This study comes on the heels of several other recent studies looking at the use of more modern connected health tools for other chronic health conditions. The Wired for Health study from Scripps basically gave patients a suite of connected health devices and a portal to review the data; no other significant support was included. That study found no difference in healthcare utilization. Another Finnish study involving around 500 patients added health coaching about every six weeks to the mix for patients with diabetes and hypertension, also finding no benefit in quality of life or metrics like blood pressure or HbA1c.
BEAT-HF was conducted at several academic centers in California and randomized patients hospitalized with decompensated heart failure to either a remote monitoring arm or usual care. Nearly 1,500 patients were randomized, with patients in the intervention arm receiving connected health devices including a weight scale, blood pressure monitor, and a texting device they could use to report symptoms.
Prior to discharge they received extensive coaching on heart failure as well as guidance on how to use those devices to send in daily reports. That data was sent to a centralized monitoring station where nurses were alerted if there were changes that exceeded pre-specified thresholds. They also contacted the patient at every few weeks just to catch up and discuss heart failure management.
At 6 months, the readmission rate was essentially identical at around 50% in both arms. There were also no significant differences in mortality which was a secondary endpoint.
A theme among all studies was adherence. In BEAT-HF, about half of patients dropped off in both the remote monitoring and the follow up phone calls within the first thirty days. That’s similar to what was seen in both the Wired for Health Study and the Finnish study. It’s also similar to what was reported in the Asthma ResearchKit study, where response to messaging prompts dropped precipitously in the first few weeks. In BEAT-HF, this adherence issue persisted despite calls from the nurses when patients weren’t sending in data.
For use in heart failure, there are several reasons that the authors offer for their negative findings. First, they suggest that maybe the metrics we’re using – weight in particular – is just too late of a signal, pointing to data related to the CardioMEMS implantable pulmonary artery pressure monitor.
The other theme across these studies was that aside from having patients collect the data daily, these were pretty “low touch” interventions. The outreach to the patient was infrequent (once every few weeks) or nonexistent. Perhaps when it comes to chronic diseases that our patients are having to actively manage every single day, going from a quarterly clinic visit to a monthly phone call (or phone calls triggered when bad things are already happening) isn’t enough.
Perhaps more proactive, high frequency interventions that leverage a mix of automated feedback mechanisms and educational messaging in addition to periodic virtual or phone based visits and alert-triggered contacts is needed. Certainly results from smoking cessation studies as well as SMS studies using high frequency messaging suggest those strategies may have value.
Or maybe it’s the one-size-fits-all approach that many of these RCTs, including this one, using digital health technology take. One strength of this technology is that there are a multitude of tools and strategies available, with varying levels of complexity, data granularity, and demands on the patient. Perhaps a more patient oriented approach in using the right tools for the right patient is needed.
Whatever the answer, studies like these are important steps towards determining how we can use technology to better engage patients in their own health management and improve outcomes.
References: MedPage Today