Connected Health: Bake behavioral science into health technology

BOSTON--Motivating individuals to engage in healthy behaviors is easier said than done, and new technologies developed to influence behavioral changes must keep in mind the science of what works and what doesn’t, said speakers at Partners HealthCare’s 10th Annual Connected Health Symposium.

After working on an online exercise motivation program at Columbia University that “was a total fail,” Stephanie Habif, PhD, said she spent an entire year conducting an in-depth ethnography with an obese woman, which entailed living with her for four months. Thanks to the intervention, the woman lost 65 pounds and finished a marathon.

“My number one question was how do we scale this?” she said.

No matter now sophisticated an app is designed, it isn’t enough to throw content in there and hope it works. “There’s actually an art to this… Translating behavioral theory into technology design isn’t happening enough,” said Habif, who has since since focused on leveraging her experience as a behavioral scientist to consult for healthcare tech companies.

Chris Niemiec, PhD, senior lecturer, University of Rochester, said three conditions that support behavioral change include autonomy, competence and relatedness. These elements, which constitute self-determination theory, inform the highest quality forms of motivation and engagement for activities, including enhanced performance, persistence and creativity, he said.

On the matter of how to incent people to change, Heather Patrick, PhD, program director, scientist, at the National Cancer Institute, said experimental psychology shows that incentives and awards outside of a person generally do not work for the long term. In the matters of weight loss and smoking cessation, incentives are great in the short term but cannot be expected to create long-term change.

Niemiec agreed, noting in the realm of weight loss, paying someone to exercise every day “gets expensive quick… It’s rarely cost effective."

“If you can’t categorize the behavior you are designing for, you’ll fail,” added Habif. She said content developers should refer to a behavioral grid that features 15 ways behaviors can change. “If you are designing for a one-time behavior, go for incentives. For a two to three week change? Go for incentives. But Long-term? No way,” she said.

Companies should take advantage of the fact that mobile devices open up accessibility to individuals and allow health workers to meet people where they are. Patrick said a two-way texting program can be more effective, in that it sends tools, tips and advice on how to deal with cravings as they are happening in real time. Additionally, “it increases the dosage of intervention without increasing the clinical burden.”

Niemic added with clinicians so pressed for time, technology’s ability to provide ongoing health-relevant messages for change means more clinical time is spend developing strong patient-clinician relationships.

However, as new health technologies are rolled out, more data need to be collected on whether or not the apps work. Habif said help is needed from the private sector to provide that evidence. “Physicians need proof of concept. We don’t have that quite yet in the world of digital health—but we’re working on it."