In the not-too-distant future, we will all be living in an ongoing clinical trial, where sensors track our lives and stream our health data directly to our electronic health records
Hindsight, as they say, is 20/20; it's predicting the future that's tough. Yet, when it comes to new technologies, there is rarely a shortage of sibyls, soothsayers and seers willing to stand up and prophesy. Yet, at best, forecasts of the future offer a direction of travel – after all who really foresaw the incredible expansion of the worldwide web or the stunning uptake of smartphones?
However, sometimes prognostications make so much sense it is difficult to deny them. One such prediction involves the beguiling idea of continuously monitoring our health. Using an array of wearables, sensors and monitors to track our every waking – and sleeping – moment, such technology offers a tantalising glimpse at our own individual health future.
This biotelemetry revolution-in-the-making sits at the convergence of a number of trends, not least the rise of the smartphone and the powerful push towards patient empowerment. What's more, the technology is being field-tested right now in another market – the wellness space – where Fitbits, Moovs and Jawbones are strapped to millions of wrists measuring heart rate, sleep patterns and activity levels.
Using such devices in a clinical setting, however, inevitably raises concerns prompted, in part, by sensational news headlines, most recently around the US class-action lawsuit facing Fitbit concerning the accuracy of its heart rate data.
The resulting inertia means that adoption of biotelemetry in clinical trials has been uneven and slow, say Daphne Kis and Dr Samuel Volchenboum, co-founders of Litmus Health, a data science platform for early-stage clinical trials.
"Part of the problem has been the devices used in clinical trials; they tend to be highly developed, one-use devices that are very expensive and have a closed data ecosystem," says Volchenboum, who by day is director of the Center for Research Informatics at the University of Chicago. "People are finally waking up to the idea that a set of consumer devices that people can bring with them can do just as good a job."
He points to the data generated by activity trackers: "These devices have been used for a long time by people who want to narcissistically track how many steps they've taken or how well they slept. No one thinks these devices will provide a perfect measure but most agree they provide a good measure.
"All measurements have error associated with them. The current gold standard for determining how well a patient slept is a survey written in the 1980s called the Pittsburgh Sleep Quality Index, which asks patients to recall how well they slept over the past month, so you're subject to accuracy issues and potentially antiquated questions," says Volchenboum.
The solution, says Kis, is to cross correlate the data with other sources. "We are running a trial at the moment that tracks how well people with Crohn's and ulcerative colitis sleep using Fitbit sensors but the next morning we also ask the patients how they slept. Cross-correlating device data collected passively with the person's subjective experience helps us to develop an extremely powerful and persistent tool."
For Volchenboum, clarity is needed between the ideas of real-time data and 'at the point of experience' measures. "You don't need a dashboard where you watch your patients sleeping but, because of the problems associated with human recall, collecting data while the patient is experiencing the event is incredibly powerful. Also, being able to feed that back to the physician in 'semi real-time' allows them to intervene if, say the dashboard shows which patients take their pills on time and which are in a lot of pain. This will help to optimise and speed up clinical trials."
A key challenge is to standardise data collection, he says. "Pharma is using these technologies in their trials in their own ways – this means we are plagued by a lack of standards across the industry which leads to difficulties with interoperability. The result of pharma's reluctance to standardise is that, even though the FDA has made it clear that they want to see these innovations in clinical trials, there is no clear understanding yet on how it will accept the data collected on devices.
"The problem is that there is neither a carrot nor a stick right now. Without clear direction from the FDA, many groups are just collecting data the old-fashioned way then transforming it into whatever format is needed right before submission. However, if we can change the way they collect data so they're collecting it in a standardised format, it will make it far easier to share and consume data. I'm hoping that, as an industry, we can rapidly develop better metrics that will be acceptable to regulatory agencies."
Privacy and data security are major issues, says Kis. "The question of how open, transparent and available these data are is becoming a key approval driver for the FDA. We're going through our own process to ensure we've built privacy and security protocols into our products from the outset because not only does it give us a competitive advantage but we know it is going to be required."
Patients also want control of their data, says Volchenboum. "A crucially important issue is that people will rightfully insist on owning and having access to their data. The idea of having clinical data in one silo and trial data lives in another and their pharmacy information elsewhere will not be acceptable. People will also have control over who else can access their data, which will open up avenues for collecting and sharing data that we can barely even conceive now."
In 5-10 years' time, Kis believes we will all be living inside one giant ongoing clinical trial. "A huge shift in healthcare is that people are starting to see that disease and health are not black and white, there is often a spectrum. This understanding affects personal decision-making; for example, a patient will want to know what impact a drug will have on their entire health, will want to know the impact of participating in a trial or not. We increasingly live in a culture in which people are involved in discussions about their health in a way that they weren't ten years ago."
Volchenboum agrees. "We throw around terms like 'quality of life' and 'vitality' but we don't really have good ways of quantifying how a person feels and how they feel compared to another person. In both sickness and in health, we will develop new ways to look at patients and how different factors contribute to their wellness, whether that's how they slept or how they move or whatever.
"Innovation in medicine is becoming faster and faster and our ability to deliver personalised, precision care is increasing too. With clinical trials specifically, the days when you could do a five-year study with two arms is over. Nowadays you're six months into a trial and a novel therapy comes out. We have to be more innovative in how we do trials, how we enrol patients and a key part of that is how we collect data."