Digital Health Promises Remain Unfulfilled for High-Need, High-Cost Populations

Published on: 

This week, companies are showcasing digital health at conferences in Las Vegas and San Francisco, but when it comes to alleviating disease burden or the cost of care for vulnerable populations, the promise of digital health is not yet fulfilled, according to a study published in Health Affairs.

This week, companies are showcasing digital health at conferences in Las Vegas and San Francisco, but when it comes to alleviating disease burden or the cost of care for vulnerable populations, the promise of digital health is not yet fulfilled. That’s according to a study published Monday in Health Affairs, which found that of the studies conducted on digital health products and services, most enrolled healthy volunteers. Few enrolled high-burden, high-cost patients, said the authors.

“We both think digital health holds really great promise to address some of the major challenges in the healthcare system, including providing higher quality of care, lowering costs and increasing greater access to care,” said Kyan Safavi, MD, MBA, the lead authors of the study, in an interview with The American Journal of Managed Care®, along with coauthor Adam B. Cohen, MD. “As physicians we wanted to know whether digital health companies have demonstrated an impact on healthcare in the United States in patients who hold the highest cost and burden of disease,” he said.

Safavi, Cohen, and the other coauthors chose 20 digital health companies for their analysis, based on the amount of their private equity funding. The companies had median funding of $67.5 million versus $5.3 million for the rest of the industry, and tended to have a larger number of employees.

They then sought to evaluate how rigorously the companies’ products and services were evaluated and published in peer-reviewed journals in patients with high-burden conditions and using a high degree of evidence, using the criteria of the US Preventive Services Task Force.

The majority of the 104 studies indexed in PubMed (75, or 72.1%) did not target patients with a high-burden condition or risk factor, while only 29 (27.9%) did. Healthy volunteers made up 32% of the studies, followed by patients with amyotrophic lateral sclerosis (14%) and those with multiple sclerosis 12%. Mental health was the most common high-burden condition category studied, with 8 studies. There were no studies for lung cancer or smoking.

Sixteen, or 15%, of the studies assessed clinical effectiveness. Eight percent assessed the clinical effectiveness of the product or service in a high-burden or high-risk-factor population. There were no clinical effectiveness studies in mental health conditions, hyperglycemia, depression, heart disease, chronic obstructive pulmonary disease, trauma, or low back pain.


Cohen, a neurologist, said both he and Safavi, an anesthesiologist, were particularly interested in what the clinical impact of these new technologies were having in terms of cost and access.

Looking at the subject from a clinical perspective, Cohen said they wanted to get a sense of “what digital health could provide, where there’s evidence, and where there’s open space…to help define where we think things are now, and in some ways to highlight where we think things will go and where they should go.”

“We think we’re in a validation phase of digital health,” said Cohen, who is the health technologies program manager at the Johns Hopkins University Applied Physics Laboratory, and is an assistant professor of neurology in the Johns Hopkins Hospital and Health System.

Companies and providers are basing their success on measures such as volume, and not as much on disease prevention and improving clinical outcomes, he said.

Many of the companies studied provide direct-to-consumer (DTC) or wellness products, said Safavi, the David F. Torchiana Health Policy and Management Fellow at the Massachusetts General Physicians Organization and an instructor at Harvard Medical School in the Department of Anesthesia at Massachusetts General Hospital.

Ideally, in the next generation of studies, researchers would be able to find “a reduction of the incidence of chronic disease,” through the use of such products in the areas such as diabetes, obesity, stroke, heart disease, and others, said Safavi. Moreover, the studies need to be performed in patients who have such diseases. Lastly, providers would want to see improvement in patient outcomes, and health systems would want to see measures released to cost and access to care, he said.

However, in order for these changes to happen, policy makers need to adjust regulations and use incentives to create a stronger market, the authors said. FDA regulation may fundamentally differ between DTC products marketed to the healthy, and aimed at the chronically ill. Another factor that could spur more evidence-based, high-value digital health products is the trend toward value-based purchasing, the authors said.

But for now, the demonstrated impact of these companies does not yet exist, Safavi said. Compared to traditional therapies and services, digital health has clear advantages, as the development time is not nearly as long and it can be iterated quickly, he said.

“Digital health is relatively cheap by and large,” said Safavi. “ And at scale, it is still relatively cheap.”

Some of the digital health companies offered products in multiple categories. By category, the largest was analytics, at 5, followed by consumer health engagement, 4; biosensors and care coordination, 3 each; population health management, telemedicine, and social networking, 2 each; and personal health tracking, mobile medical apps, payer administration, and digital medical devices, 1 each.

Using the example of activity trackers, Cohen and Safavi said there would need to be more evidence-based data that they can reduce obesity, and in turn, related diseases, such as heart disease or stroke. But to get there, there are a number of unanswered questions. Do patients consistently use the trackers? Does the product work as intended to prevent a condition, such as obesity? If obesity is prevented, does the incidence of co-morbid cardiovascular diseases drop?

“To me, that’s the opportunity here,” said Safavi.


Safavi K, Mathews SC, Bates DW, Dorsey ER, Cohen AB. Top-funded digital health companies and their impact on high-burden, high-cost conditions. Health Aff (Millwood). 2019;38(1):115-123. doi: 10.1377/hlthaff.2018.05081.