Understanding Patient Behavior Through Wearable Sensors

Read Transcript

I think this is one of the more exciting things in behavioral science in the since we've done behavioral work I think. If you think about how we've traditionally done this work someone has to come into our office. We have to see them on a regular basis, we see them once a week for an hour, the rest of their life we have no idea how to change it, how to modify it.

We are not monitoring it we are not providing any sort of impact into it on a daily basis. The mobile and wireless technology has allowed us to basically reach into peoples lives, and help them change their behavior in real time. A lot of the things that we weren't able to do in the past just because we can see into peoples behaviors over the course of the day.

We're able to do much more in terms being able to actually adjust, and adapt our interventions to what they doing in the dietary area, we're able to sort of monitor much better now what they eat, how they eat against the camera sort of technologies now people also take a picture of their food.

This is what I've eaten for the day, so we have some sensor there. We can fairly quickly sort of get back and sort of correct sort of misperceptions they have about the type of foods that they are eating, how much they are eating, and make those changes in terms of feedback. We are now able to track people's physical activity levels, especially sedentary behaviour just sitting for more than a half an hour.

Being able to prompt people when they're doing that, being able to just to move and do other things. In the smoking area, being able to actually sort of monitor and track smoking behavior, and then be able to schedule it and break up the normal sort of cues that they have for smoking, and then gradually reduce that over time until they're more able to quit.

Medication adherence as well, so not just adherence to behavioural interventions, but the basic sort of medical adherence. It's another place where the field has really had a lot of traction, able to be able to remind people in real time when they're supposed to be taking their meds.

Track that electronic pill boxes that allow us to be able to sort of follow people. Even if they are like complicated sort of regimens that they're taking across the multiple diseases. We do all this work in sort of better meds, and at the end of the day, we know there's still on average about only half of the people take their meds the way they're prescribed anyway.

So, being able to use some of these technologies to increase that compliance, adherence to their medication regimens is actually a very useful part of this. But, one of the things that we know doesn't work is sort of the burdening people to death with monitoring. So, that we are asking them to just provide an awful lot of of information in large amounts of time that they just they quickly start burn out from that.

So, we've had to back off on the amounts of monitoring we ask people to do, but it still has to be an act of process, right? We can't just passively sought of do things for people it still has to be something they are actively engage them. But I think one of the biggest sort of areas of failure for us is that we keep forgetting that the whole reason that we evolve with all this learning principles is so we can control the environment.

Not so the environment could control us. And so we always have to think about these applications as ways to assist a person in making the changes that they want to make. Not imposing on them changes that they probably don't really have much interest in. So, we have to take them from where they are a little bit more than we've done in the past.