Digital health Shutterstock.com
After absorbing several years of increasingly extravagant promises about the remarkable potential of digital health, investors, physicians, and other stakeholders are now unabashedly demanding: “Show me the data.”
By now, most everyone appreciates the promise of digital health, and understands how, in principle, emerging, patient-focused technologies could help improve care and reduce costs.
The question is whether digital health can actually deliver.
A recent NIH workshop, convened to systematically review the data on digital health, acknowledged, “evidence is sparse for the efficacy of mHealth.”
As Scripps cardiologist Eric Topol and colleagues summarized in JAMA late last year,
“Most critically needed is real-world clinical trial evidence to provide a roadmap for implementation that confirms its benefits to consumers, clinicians, and payers alike.”
What everyone’s asking for now is evidence – robust data, not like the vast majority of wellness studies that experts like Al Lewis and others havedefinitively shredded.
The goal is to find solid evidence that a proposed innovation actually leads to measurably improved outcomes, or to a material reduction in cost. Not that itcould or should, but that it does.
Applications ranging from GPS-enabled asthma inhalers to cloud-based EMR services to iphone-based medical diagnostics typically highlight the transformative potential of these technologies, and often emphasize the exceptional value said to reside in all the collected data. It might even be true – but let’s see the evidence.
Given the striking disparity between digital health’s breadth and depth, between the sheer number of health-oriented apps and the data that even a handful of these products really do something substantial, it’s not unreasonable to presume digital health gadgets are, at best, amusing wellness devices — wellutainment — until proven otherwise.
However, before we get too self-righteous in our critique of the digital health evidence base, we might take a moment to recognize how fragile the data are for much of what we do in medicine. The evidence for the utility of digital health devices may be weak to non-existent, but it’s not much better for a startling number of medical tests and procedures.
A major study published by Vinay Prasad and colleagues last year, for example, found that about 40% of established medical practices failed to stand up to scrutiny when deliberately studied. In an accompanying editorial, legendary Stanford statistician John Ioannidis observed, “the introduction of interventions with limited or no evidence of benefit continues at a fast pace,” adding,
“Once we divert beyond traditional treatments (eg drugs or devices) to diagnostic tools, prognostic markers, health systems, and other health care measures, randomized trials are a rarity. For example, it has been estimated that, on average, there are only 37 publications per year of randomized trials assessing the effectiveness of diagnostic tests.”
The results of such careful testing can often be surprising; for example, for years, it was standard practice to utilize a pulmonary artery catheter (PAC) to optimize the care of critically-ill patients, before randomized controlled studies demonstrated the intervention was generally of little value. A recent “obituary” for the procedure concluded, “there is no evidence that the use of the PAC has improved patients outcomes.”
The real issue, then, isn’t whether digital health is being excessively scrutinized: we unquestionably should continue to demand evidence for efficacy and impact. But we – entrepreneurs, investors, care providers, stakeholders — also should recognize what an incredibly high bar this represents, a bar many traditional and long-established medical interventions and approaches would likely fail to clear.