Objective:
To assess the effect of digital monitoring on bleeding rates in patients with hemophilia A using prophylaxis.
Methods:
A total of 294 eligible patients with hemophilia A were included in our observational study. Eligible patients used clotting factor concentrates and had no active inhibitors. Patients used a digital health tool powered by MicroHealth to log bleeds and infusions via smartphone, texting, or online. The study observational period was August 2014 to January 2015. Patients using the tool could choose to invite their care professionals for monitoring. For each patient, Hemophilia Treatment Center (HTC) monitoring was defined as having at least one HTC professional: 1. Linked to the patient and with online access to the infusion logs; 2. Receiving notifications when the patient had bleeds and/or had adherence below a threshold, and; 3. Having two or more patients under monitoring. There were a total of 35 and 259 patients in the monitoring and non-monitoring groups, respectively.
We conducted Bayesian analysis using linear mixed models and a negative binomial distribution to compare the relative risk of bleeding rates for patients with and without HTC monitoring.
Summary:
Patients using HTC monitoring had a relative bleeding rate of 0.60 vs. patients without monitoring, which is equivalent to a 40% reduction in bleeding rates for monitored patients (95% credible interval: 0.38—0.96).
Patients on the monitoring and no monitoring groups were comparable except that the monitoring group had 23% more pediatric patients (p<0.001). However, bleeding rates between pediatric and adults were comparable (p=0.500). Subgroup analysis showed no differences in the reductions of bleeding rates due to monitoring between pediatric-only and adult-only subgroups (p=0.353).
Conclusions:
The use of digital tools for chronic care monitoring is a growing global trend.
A reduction in the annualized bleeding rate of 40% (~2 bleeds a year) is both statistically and clinically significant and may have a cumulative protective impact on patients’ long-term outcomes. Observational studies are subject to sample bias, however, patients in both groups were technologically savvy and motivated enough to track their condition using a digital health tool. Given that this intervention is free, safe, and fits the accountable care model, we encourage clinicians to explore its adoption. Confirmatory studies on this topic are encouraged.