Wednesday, 24 June 2015
Apple ResearchKit: When Users Become Participants
In March this year, Apple announced a new open source software framework called ResearchKit, which plans to turn the hundreds of millions of iPhones into a powerful tool for collecting data for medical research. The idea behind the framework is simple: by using the iPhone’s built-in sensors, and the growing number of personal trackers that can be connected to the iPhone (e.g. FitBit, glucose monitors), researchers can gather activity and biometric data from participants who have given their consent to be part of their research study.
Collecting data in this way means medical research is no longer limited by geographical location; participants don’t need to travel to a hospital or University to take part in a study, meaning data can be collected more frequently and at lower cost. Furthermore, the ease with which the data can be collected (automatically in some cases) will likely reduce dropouts and participants lost to follow-up appointments, as well as making studies less reliant on inaccurate self-reports.
But with an increase in the quantity of data, there is likely to be a compromise in the quality of the data. The activities are conducted in less controlled environments, meaning the data will be more noisy. There is, for example, nothing to stop a participant giving their iPhone to somebody else to complete. Demographic bias is also a significant problem, with studies showing iPhone users are more likely to have a graduate degree and have higher income than your average Android user.
Five apps have already been built that target some of the most costly medical conditions in the world: asthma, diabetes, Parkinson’s disease, cardiovascular disease and breast cancer. The early sign-up rates for these apps have been unprecedented, with Stanford University announcing that signed up for their cardiovascular disease study in the first 24 hours. Whether this sign-up rate continues once there are more apps competing for participants remains to be seen, but it is worth noting that these apps have only been released in the US so far.
It is early days for ResearchKit, and there are plenty of questions that need to be answered regarding the quality of the data collected. However, Apple’s decision to make the framework open source means anyone can contribute to the framework and even develop it for other platforms to further extend its reach. This means the future of ResearchKit, and its contribution to medical research, is in the hands of the developers and researchers who use it.