Behind the curtain of technology and behavior change

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon
 - People Physicians Patients

BOSTON—The concept of continuous care improves patient engagement with the healthcare system and improves both clinical outcomes and efficiency, said Kamal Jethwani, MD, MPH, head of the research and program evaluation initiatives at the Center for Connected Health, speaking at the 2014 AMDIS Fall Symposium.

Several steps lead to continuous care, he said. The first is data collection, followed by putting the information into relevance and making it actionable. “Getting to action is really difficult without the right motivational tools. People are motivated by different things.” Jethwani showed how radio buttons for four factors—social networking, gamification, incentives and coaching—can be adjusted to meet different motivators.

He shared an example using remote monitoring for heart failure. Both patients with heart failure and without dropped their rate of hospitalization by 50 percent. “By monitoring heart rate and blood pressure, we also see drops in falls, urinary tract infections, chronic obstructive pulmonary disease and more. We realized that when patients start monitoring themselves for anything, they make healthier choices across the board. There are very few things in medicine that can do that.” The monitoring even increasing patients’ lifespan.

Jethwani said there is power in simple messaging. For example, before the days of smartphones, the center tried to increase use of sunscreen. Two groups got electronic caps but both the control group and intervention group decreased their use. The researchers tried adding a weather report to the messaging that included a note about applying sunscreen. That gave them a reason to open the message and get the reminder.

“This exemplified the value of giving patients something they need and want,” he said. On experiments on medical reminders, smoking cessation, prenatal care and more, “every single time we combine these concepts, we get amazing results.”

Another example is the Text 2 Move program which aims to get patients with diabetes to walk more. Messages are based on each person’s profile and the engine behind the program did machine learning, analyzing patterns and messaging accordingly. The algorithm can detect when patients are moving less and offers reminders and suggestions in a “hyperpersonalized way. Engagement with the system went up, loyalty went up and there were fewer no-shows at appointments.” Also, patients using the program had better results than those just taking the medication.

More personalized care is “simple and extremely difficult,” said Robert Havasy, MS, corporate team lead for product and technology development, also at the Center for Connected Health. “Almost all interventions require data but knowledge does not equal motivation.” He said ability matters more than motivation when it comes to taking action. “Behavior is the intersection of ability and motivation. If you are having trouble getting someone to do something, it’s better to make it easier for them to do it than to give them more motivation.”

To get something to stick, the technology has to disappear. “It seems simple but, in fact, is very complex. We can’t just reach patients. We have to engage them.”

All systems need an integration layer “which proves critical because it is the flexible piece that allows us to do many of the things that appear simple,” Havasy said, such as gather data and pull data out of patients’ records. The Connected Health system receives data from a variety of third parties. “It’s incredibly hard to find all of the little interfaces but it’s incredibly important to get this foundation built. The rules engine has gotten more and more critical.”

Connected Health built its own personalization engine “to digest the information and turn it into something useful for both sides of the equation.”

Havasy said the center’s staff thought this was a data interoperability problem. “It turns out it was a data problem all along. While device connectivity is an important foundation, our focus needs to remain on data interoperability so this can go where it needs to go for whom and at the right time.” Another area that needs more exploring is how to turn this into something that can be scaled to a national health information exchange.

When asked about the cost and financing of these programs and tools, Jethwani said the willingness to experiment in this area has gone up significantly. “The jury is still out on exactly the return on investment but I think there’s something there.” He cited a study that found that 73 percent of patients go to the internet first when they have any symptoms. “That number was probably in the single digits two or three years ago. There is a consumer push to make healthcare more accessible and efficient. Even if the return on investment is not compelling it will be the right thing to do for our patients.”