About that meta-analysis of mHealth efficacy studies

By: Brian Dolan | Jan 22, 2013        

Tags: | | | | | |  |

Brian Dolan, Editor, MobiHealthNewsTechnology moves too quickly for the world of academic publishing. That’s a long held belief by many but also a common topic of conversation among those working in digital health today. How to best study efficacy can be a tricky subject for digital health, however, the argument that efficacy data is unimportant rarely enters the debate — it is, especially for those companies that hope to convince payers that their services are worthy of reimbursement. Of course, the benefits of rigorous study go well beyond that in many cases, too.

There are some digital health studies, however, that are best published elsewhere. Any sensible person would agree that a scientific journal is no place for them. Take meta-analyses of mHealth efficacy data. Are you looking to determine whether the scientific evidence supporting clinical efficacy of mobile health was robust back in mid-2010? Well, you just had to wait until January 2013 to find out — in fact, you probably did last week.

The fact that the scientific body of efficacy data was lacking as of September 2010 should come as a surprise to no one. A quick poll of those mobile health companies with published efficacy studies informed me that the time it takes from when an mHealth service is ready to begin an RCT to the actual publish date of the study in a medical journal is at least two years. That might be appropriate for some intervention studies, but what good is a meta-analysis report on the state of current affairs if it arrives two years later?

As my colleague Neil Versel wrote last week: “British researchers examined 20 years’ worth of published studies on mobile interventions delivered to healthcare consumers and mHealth technologies intended to improve care delivery processes. In both cases, they were underwhelmed with what they found.”

The 20 years, of course, stretched from January 1990 to September 2010. Given the typical two years (minimum) it takes to get to the publish date, that means the study was only looking at mHealth interventions developed by mid-2008. That pre-dates MobiHealthNews by a few months. Apple launched its AppStore around the same time these studies were just beginning. At that time, Matthew Holt and Indu Subaiya had hosted just one Health 2.0 event and a single Spring Fling!

So, yes, scientific mobile health research was found to be severely lacking as of September 2010. Since then, much has happened, including the publication of WellDoc’s landmark efficacy study in 2011. In a comment on our post last week, Wireless Life-Sciences Alliance strategic advisor Jack Florio noted: “This field has been evolving with lightning speed. It is likely that studies which were started in 2009 and 2010 have not reached the literature. This means there is an additional 5 years of past and ongoing work which is not accounted for in this research.”

He’s right, of course, and on that note, a much more current meta-analysis found that the future of mHealth efficacy data is bright.

Alain Labrique, Ph.D., the founding director of the Johns Hopkins University Global mHealth Initiative, and his team recently conducted a survey of the mHealth studies that were registered in the government’s clinicaltrials.gov database as of November 2012. Labrique writes: “We identified 215 unique mHealth studies that were registered in the clinicaltrials.gov database, of which 8.4 percent (n=18) were observational in nature while the remaining 91.6 percent (n=197) were interventional. Of the 215 studies, 81.8% (n=176) studies used a classical randomized trial design and 40 new studies were added to the database between May and November 2012 alone. Based on these results, we posit that the field is entering a new ‘era’ where a body of rigorous evaluation of mHealth strategies is rapidly accumulating.”

Labrique concludes: “The transition into an era of evidence-based mHealth supports our position that innovation in this domain can be evaluated with the same rigor as other public health strategies, attenuating some of the hype previously associated with mHealth.”

We’ll continue to cover that transition in near real-time here at MobiHealthNews.

  • Jack J Florio

    I fully concur with Alain Labrique’s, final comment above. This is a premise that the industry must support.

    However, I come from an industry of pharmaceuticals and medical devices where the product being evaluated was static and the studies took a long time evolving through several phases of safety and efficacy. Even when final results come in, there was little that could be done to change the product.

    mHealth however is very patient centric and highly flexible. The technology can be adapted to the individual patient circumstance and interaction to improve the desired outcome. The technology can also be adapted along the way to enhance efficacy.

    As an industry, we need to band together and work to define new clinical trial designs and parameters for mHelath that keep the rigor of science while incorporating the flexibility.
    It is in this way that we will be able to unlock all of the value that this new paradigm that connected health technology has to offer to improve health outcomes in the short term.