Related Experiment Videos
Prompted Text-Based Vital Sign Recording Versus Unprompted Electronic Medical Record Entries in Patients With
Jenny Wu1, Julia Chen1,2, Lillian Min1,2
1Department of Internal Medicine, University of Michigan, 1500 E Medical Center Dr, Ann Arbor, MI, United States, 1 734-936-4000.
Background:
Long-term remote patient monitoring of weight, pulse, and blood pressure has been shown to significantly reduce mortality and hospitalization rates among patients with heart failure. Despite its proven effectiveness, maintaining patient engagement in remote monitoring programs remains challenging.
Objective:
This study aimed to evaluate the impact of 2-way text-based communication on prompting patients to record key vital signs and to compare it with unprompted patient reporting through electronic medical records in terms of engagement and clinical outcomes.
Methods:
We analyzed data from patients participating in the University of Michigan Advanced Heart Failure Program who reported daily weight, blood pressure, and pulse using either the MiChart Patient Outreach Texting Application (MPOTA) or patient enrolled flowsheets (PEFs). The study's primary metric was the consistency of patient-reported vital signs, with secondary descriptive metrics including variations in hospitalization and emergency room visits pre-enrollment and postenrollment in the programs.
Results:
A total of 890 patients were included, with 301 enrolled in the MPOTA group and 589 in the PEF group. The engagement rate for the PEF group had a median of 2.29% (IQR 0%-23.93%). In contrast, the MPOTA group showed a significantly higher median engagement rate of 66.67% (IQR 30.67%-88.24%). There were no significant differences in hospitalization or emergency room visit rates across engagement categories (none, low, medium, and high) or between programs. Mean hospitalizations declined by 21% in the MPOTA group (from mean 0.53, SD 0.90 to mean 0.42, SD 0.88; P=.06) and 18% in the PEF group (from mean 0.50, SD 0.86 to mean 0.41, SD 0.80; P=.03) after the initiation of each program. This reduction was small and statistically significant only for the PEF group. Mean emergency room visits did not significantly change in either group. Regression analyses showed no significant association between engagement level and hospitalization or emergency room utilization, although medium engagement was associated with a nonsignificant trend toward fewer events. Despite improved engagement with MPOTA, this did not translate into significant reductions in hospitalizations or emergency room visits. All analyses of clinical outcomes were exploratory and underpowered, and no significant associations were found between engagement level and utilization.
Conclusions:
Although MPOTA was associated with substantially higher patient engagement levels compared to unprompted patient reporting, neither demonstrated significant differences in hospitalization or emergency room visits across engagement levels or between programs. Small reductions in hospitalizations were observed, but these were significant only in PEF, not in MPOTA. Observed changes in utilization were exploratory, small in magnitude, and underpowered to detect clinically meaningful effects. These findings suggest that mobile text-based communication may be a useful tool for improving engagement in remote monitoring programs for patients with advanced heart failure; however, further research is needed to assess its impact on clinical outcomes.