Initial research experiences with the validation of wearables

I started my PhD research several months ago and already feel like there's not enough time to implement my ideas. Setting up research studies takes a lot of time! But the journey is moving along nicely. In this post, I'll share my progress, the validation of the first wearables, the bumps in the road, and my initial experiences in the field.

In order for my research to progress, I first have to determine which wearables are capable of wireless patient monitoring in and outside a hospital setting. At the moment, I'm busy setting up a comprehensive validation study with two wearable patches for the medium care and nursing departments. "What does good validation entail? Trust me, it's a lot of work."

What does good validation entail? Trust me, it's a lot of work.

What does good validation entail?

The first step is to ask yourself what you want to measure and in which situations. For my PhD research, I wanted to determine the added value of using a remote, wireless monitoring device to measure vital signs in post-surgery patients with the aim of identifying deterioration at an earlier stage. My research focused on patients in nursing units and patients who were sent home to recover after surgery. A study like this requires a comprehensive validation approach that takes into account several factors.

I started with non-mobile patients requiring intensive care and constant monitoring on a medium care unit. When validating different vital signs, it's important to realize that you're working with both normal and abnormal values, such as an extremely high heart rate and respiration rate.

I hope to obtain more reliable measurements when the patients become mobile (e.g. after they're transferred to a nursing unit). Once home, patients should be able to move freely and we still hope to receive reliable measurements, albeit less often.

Medium care measurements

This calls for a step-by-step validation process. We are using patches to measure the vital signs of patients in medium care units. These measurements are then compared to the current standard – a bedside monitor that takes continuous measurements. 

Erik Huizinga, a master's student in Technical Medicine, is helping me with this process. He is researching whether unusual vital signs can be detected and how these relate to the available information we have on a patient. This can include the vital signs reported by nurses and suspected complications that require further diagnostics, such as sepsis or pneumonia.

Previous findings on medium care units

I recently had the opportunity to spend a morning with the nursing staff at a medium care unit. The first thing I noticed was the excellent quality of care they provide! 

"The word 'wireless' is not really part of the lexicon here, despite the fact that most nurses would welcome the option with open arms." I also noticed the logistical challenges regarding the implementation of patient wearables. On this unit, patients are being continuously monitored and are covered in tubes, drips, and other measuring devices. The word "wireless" is not really part of the lexicon here, despite the fact that most nurses would welcome the option with open arms. And here I am adding to the measurement load with wearables.

Patients no longer require constant monitoring on the nursing unit and therefore have fewer wires and tubes. This makes it easier for me to test out some of my wearables on the patients. However, in order to validate these wearables, you need to think about implementing a temporary yet continuous reference standard to determine the accuracy of the measurements. And they're not used to continuous measurements like this on the unit.

Bumps in the road

And there are other bumps in the road. At the moment, I'm facing the following challenges. 

"While some companies claim their patches can monitor vital signs for three continuous days, this isn't always the case in practice." While some companies claim their patches can monitor vital signs for three continuous days, this isn't always the case in practice. The patch may not stay put on a sweaty patient, the battery may die, or data may not be sent due to a Wi-Fi connectivity problem. There are plenty of reasons for intermittent data loss. What about the patches that are ripped off accidentally when a patient needs acute care!

And then there's the task of retrieving the data from the wearable. Some wearables are linked to an iPad and a secure cloud, while others have to be installed on the hospital network. These things take time and effort!

Of course, I'd love nothing more than to validate wearables based on a golden standard, but on the road from intensive care to at-home recovery we need to consider other solutions that may be more practical to implement. 

Capnography, for example, has become the golden standard for measuring respiratory frequency. This is the best and least invasive method for sedated patients in a controlled surgical setting. But it's almost impossible to use on alert, mobile patients that require monitoring over an extended period of time. 

In short, getting wearable measurements is a real challenge!

Wealth of information

Nevertheless, this extensive validation is certainly worth the effort. Not only will it help us find a reliable wearable that can monitor patients remotely on the road from intensive care to home recovery, it will give us a wealth of information that can help us implement this new healthcare innovation. Input from enthusiastic nurses, doctors, and patients is an absolute must!   

If you have ideas or suggestions for important aspects I should include in the validation process from ICU to home, let me know! I'd love to get in touch via Twitter or LinkedIn.

As a Product Researcher at FocusCura, I straddle two worlds: I focus on advanced home healthcare and I carry out scientific research as a PhD candidate at Utrecht University Medical Center. This blog offers updates on my PhD research, "A Safe Path from ICU to Home" and research developments in the field of cVitals.