Industry survey delves into wearable device use

While adoption of decentralized trials has exploded in recent years (especially in the face of the global pandemic), industry professionals still have a number of questions about the technology, including wearable health sensors. In the recently released second volume of its Device Census Report, Litmus Health offers a comprehensive look at the wearables field, through the eyes of clinical researchers.

Litmus Health chief medical officer Samuel Volchenboum (SV) spoke with Outsourcing-Pharma (OSP) about the results of the survey, and the story the report might tell about the future of decentralized trials.

OSP: How has the decentralized trial field evolved in recent years?​

SV: I think the singular biggest change has been the increased acceptance and even desirability of the decentralized trial. While there used to be a lot of skepticism around the quality of data that could be collected in a decentralized trial, the advent of wearables and the increased usage of smartphones in trials has meant that clinicians no longer have to sacrifice data quality for trial accessibility; in fact, more pharmaceutical companies are interested in utilizing wearables than ever before.

Part of this increased adoption is due to the fact that wearable technology has also rapidly advanced. Now numerous devices are able to capture more advanced metrics — for example, blood oxygen, VO2 max, PPG, and even electrodermal activity. More importantly, many of those devices are no longer prohibitively expensive.

Also, the culture around these devices have changed. We’ve seen this effect accelerated due to the pandemic. People are increasingly interested in learning about their health metrics and are willing to participate in studies that track them.

OSP: Could you please share in a nutshell why your company decided to conduct this survey?​

SV: Two years ago, we realized that, despite pharma’s increasing adoption of real world data, there was no comprehensive review of wearables for the clinical field. Yes, there were multiple consumer guides comparing different devices, but they lacked information relevant to a pharmaceutical audience; as a result, every clinician and researcher we spoke with was essentially starting from scratch when it came to device selection.

We wanted to address this gap in the market and truly investigate these devices through the lens of the clinical trial. This included research around sensors and functionality, like battery life; however, it also meant honing in on device makers’ approaches to documenting data transformations and accessibility of research-grade APIs and SDKs.

Unsurprisingly, we found that while numerous wearables provided this information, it’s often buried or near impossible to find. By developing a comprehensive and demystified report that compiles this information, we sought to empower researchers and clinicians to easily determine which devices would be best suited to a study.

We were pleased that this first volume report was a huge success back in 2018. However, as you might expect, the market has evolved considerably in the past two years.

If anything, device makers’ sensors and handling of data has gotten even more complex. We saw the need for an updated report that dug deep into today’s device leaders — particularly since the field has narrowed considerably.

OSP: What are the key differences between Vol. 1 and Vol. 2?​

SV: The main differences between Vol. 1 and Vol. 2 are depth and scope. In Vol. 1, we surveyed nearly 200 different devices and a wider field of device makers. From there, we chose our top list of devices.

However, in Vol. 2, we reviewed less products and device makers, but took a deeper dive into each one. Part of this is purely a result of the industry narrowing. The field of device makers producing wearables suitable for clinical research is smaller than it was in 2018.

Some companies profiled in the previous report are either now defunct or have pivoted more towards the consumer wearable space over potential clinical applications.  

In short, some clear leaders have emerged, and we wanted to take the time to really hone in on what differentiates these top players. Many of these leaders also now have multiple devices suited to the clinical environment, so we wanted to take the time to profile multiple devices per device maker, when relevant.

We also wanted to further emphasize and explore device transparency. Readers will see that in this report, we’ve really sought to give as clear of a picture as possible to what it would be like to work with a particular device maker while running a trial.

OSP: How did you determine which devices and brands to cover?​

SV: A large contributor in our decision on which brands to cover was research presence. There are a few device makers which are already pretty well known in the clinical space. Actigraph and Activinsights, for example, make devices specifically for the usage of clinical trials, and combined are present in thousands of studies that utilize wearables.

Other brands like Apple, Fitbit, Garmin, and Polar, while mainly consumer focused, are also used in a considerable amount of clinical studies. We also wanted to cover device makers that we feel have the potential to be a larger player in the future, even if their clinical presence is currently limited.

When deciding which devices to review, we had a list of criteria that needed to be met in order for us to recommend a device’s use in a clinical trial. We required the device to be worn somewhere on the body (as opposed to a patch or environmental sensor), to capture data continuously, and to, at minimum, collect heart rate data.

OSP: What are the most interesting or useful findings?​

SV: We believe the main takeaway is this: right now, there is no perfect wearable. Some devices are technologically advanced, yet there is still lack of access to minimally transformed data. Other device makers with a high transparency score haven’t advanced as quickly as we’d like to see in a technological sense. Detecting heart rate is a minimum, yet we are capable of so much more than that now.

Another interesting thing we found is there’s currently a huge emphasis around heart metrics. Apple, Samsung, and Fitbit have all scrambled to get FDA clearance for their ECG sensors to detect Atrial Fibrillation (AFib). In particular, Apple just wrapped up a cardiology study with over 400,000 patients that focused on AFib.

Finally, we found that the pandemic drove more consumer-focused brands to finally venture into the clinical space. For example, OURA has recently been in the public eye given its new partnerships with the NBA and NASCAR to monitor player health — something that likely wouldn’t have occurred without COVID.

OSP: Were there any surprises?​

SV: It’s been encouraging to see how device makers have evolved their offerings over these past two years to create more transparency around their data collection principles and approach to data transformation. Because this is a topic of conversation that is almost entirely absent from the consumer market, it would be easy enough to deprioritize improvements on these fronts. However, we’ve seen quite the opposite — the majority of device makers in the report achieve at least a medium score for data transparency.

However, it is interesting and surprising to see that considerations for transparency have not yet reached the hardware side. As we state in the report, the embedded sensors in these devices often come from unknown and varied manufacturers.

Just as researchers and clinicians must examine data legacy, so too must they be aware of hardware lineage. I think it’s fair to say that while the importance of raw data is slowly becoming accepted, the additional importance of transparency about a sensor’s origin remains an overlooked challenge in the industry.

OSP: How do you expect or hope people in the research field will use this information?​

SV: We hope researchers and clinicians can look to this report when deciding which device will be best for their clinical trial. By providing a comprehensive guide that reflects the concerns the pharma industry has when incorporating real world data, including device price, sensors, and data transparency, we hope to support the ongoing adoption of wearables in clinical trials.

OSP: Any plans for future surveys, in this field or others?​

SV: In time, we plan on releasing a third volume of the device census report in the future that continues to focus on device transparency. With new players coming into the wearable space such as Amazon, we know it is important to continue to iterate on our research and document the changes in this industry.