Some health plans are interested in more than just consumers' health data. Demographic, socioeconomic and purchasing information are helping insurers focus their marketing strategies to gain a competitive edge.
Phil Fiero, senior vice president of analytics at Denver-based Welltok, said predictive and prescriptive modeling allows healthcare organizations, like insurers, to gain a holistic view of potential customers by using data unrelated to healthcare.
He spoke with Becker's Hospital Review about how prescriptive models can help insurers extend their plans.
Note: Responses have been lightly edited for length and clarity.
Question: How can insurers use analytics to sharpen consumer outreach?
Phil Fiero: One particular plan we worked with was based in New Jersey. It was a small, Medicare health plan looking to expand rapidly into six counties in New Jersey. They wanted to take a different approach than a lot of organizations take.
Outreach is expensive, and we found that the six counties they wanted to expand into had a little more than 350,000 households. That would have cost them a lot. The idea was — is there a way to cut that down to just the individuals who are most likely to respond to outreach? And find people who wouldn't just sign up for the plan, but also stay 12 to 18 months?
Q: How does the model identify those potential consumers?
PF: We took the plan's current population and created a model to profile that population based on demographics, socioeconomic status, purchasing, etc. The insurer offered a $0 premium plan, so policyholders had a lot of common characteristics. Most were ages 65 to 72, single family households, predominately Hispanic and black and Republicans. We identified the characteristics of people who have been in the plan and applied that information to find people in the six counties that resembled them.
In doing so, we cut 350,000 households to 43,000 households that were five times more likely to join the plan.
Q: How did the insurer use the data?
PF: The information was utilized by the plan in two ways. One was standard direct mail. They sent mail out to 43,000 households and had a much higher response rate. The other thing we were able to do was a geospatial analysis. We found that in certain areas, there would be a high density of those 43,000 target households. We helped the insurer identify certain neighborhoods to provide community events. They had a big orange wellness bus to provide free checkups, and we actually got all the way down to certain intersections where there were people who would be interested in the plan, and had them park the wellness bus there with the most concentration that would utilize the bus.
Q: How have insurers reacted the model?
PF: The reactions have been very positive. The model is both strategic and tactical. It's strategic because they get a much better knowledge about their members. Cultural differences, caregiver status, purchasing – they help inform future marketing strategies. That's one very positive outcome. The second is allowing companies to understand what their marketing spending should be. Often, it is hard to forecast budgets. What analytics can do is add some science around that. It gives a better understanding of what types of outreach are really resonating.
Q: What will be analytics' role in the future regarding health plan expansion?
PF: As there's further disruption with Affordable Care Act exchanges, understanding more about population has been critical. We find organizations are really understanding the incumbent members much better, whether or not they are trying to reach them or not. The more lives they have, the more of a presence they have. So I see this expanding quite a bit.
More articles about payer issues:
Cigna taps 'McDreamy', other TV physicians for ad campaign
AMA, physicians press NY state to reject Anthem-Cigna deal
Independence Blue Cross to open customer and community center