Health data Apple Watch more unreliable than expected
Apple uses algorithms to analyze health data from the Apple Watch. These algorithms are regularly updated. According to a Harvard professor, this is not good for reliability.
Read on after the ad.
Health data Apple Watch more unreliable than expected
A big advantage of the Apple Watch is the extensive software support that Apple is known for. Apple typically improves the ease of use, security, and features of its products through software updates.
However, in the case of medical research, the regular update schedule is not necessarily a good thing. That’s according to recent research by JP Onnela, associate professor of biostatistics at the Harvard TH Chan School of Public Health. Algorithm changes can cause inconsistencies in health data from the Apple Watch.
The same heart rate variability data was collected at different times during the study. These came from one Apple Watch. Although the data was collected at different times, the data should logically be identical. However, this was not the case.
Algorithm changes are the culprits
According to Onnela, the problem lies in adjustments to the algorithms that Apple uses to analyze medical data. “These algorithms are what we would call ‘black boxes’ – they are not transparent. It is therefore impossible to know exactly what is in such a box”.
The changes in the algorithms are attracting critical eyes from scientific researchers. Small differences in health data may not be a problem for the average consumer. For science, however, inconsistency is a disaster. Accuracy is essential to arrive at reliable conclusions.
Apple changes the algorithms regularly without any warning. In addition to heart rate variability, researchers have found similar inconsistencies in data from the sleep app.
Want to always be up to date with the latest Apple news? You can do this via the website and our free iPhoned app. You can also register for our newsletter.
