Professor Brian Caulfield

Insight Wellbeing: Your fitness tracker and you – Prof Brian Caulfield on the pros and cons of wearable sensors

Submitted on Tuesday, 16/01/2024

Do wearable sensor devices help or hinder your January goals? Professor Brian Caulfieldon the ins and outs of using trackers for sport or wellness

January is when we spend more than we can afford on gym membership and runners. In recent years, fitness trackers have pushed other products aside. According to Sports Ireland, over 50% of Irish adults report using sensing devices to measure their physical activity.

With the rise of the fitness tracker has come a rise in research into these ‘consumer wearables’ that track metrics such as steps, speed, distance, energy expenditure and heart rate. Research is showing that the accuracy of over-the-counter fitness trackers is uneven, varying from very accurate to wildly inaccurate depending on the specific device and the measurement target.  Beyond the quality of the sensor itself and the algorithms used to process the signal, there are many variables that can interfere with a sensor, ranging from how well the user applies it to the body to skin colour.

Research on wearable tracking adherence has shown that between 30 and 70 per cent of users abandon their fitness trackers after a few months. There is a ‘ditch your wearable’ discussion  emerging among elite athletes that is starting to filter down to recreational fitness enthusiasts.

Researchers in France have posited the idea that, for some participants, putting numbers on fitness recasts it as ‘forced labour’ rather than leisure. In other words, the pleasure of, say, running for its own sake is replaced by a distracting fixation on monitoring and numerical achievement.

So, is it all bad, and should you leave your new wrist ornament in the box? Well, it’s not that simple.

The relationship between a human and their tracker falls at the intersection between sports science, behavioural psychology and technology. Trackers can have a beneficial effect on adherence to fitness regimes, even when they are not entirely accurate, and even if the wearer only finds them supportive for a limited time.

The success of your relationship with your wearable starts with defined goals. If your goal is to get out running a few times each week and cover an increasing distance over time, the tracker might well be the running partner you need. Even if the distance readings you get are not precise (as they may not be where there is interference with the GPS signal, as in a city), they will still give you a sense of your progress over time and a feeling of achievement that will keep you motivated.

If your goal is to have better sleep in 2024 and you are using a tracker to assess your sleep quality, it’s probable that your own assessment of your night’s sleep will be at least as valuable as the tracker results. Wearable devices rely on sensors that read the body’s movements during the night. Not all data is recorded and the results you get are based on an algorithm that gives an estimate of your sleep quality.

However, the tracker is a sort of co-pilot on your mission to get better sleep which will likely result in you implementing a more optimal regimen, improved sleep routines etc. We all need a guardian angel sometimes.

As a long time runner, I use a wearable tracker but I ignore the majority of the metrics that it provides to me. A large part of the value for me is the social element – I use the data as part of my Strava membership, which links me in with a running community. Being part of a fitness community is highly motivational. I’m quite relaxed about the accuracy because it’s not really the point.

However, when I wear my researcher hat in the Insight Centre, accuracy is a concern.  The Insight SFI Centre is part of a European initiative called INTERLIVE which serves to develop best practice guidelines for validation of wearable sensors, and reporting the results of that validation. If manufacturers are making claims about the accuracy of their devices, consumers need to have transparency around the basis for those claims. If, for example, the accuracy of the product’s heart rate readings is based on a study of young white males under sedentary conditions, that level of accuracy might not extend to users from outside that category.

Accuracy of readings is not life-or-death when it comes to a morning workout but sometimes it matters, particularly when it comes to health management. Anyone embarking on an exercise regime who also has a heart problem may be relying on their heart rate readings to report anomalies to their doctor. If that reading is algorithmically derived  instead of being directly measured by the sensor due to a poor signal quality then it is important that the user and their clinician understand this when interpreting the data.  Manufacturers need to report this alongside the HR data.

This is why best practice validation and reporting models are badly needed, so that consumers can make realistic comparisons between various products on the market.

If motivation is what you need, a fitness tracker might be just the thing. Data overload is a consideration however, so be judicious about what you track. Some measures, like ground contact time, for example, have little value for most use cases. Others, like distance and time or metres climbed per week, can actually help you to build a sense of achievement, even if the margin of error is significant.

Bear in mind that also, because of the AI algorithms they use, these devices will often get more useful with time as they learn more about you. They have their uses, just don’t set your watch by them.