoring. The case manager group received personalized lifestyle advice, medication reminders, and regular follow-ups from human case managers. Blood pressure readings were collected at baseline and 6 months, and TAM metrics were assessed through questionnaires. Compliance with health behaviors and engagement in 722 goals were evaluated through app interaction logs. The sample size was determined based on a non-inferiority design comparing two independent groups. The primary outcome was the change in systolic blood pressure (SBP), which was defined as the post-intervention value minus the pre-intervention value. A one-sided significance level of 0.05 and a statistical power of 80% were assumed. The sample size calculation used a pooled standard deviation of 11.6 mmHg derived from the pilot study. A non-inferiority margin of 3 mmHg and an expected mean difference of 1.34 mmHg between the AI intervention group and the control group were also assumed. The required sample size was estimated to be 604 participants per group. After allowing for a 30% dropout rate, 863 participants will be required in each group, resulting in a total sample size of 1,726 participants for the two groups. Statistical analyses included paired t-tests for within-group changes, independent t-tests for between-group comparisons, and linear regression models to assess associations between intervention type and TAM metrics, adjusting for baseline values. Discussion: This study will provide preliminary evidence on the feasibility and potential benefits of AI-driven interventions for hypertension management. It will further evaluate the AI assistant's ability to improve blood pressure control, increase engagement in 722 goals, and enhance technology acceptance. These findings will deliver important insights for clinicians and inform future clinical practice.