In a wide-ranging interview with Mobihealthnews, the Center for Connected Health’s Founder and Director Dr. Joseph Kvedar recently discussed the center’s various wireless health programs, belief in text messages for healthcare, assessment of the likely early adopters of wireless health services and the need to not lose sight of the challenges facing this emerging industry.
Mobihealthnews: While many readers may already be familiar with the Center for Connected Health at Partners, can you give a brief overview of the Center’s work and raison d’etre?
Kvedar: Sure, our goal is to provide care to patients where the patient is and when the patient needs it, which is to say: outside of traditional care settings like at work or in the home place. This is very consistent with the vision of mobile applications since they too follow the patient. We also have a second goal, which is to give the patient or the consumer the tools that they can use to be their own provider whenever possible. They are a very complementary set of goals.
To accomplish these goals, we have four design principles that we apply:
The first is accurate collection of data (usually physiological information): blood pressure, glucose, step counts, that kind of thing. The second principle is sharing that information back with you in a way that is meaningful to you in the context of your illness. The third component is coaching and the term coaching is very broad to us. It could include sending a reminder to a person or contact with a human being. The idea is you take that behavioral and physiologic data and arm a coach with the same data you are giving the patient and it turns out to be very powerful. The fourth is connecting that information to a healthcare provider when one is needed.
We have programs in a number of chronic illnesses: heart failure is our most advanced program. For diabetes we have an up and coming, growing program. We have done a lot of interesting things with high blood pressure, both in terms of our use as a provider but also in terms of in the market place as an employee benefit. We have a program up and coming in activity monitoring and weight control. Down the road we will be working on mental health, depression and so forth.
That’s a very quick fly by of what we do.
Are these programs still pilots in the development phase or are they commercially launched? Can you drill down on some of these programs and how far along they are?
Well, not all of them are pilots: Our heart failure program is a bona fide service now. It’s well integrated into how we care for heart failure patients as a system. The work on hypertension is now offered as a service. We now have legitimate customers and we are spinning that out as a business in the next few months. I would say [our diabetes program] is a pretty advanced pilot, but it’s definitely still a pilot. We also so research and development work going on with medication adherence technologies and new kinds of things that are up and coming as well.
How do these programs make use of wireless technologies?
We have done a lot of wireless is the use of SMS text messaging. We have been very fond of text messaging because it enables us to reach the broadest number of users. If you start out having to narrow your sample size because you are treating people with a certain illness or a certain level of engagement with the system already, then demanding that they have a specific phone or carrier can really, really dull the impact of your intervention. We have been more excited about text messaging and have used it really as a reminder tool for the most part.
We did one study that was exciting where we compared a group of individuals who didn’t get daily reminders with a group that did. The goal of the intervention was to remind people to put on sun screen. This was a sunscreen study. We had a device that we could use that could measure whether people were actually putting sun screen on or not. Both groups had that same device. Every time they opened the tube to put on sun screen, it was a measured event that we could track. The intervention group had a daily text message that shared with them a weather report and reminded them to use their sun screen. That group had twice the level of adherence to the care plan as the control group. It was quite dramatic and statistically significant.
We have taken that data to roll out a number of things. We are just in a month or so going to start enrolling patients from one of our neighborhood health centers in two programs where we are going to use text messaging as a tool to remind patients to engage with the healthcare system on critical events. The first one is teenage pregnancy and making sure those teenage women or girls come in for their prenatal visits. They are at great risk if they don’t so we are going to use texting to remind them to come in. The other one is with substance abuse patients who are on a particular regimen where they have to undergo testing to continue on that regimen. We will use texts to remind them to come in so they can continue on that regimen.
Again, we see a big future in using text messaging as a reminder tool, of course, there are certain populations that don’t respond to texts. One of the big things we are big on here is segmenting so that we don’t use a certain technology where it will fall on deaf ears or even annoy someone. We estimate that there is about 20 or 30 percent of the population, and it’s much higher in the younger generation, where text message reminders will be very powerful.
Beyond younger age groups, what other demographic groups do you think will be interested in adopting these wireless-enabled health services?
We have done research on parents of children with Type 1 diabetes. We have found a very rich market there for this type of thing. We have developed a prototype, we haven’t been able to bring it to the next level yet, where the child tests their glucose and it gets uploaded to the mobile phone and then that is automatically messaged to Mom or Dad that the child had uploaded the glucose and what it was. So they could keep track that the kids are in fact testing and that the kids are taking their insulin.
Was that uploaded automatically or did the child have to enter the data manually into the phone?
You can do both. One of our core operating principles is that automatically uploaded information is much more powerful than self-entered. You can arrange it both ways. As I am sure you are aware, the current technology and the automatic upload really limits you because only certain glucometers can do it and it might require a cell phone sleeve. There are various ways to solve this, but none of the solutions [for meter integration] are perfect. Understand that, but that’s why to get really wide spread adoption you might have to forgo that automatic upload first, but have some way that if the kids know that if they put in a number that wasn’t proper, that they will be checked on later. That should keep them honest. The issue is that people tend to not always put in the right information if they self-enter. Whatever the age group and whatever the disease.
What else are we interested in mobile? Well, we are quite interested in embedded mobile, but we haven’t done anything with it yet. This is the idea that your sensors will have a mobile chip in them and have an assigned phone number and, like the Kindle, once you buy the product and register it, it is forever feeding information into your account. We think there is a huge opportunity for that. We are waiting on some really early-stage relationships that I can’t speak in any detail about, but we are looking into the possibility of testing out some of that equipment.
What is an example of this — doesn’t have to be one you are pursuing just to flesh out the idea of “embedded.”
Well, like a blood pressure cuff in our hypertension program. It would have a mobile chip embedded into it and a phone number assigned to is so that whenever a user took their blood pressure the data would just be sent along. We find that as a very compelling vision. The caveat, of course, is that we know it will add cost so we have to always do the ROI around it. The ease of use, though, is very compelling. We find that if we want accurate, frequently uploaded information about an individual we have to make it so easy for them. The least little thing that adds works is often a deal breaker. People are busy and they have all sorts of priorities. We like that concept of embedded mobile.
During this discussion so far you have noted a number of challenges that uptake in these services face, are there any others that should be mentioned?
I have a touch of worry about the reliability of the network. That phenomenon is still there. The cellular network is really built so that kids can swap a lot of texts and if one doesn’t go through then who cares? For really robust healthcare use, things need to be very reliable with high quality of service. Network availability is still a challenge in many areas of this country. I just wouldn’t want to lose sight of that. To make something really high quality for healthcare we need a little better network infrastructure. I am very bullish on mobile, I just think that sometime the vision gets out a little ahead of the implementation.
Do you think the vision getting out in front of the implementation is any different from any other new technology? Or is there more of that going on in wireless healthcare right now than is typical?
No, but there is more complexities that people don’t want to talk about. If I go back to the days when the Internet was getting hot, say the mid to late 90s. Even when you had to sign up for dial-up service like MindSpring or AOL, if you had Internet access then you had Internet access. Back then most people were using machines that ran Windows so you could build software or just use the browser as a universal tool to run software and get things done. As you know, better than I, that just doesn’t happen in the mobile industry. There are so many different devices and they change rapidly. They change the software without telling anyone. You go on one carrier and it doesn’t work with the other carrier. You can only get this device on AT&T, well, I’m not an AT&T customer. It just throws up a lot of barriers for getting a horizontal solution. Most of the people who are bullish in building all these visionary tales of wireless health are sort of ignoring these issues. They will come up with the iPhone app and say, but a third of all doctors carry iPhones. As you and I know, there are 40 million people with diabetes, how many of them have iPhones? If you really want to get diabetes taken care of you have to solve that problem. I think those are just some of the challenges that I don’t see talked about a lot. That’s sort of skunk at the picnic view. There is so much potential here and I don’t want to see it go the way of the Internet Bubble and be remembered as the Mobile Health Bubble.
Clearly, fragmentation in the wireless industry — exclusive phone deals, included — make it difficult for any company to offer horizontal services cross-carrier. It’s a long held frustration, but hopefully it’s getting better.
Yes, the business model that has been so successful for the mobile phone carriers to date, makes for a very challenging context in which to innovate in.