Researchers’ Top Tips for First-Time Wearable Users
Wearables can offer a lot of benefits to your research, but they also come with unique challenges. To help you stay on top of them, we asked our customers to give their top tips for future researchers using wrist-worn wearables for the first time.
1. Run a pilot
Dr. Ann Hsing of Stanford learned a lot of lessons during her WELL For Life Wearable Pilot Study and one of the biggest was the importance of an initial pilot. Before purchasing all of the devices for your participants, run a thorough pilot in advance to check whether the collected data is what you need and to better understand the participant experience. You’d be surprised how many people get excited and decide to snatch up the devices first.
Researchers recommend buying a refurbished device or using one you already own and trialing it on a platform’s free version. Make sure you and your team become very familiar with the watches and apps. Be as meticulous as possible, walking through both the researcher's and the participant's experience.
When piloting your study, look closely at the exported data files to make sure the data is exactly what you want. This allows you to ensure you’re collecting the correct metrics, in the format you need, at the sampling rate that best supports your study’s objectives.
2. Make protocols and onboarding instructions as detailed as possible
Dr. Sandra Webber from the University of Manitoba didn’t run into many difficulties during her Long Covid study which she mostly attributes to the detailed protocols she and her team wrote.
Since participants will not likely be coming into the lab regularly (if at all), clarity is necessary when using wearables for remote data collection. If participants don’t know precisely what they need to do and when, you and your research assistants could be facing a barrage of emails and calls. More importantly, you run the risk of low adherence and missing or inaccurate data (some third-party platforms offer an adherence dashboard and task tracking to help rectify this).
3. Have resources in place to help participants remotely
One challenge Dr. Kristin Scott and her team faced in their study on healthcare worker resiliency was that some participants weren't as tech-savvy and required a bit more support than expected. To counter this, Kristin used videos and PowerPoints to support participants remotely. Since adherence depends on familiarity with tech and comfort with new devices, she recommends having support resources in place for participants who may be unfamiliar with these devices.
This is also where the importance of Tip #1: Run a Pilot comes back into play. Should any confusion arise in the pilot, you’re able to build out the necessary support resources to prevent the situation from occurring on a larger scale.
4. Collect an additional day of data
One valuable trick from researchers with past experiences using accelerometers is collecting an additional day of data. The reason is that people often change their behavior when they know they're being monitored, such as becoming more active if their steps are being tracked (you’ve probably heard about the ‘Hawthorne effect’ if you’ve ever taken an introductory psych course). The good news is that most of the research has shown that people don't keep that up for more than a day or two. So if you want data that is representative of a participant’s normal week, collect data for eight days and drop the first day.
5. Only collect the metrics you need
Although tempting, it’s important not to get caught up with all of the data wrist-worn wearables can provide. Smartwatches offer you tons of different metrics to collect, but not all are necessary for your study. After studying the correlations between subjective and objective wellness measures in runners, Colten Brand’s advice is to keep your study streamlined and focus on the data you need for your study objectives. Otherwise, you might end up with an overwhelming amount of data to analyze.
This leads us to our final tip…
6. Consult with analysts who specialize in big data
Using wearables in research has made it easier than ever to collect physiological data. However, interpreting the data from these bands can be a whole other story if you’re not familiar with it. Once again we’ll bring up Tip # 1 (we told you it was important) because running a pilot and carefully inspecting what your exported data will look like should not be overlooked.
As Jenn Edelschick from Duke discovered during her cystic fibrosis research, you get a lot of data from the devices and it can often be overwhelming to consolidate if you don’t have a full-time statistician on your team.
Before beginning your study, check to see if the device company or third-party platform you’re interested in has an analytics service to help organize, analyze and report all of that data back to you in a usable way. This will allow you to save months of time and energy on analysis while making the most out of your data.
While we can’t exactly be impartial, our data analysts have the receipts to show they’re some of the best. You can consult with our analytics team about your study or use our analytics reports to save you months of time.
Labfront's analysts understood our needs as researchers and saved us weeks of work when they prepared our datasets for analysis. - Bryan Edwards, Oklahoma State University
And there you have it, the top recommendations from those who have ventured into the world of physiological data collection. We hope you keep these tips in mind when launching your own study.
Have you used wearables in your research and have a tip of your own to share? Get in touch with us at email@example.com.
Alix doubles as the marketing and pun specialist at Labfront. She usually operates quietly behind the scenes, but give her a karaoke mic and all bets are off.