Researcher’s Guide to Wearables

Apps for Ecological Momentary Assessment are rapidly changing the manner in which psychology and behavioral health research is conducted. Close behind this wave of development is the use of wearable technologies in the same fields. This is a guide for researchers wishing to augment self-report EMA data with objective data collected from wearable devices and engage in some form of ambulatory assessment. Researchers are currently using wearable devices to capture a range of health metrics including, activity, blood pressure, heart rate and HRV, sleep, skin conductance and EEG (almost!). All of these can be integrated with mobile EMA (mEMA) measures There are three problems you will encounter with commercial wearables regardless of which metric you are attempting to capture, ilumivu is working to provide solutions to these:

  1. No raw data: you will find it next to impossible to get raw data from many commercial companies, you will instead get data processed through their algorithms (e.g. “steps”, “stress”, “calories burnt” or “hours of sleep”) without gaining access to the actual measures that went into these calculations and certainly not to the algorithms themselves. The accuracy of many of these processed metrics is questionable and devices can vary widely on metrics with the same label.
  2. Coarse temporal granularity of the data: most companies will only give you access to the summary data per day so you may know how many steps your participant took that day but have no idea how these “steps” were distributed throughout the day so you can not synchronize these data with the self-report mEMA data they are simultaneously providing.
  3. No aggregated data sets: the revenue model of the companies providing these devices gives them no incentive to provide an API (an expensive undertaking) that would allow you to capture aggregated data sets from a group of participants. If you are providing devices to your participants and can wait until you collect them all back to access the data then no problem. If you are hoping for real-time insight into participants activity or want to use these data to trigger mobile EMA (mEMA) surveys then it’s not going to work.

ilumivu is working around these barrier by finding devices to which we can connect directly bypassing the restrictions of companies that are focused on consumer sales so that we can utilize these devices for research purposes. Currently we are integrating Smartwatches running Android Wear. This includes devices from numerous manufacturers including Motorola, Polar and Wahoo. The Android Wear platform allows us to write code directly on the device and to pair the device with a Smartphone so we can perform calculations and  send notifications to either device. For instance, we can capture the raw RR Interval data from a watch, calculate HRV on the device and use it to trigger mEMA alerts on the phone when HRV passes a pre-specified threshold. Read more about activity tracking and heart rate monitoring integrated with ecological momentary assessment.