Can We Be Friends?

Quick sidebar: Economist Arthur Laffer (of the famous Laffer Curve) wrote an editorial in the WSJ this morning that makes a similar argument as I did in this blog for stripping the insulation from healthcare costs (he calls it the "healthcare wedge") that may be an alternative to the current plans being offered for healthcare reform. He argues the point more eloquently than I did, though. Check it out!

It's no secret that most Americans have a deep mistrust of their health insurance company. We firmly believe that it's in the company's best interest to take as much as the market will bear from us in terms of monthly premiums, and give as little as possible in the way of payments when we seek care.

But what if there was a way that our insurance companies could cut costs, and we could be truly, genuinely be healthier because of our relationship with said evil company?

I think there's a way to do it, and it's through predictive analytics.

Health insurance companies have a wealth of historical information about subscribers in the form of claims data. Every time I go to the doctor, or have a test done, or see a specialist, my healthcare professional would like to get paid for their services, and so they file a claim with my insurance company.

Over time, the millions of claims being filed by healthcare professionals for services performed on all of us becomes a treasure trove of data for analysis. Each one of those claims is coded with certain pieces of information, and something that's present on every claim is an element called a "service type code", or in other words "what kind service did I perform on the patient?"

Imagine if we performed an analysis on the dates of service and service type codes and found a statistically significant pattern of claims that looked like this:

Patient: Male, age >55, smoker, BMI >28.7
Claim 1, Day 1: Office visit
Claim 2, Day 3: Lipid profile
Claim 3, Day 30: Office visit
Claim 4, Day 35: Emergency room admission, heart attack

This is a simplification, but it's to illustrate a point. If this pattern had predictive power, wouldn't it be a great win for both the patient AND his insurance company if they made a call to his doctor on day 31 saying, "we've noticed that your patient has just fallen into our high-risk profile for heart attack in the immediate future. Please make a wellness outreach and begin an immediate statin and anticoagulant regimen if appropriate." (This assumes healthcare professionals get paid for proactive outreach, which is a topic for another day.)

I know I'd appreciate not having the heart attack, and certainly the health insurance company would like to avoid paying the tens of thousands of dollars it would cost to save my life. I benefit, and they benefit.

If this were done for all of us, we might have more reasons to like our health insurance companies!