Understanding Patient Behavior to Improve Care and Experience

By Matt Phillion

A recent survey of 70 national and regional health plan leaders looks at the impact that advanced health plan member personalization can have on creating a more sustainable healthcare ecosystem to improve outcomes.

The survey, commissioned by Lirio and conducted by healthcare consultancy Sage Growth Partners in the first quarter of 2025 uncovered some interesting findings:

• 84% of the health plans surveyed said the most significant barrier preventing their membership from taking action following plan outreach is that their members are unmotivated to change, despite being aware of necessary actions to take;
• 60% of respondents cited “difficulty engaging members to improve their quality of care” as a top challenge;
• 40% of survey respondents rated their current solutions as “not well at all” in terms of providing adequate machine learning and AI tools to optimize workflows for increased engagement and improved outcomes;
• The majority of respondents indicated that population health analytics will be their most likely near-term technology investment, despite broad recognition of the impact AI can have on member perspiration and experience; and
• 77% of respondents stated that it is “very important’ to incorporate personation for chronic care outreach, such as diabetes care management.

“That first statistic is very interesting. These leaders are saying people aren’t motivated, and that shows they’re missing the boat in understanding their members and how they’re thinking,” says Amy Bucher, PhD, chief behavioral officer of Lirio. “There’s a difference between getting you to do what I want you to and getting you to do what you want to do. The science of personalization is one of translation.”

The healthcare plan wants the patients to be as healthy as possible to minimize costs and unexpected care, but the members want the same thing, Bucher explains.

“They may just not express it in the same ways,” she says. “There’s this missing part on the part of plans on what personalization can do. It’s not just reviewing and explaining benefits. It needs more patient-friendly language. It might be a reminder to go see a doctor in a way that will facilitate your ability to take that annual trip.”

Technology can play a big role in personalization, notes Marten den Haring, PhD, MSc, CEO of Lirio.

“We talk about AI as though it just arrived, but the reality is we’ve had forms and flavors of AI techniques for more than a decade. If you look at social media as an example of AI and machine learning gone wrong, you can see what happens when you feed people what instantly gratifies them,” he says. “It’s necessary to take a step back and try to understand what’s best for us. If you were to watch an algorithm examining how I go about my day and learn more of what I’m doing, it’s going to pick up on more of my bad habits and would not necessarily learn to adjust course and lead to a better change.”

There’s still a skill gap in this area of technology because the pool of people who are experts both in the techniques needed as well as how to understand the implications, den Haring explains.

“We need to be intentional in this. Healthcare is where the real impact or harm of these technologies will show itself in the next few years,” he says “It’s very clear big tech is bringing this toolbox to healthcare, so what are we going to? And will we have learned something from the social media experiment over the last decade to do something different? It’s all about how we get to the bottom of each individual and how we support them in making better choices for their health.”

The architecture of choice

Getting to that point requires a level of involvement with the individual patient, den Haring explains.

“It implies involvement of the patient in the choice architecture,” he says. “An element of nudging people not only to what they want, but to keep nudging them about what choices they have with their overall health and how they can improve it.”

“Facebook, for example, has a real goal of keeping people on the page to stay and interact so they can monetize that,” says Bucher. “In healthcare, if we can get that translation right, we get out of the mindset of saving money or raising HCAHPS scores and into getting them to make healthy choices.”

Healthcare is specifically aligned to make that shift, den Haring notes.

“There’s an absolute moral compass in healthcare that maybe doesn’t exist in other industries,” he says. “It’s rare you find a clinician who doesn’t want to act in the best interest of their patients; the principle of nonmaleficence is what guides them. The bigger concern is if you believe a new technology can make an impact in healthcare, the first reaction is often, ‘Let’s go hire people from other industries who have done this before.’ The danger is that these people bring their toolbox without realizing that healthcare is different, without fully appreciating the risks and the moral compass.”

The intent of adding technology can be to help, but it must be done in a way that it ultimately helpful, Bucher explains.

“Clinicians have been really frustrated by tech thus far: challenges with the EHR is a top cause of clinician burnout,” she says. “We can offset some of the administrative burdens to free time to activate patients so they can become more empowered consumers and interact more proactively with the system without adding to that same administrative burden.”

Something Bucher finds frequently in audits is that outbound phone calls are handled by nurses.

“These are people who went to school to provide clinical care and they’re sitting at a desk: what if technology can do that proactively so that these nurses can respond to people who need help?” she says. “I think we can add real meaning back to that clinical experience.”

“As long as you understand what the objectives are of implementing a technology, you can work backward from those objectives. The worst thing you can do is just throw the tech in there,” says den Haring. “In our industry we have to remain laser-focused—What are the quality objectives? What are the safety objectives?—and fine-tune the technology to help optimize those.”

This also means that we need to make sure the tech stack and machine learning models are uniquely tuned to healthcare rather than taking a generalized tool and retrofitting it onto healthcare, den Haring notes.

“We want to help patients understand the right choices for their health, the consequences of those choices, and to help them make better decisions,” says den Haring.

The importance of behavior

Organizations need to think about behavioral barriers that stop people from taking actions to better their health, says Bucher.

“There are certain behaviors a person struggles with and certain ones they don’t,” she says. “We research that very granularly, using behavioral science models to help patients overcome those barriers. If we can help them with problem solving techniques or increase their self-efficacy, they’re more likely to make that first step. You can get the ball rolling to show them: this isn’t so hard. You’re making progress. It gives them something to latch onto.”

We also know that consumers have previously formed opinions, and that they know what they think they want from a digital experience, den Haring says.

“We have to design for where they are—having an ethical, thoughtful portal that does not provide a good experience isn’t going to change behavior,” he says. “Healthcare needs to figure out what its catalogue of products and services need to be. You’ve received a diagnosis, here is a list of things you need to do that will make a difference your daily life and help consumers trust that someone cares about their health and isn’t just trying to target them for an annual visit or refill a prescription.”

It’s about using data known about the patient to personalize the interactions and make better recommendations.

“We need to make this about the patient, the consumer, the member, the human,” says den Haring. “Trying to help people live their best lives, making better choices for themselves. At the end of the day, it’s all about quality of life and longevity. We make tens of thousands of choices a day. How do we rewire systems to help serve and live our best lives, and how do we change from a mentality where we’re making sure you check a bunch of boxes for compliance to really start to work toward how your daily choices are impacting your life?”

“I’m excited to see us break down silos,” says Bucher. “We experience this across many organizations. Even going to the PCP for regular care but pharmacies for vaccines. We can create something that feels more like a journey. A lot of healthcare takes place outside the clinical system entirely. My dream as a behavioral scientist is to tap into the patient’s motivation, to understand what their journey looks like, and empower them so they don’t need us every second of every day.”

Matt Phillion is a freelance writer covering healthcare, cybersecurity, and more. He can be reached at matthew.phillion@gmail.com.