By Sarah Klein
Quality Matters spoke to Walker to find out how this information is and could be used to influence health outcomes and utilization.
Quality Matters: One of the key takeaways from your work is that physicians, nurses, and even some health coaches often assume patients think and act like them or would do so if given the right information. But your research suggests that the largest segment of patients—the ones you term “Willful Endurers”—are often dismissive of health care providers. Can you tell us more about this group and what providers may need to know about them?
Brent Walker: This group is very hard to reach—even the name of their segment is hard to say. They live for the moment and the day and have a hard time thinking about what their health will be like six months or a year from now. They’re also very independent and disengaged, and when they do use the health system, they have the hardest time with medication adherence and follow-up care. They are also very suspicious of providers and feel that society in general doesn’t understand them.
Quality Matters: So how do you approach them?
Brent Walker: We worked with a company that had manufacturing plants across the United States. They had a large blue-collar workforce spread out over three shifts and were trying to get people with metabolic syndrome to come to screenings and a 12-week coaching program because their health condition was driving so much spending for this employer. But the company wasn’t getting the desired response among eligible employees, even though they were offering money and an incentive to attend the screening. We went and looked at the way they were advertising it—the message was geared to “Self Achievers” who will show up even without a financial incentive. Those weren’t the folks who needed help. We altered the pitch to appeal to the “Willful Endurers,” who were suspicious that this was strictly a cost-cutting move rather an effort by the company to help them. We told these employees that the reason the company is making this investment in this program is because you’re good at what you do and the company wants to make sure you’re around a long time. This was true. By emphasizing their value to the company, and reframing the company’s interest in the program, we got 72 percent of eligible employees—those at risk for metabolic syndrome—to show up for a 12-week intervention program.
Quality Matters: Can you give us another example of how segmentation might dictate a provider’s communication strategy?
Brent Walker: Another good example is communication with the “Balance Seekers” and “Self Achievers,” because on the surface they look a lot alike—they’re very proactive about health and wellness, but their motivations and communication preferences are different. If you approach both segments with the same advertising message, such as a patient education or disease management program that puts the doctor front and center, it’s going to fall flat with “Balance Seekers.” They don’t view physicians as having the same credibility as the “Self Achievers” do.
You see something similar with people who are against getting their children vaccinated. A statistically significant number of these people fall into two groups—one that’s very proactive and wellness oriented (“Balance Seekers”) and another that’s disengaged (“Willful Endurers”). We have more work to do to understand how these psychographic segments differ when it comes to caregiving as opposed to self-care.
Quality Matters: Does training providers just to be aware of psychographic differences help them in their communication approaches? Or would it be more worthwhile to focus on care managers who have more frequent contact with high-need patients? Also, aren’t some providers going to know this intuitively and others going to struggle with it if they are less focused on patients’ personalities?
Brent Walker: The feedback that we are getting from providers is that these insights not only help them understand and interact with their patients better, but it helps them be introspective about themselves. They begin to realize that their own personal lens biases how they communicate with patients and interpret their behaviors. This results in true patient engagement, with providers talking with patients, not at them. While there are many gifted providers who will know this intuitively, we want to offer a structure that systematizes these insights across an organization for the benefit of all patients.
Quality Matters: How do you know the segments are accurate?
Brent Walker: We relied on factor analysis and statistical clustering procedures to identify response patterns in our survey, to group respondents according to similar answers. We looked at over 90 possible psychographic segmentation models, and went with the one we thought offered the right balance of predictability and practicality. That is, we can predict the segment to which a patient belongs with 91.1 percent certainty, but we limited the number of segments in our model to five, because more segments become challenging to manage for a health care organization.
Quality Matters: Do you find people move between segments? It seems plausible that as a patient gets sicker, or faces a life-threatening condition, their view of providers and willingness to engage might change?
Brent Walker: We do find people are actually a mix of the five segments, with primary, secondary, and tertiary preferences; few people are, say, 100 percent “Willful Endurer.” We focus on the primary preference because it simplifies the approach because the majority of people will keep to their primary preferences and if we can bend or move what is essentially a bell curve, we will be successful across a population.
That said, we have seen some interesting insights across successive market research studies. For example, “Balance Seekers” with cancer are even less likely than healthy “Balance Seekers” to believe their physician is the most credible source of health information, unlike the other segments. Perhaps the condition accentuated this aspect of the segment.
Quality Matters: You winnowed the list of 400 questions down to 12 so that organizations can quickly survey patients to determine which category they fall into. What are some of those questions? And how do you know those are the right ones?
Brent Walker: We ask them the degree to which they agree or disagree with the statements: “I can control my health care and health outcomes regardless of my family history” and “I believe natural or alternative holistic medicine has benefits to my health and wellbeing.” We also ask to what degree do you agree or disagree with the statement: “I am willing to have my medical care delayed or have access to a physician appointment delayed so everyone can get the care they need?” We didn’t pick these and the remaining questions. We let the statistical analysis tell us which ones were most predictive of the way they would answer the other questions.
Quality Matters: What does that last one tell you about someone?
Brent Walker: It gets at whether they prioritize self and family over society. Some people feel their health care needs are so urgent that everybody else be darned. I have to take care of my family first. Other segments will say “Look, what is good for the common good is good for me and I am at my best if everyone else is at their best.” It is a different lens of how people interpret the world.
Quality Matters: Are you finding the health care industry is interested in the type of information?
Brent Walker: I know that some of the big health plans have been using this type of segmentation for member acquisition and retention. I am finding a lot of health systems and hospitals are interested in it, but it is hard to operationalize it. It requires a lot of training and coordination—how do you get everyone in the organization to buy into it and learn how to communicate with each segment? It can be a herculean task. Without that, psychographic segmentation just remains a really fun, interesting, academic exercise.
Quality Matters: It seems as though you’d need to make a case that this can have an impact of outcomes. What evidence is there now and what are you researching?
Brent Walker: We have a variety of case studies in which the application of psychographic insights has led to positive behavior change. We are currently in a pilot project with a major health system to measure the clinical impact of patient engagement using psychographic insights—for that we’re focused on diabetes and musculoskeletal issues. In partnership with another company, we are looking to see if automated patient communications tailored to psychographic characteristics improve medication adherence and reduce readmissions, among other things.