Over the past few years, mobile health (mHealth) technology has advanced at a rate of knots. From wearable fitness trackers to sleep tracking gadgets, these apps and devices have become an ingrained part of everyday life. Fitbit alone sold more than 13 million devices in 2015.
This technology is also making inroads within healthcare. As the burden of chronic diseases continues to grow, ever more patients are using home monitoring devices and ever more doctors are showing an interest in telecare solutions.
Unfortunately, a large pool of applications remains untapped. For example, mHealth has yet to break into clinical research. While the possibilities seem exciting – in essence, being able to keep an eye on study subjects remotely – the need for regulatory compliance severely complicates the task.
Over the past year or so, however, the situation has begun to turn around, with pioneers in the field addressing the question: might mobile sensors and wearables enable a new template for drug development, which is not only patient-centric but also saves time and reduces costs?
Piloting mobile health tech for clinical research
Medidata, a cloud solutions and data specialist working with life sciences companies, has long been interested in using mHealth within clinical trials. As part of its efforts in this field, the company sponsored a pilot open-level trial, MOVE-2014, which aimed to find out whether mHealth tools could be used to drive better health outcomes in subjects with type 2 diabetes. Subjects were equipped with a smartphone and a Fitbit activity tracker, as well as a mobile app called Patient Cloud. Over an eight-week period, the company tracked the effects, ultimately demonstrating that these tools could be used with great success.
"The key thing we were looking for was, is it possible to run a clinical trial with a high level of rigour, including a lot of information gathering outside of the investigative site, in a non-traditional setting?" says Medidata’s managing director of mHealth Kara Dennis.
With that goal in mind, the team found the trial highly informative on a number of different metrics.
"We were able to train the site in how to instruct subjects to put on the wearable devices and use an app on their phone," says Dennis. "We had very high compliance with the use of the app and the activity tracker, and we felt the data we gathered from both those tools was comprehensive and high quality."
Not only were the subjects more than 90% compliant in using their trackers, but the subsequent data analysis threw up some interesting results. For instance, there was a strong correlation between activity levels (as measured on the Fitbit) and self-reported pain. This suggests trial designers might be justified in using activity levels as a proxy for pain, giving some quantifiable grounding to a subjective measure.
However, because this trial was fairly small, these kinds of relationships were not the primary point of focus. The really exciting conclusion, as Dennis sees it, is that mHealth technology and wearables might be reliably incorporated into the drug development process. And while these tools would primarily be useful from a data collection point of view, providing a greater insight into patients and their response to therapies, there would be a number of additional advantages.
"Gathering data outside of the investigative site reduces the need for subjects to have to go to the doctor’s office, so that could help with recruitment," Dennis explains. "It could mean you could recruit patients from a more expansive geographic radius, maybe running completely remote trials, which means you can likely recruit more quickly. There’s also potential that you could retain subjects – you wouldn’t have as many dropouts."
There would also be scope to save money at investigator sites. Currently, investigator expenses comprise a high proportion of the overall costs of a clinical trial. Amassing patient data remotely, without needing to involve a doctor, could drastically cut down on the overheads.
What’s more, real-time data collation could lead to faster and more thorough clinical insights, allowing unsuccessful drug candidates to fail more quickly and successful ones to take their place. Because you have an instantly available biomarker in the form of mHealth data, it would be easier to introduce an adaptive trial design.
Of course, while these opportunities speak for themselves, they are unlikely to come to pass without a thorough assessment of the challenges. While collecting mHealth data is reasonably straightforward, doing so in a compliant way is not.
"We build all our software to a set of regulatory compliant standard operating procedures – so all the ways we manage data, and ensure privacy, security and anonymity throughout the trial, those are based on 15 years of deep expertise and knowhow in regulatory compliance," explains Dennis. "All our mHealth software conforms to the same principle. It’s how you manage and protect and oversee data – we’re very immersed in how to do that in a regulatory compliant way."
She concedes there is an additional challenge when it comes to clinical trials, namely that the data needs to be fit for purpose.
"Does it meet basic standards of comprehensiveness, accuracy, attributability? Have you gathered all the data, have you attached the correct device data to the correct subject, is the device reliably producing consistent signals in response to consistent stimuli? Regulatory agencies are really interested in understanding whether this data is valid as a clinical observation, and that requires a lot of work in terms of understanding the relationship between mobile health data and other endpoints," Dennis explains.
For Medidata, this is a key point of exploration. Together with its life sciences clients, the company is working to develop some answers. It is also looking to cultivate a better understanding of the device landscape, maintaining a dialogue with regulatory agencies and educating sensor manufacturers about the specific requirements of mHealth trials.
It seems clear that this dialogue will become more salient in future, as mHealth becomes ever more ubiquitous and its benefits impossible to ignore.
"There are a number of different areas where we’re starting to see a lot of demand," says Dennis. "So from an operational perspective we’re just getting ourselves ready to do a lot of this work in 2016."