February 14, 2008 
I think an onion is the right analogy for healthcare for three reasons: (1) it can make you cry; (2) every time you pull off a layer you learn more; and (3) what you see from the outside is a lot different than what you see from the inside.
- It can make you cry.
When you have the Congressional Budgeting Office projecting the healthcare costs will be 49% of GDP by 2082, you know things have to change. This is a front page topic almost everyday across the country. But, like an onion, if we don’t handle this right, it will make you cry out of frustration and pain. Change is not easy especially in a complex system that we have today. Finding the right mix of push and pull is going to be important.
Quality is still an issue across the system. Biting a bad onion or having a quality issue with your care can make you cry. Look at the USA Today article from the other day about Too Many Prescriptions, Too Few Pharmacies or an entry on my blog about the Institute for Healthcare Improvement.
- Every time you pull off a layer you learn more.
This applies so many ways to healthcare given our system, but I think of this from two perspectives – data / information and process. We have so much data in healthcare, but without the right model to make it into information, it just sits there. And, as we layer data (e.g., medical plus pharmacy plus lab) or integrate healthcare data with demographic data, we can learn so much more about our patients and how to care for them. This ranges from simple questions such as how to motivate behavior (e.g., cost savings versus loss avoidance) to how to deliver information based on their learning style.
Every question you ask (or layer you pull off) reveals a new set of data that can be transformed into information while at the same time creating new questions. Does the relationship you found in the data simply indicate correlation or is there actual causality there? I look at the data that CVS/Caremark presented around saving 30% of healthcare costs by driving compliance and adherence and wonder why people aren’t jumping up and down trying to capture this savings.
- What you see from the outside is very different than what you see from the inside.
There is a concept in Six Sigma about designing the process from the outside-in. Imagine sitting in the middle of the onion…all you see is onion all around you. That is a common pitfall when solving problems in the industry that we work in. We are too close to the problem and the historical solution. If all we see is the onion, those on the outside (our patients / members / employees) see the onion in relation to other food options. Their expectations for healthcare are produced by other companies that they interact with. They expect web solutions that work. They expect excellent service. They expect to be valued as a customer and of course need the power to walk away and chose another option.
This is a common problem in healthcomm (healthcare communications). We present information in a channel that we believe is effective based on our experience and paradigm (i.e., written, verbal, kinetic). We use language that we think is helpful. A few of my favorite examples from my PBM days are:
(1) Telling patients that they need a renewal (prescription). They don’t know what that means. It means they need a new refill since their original prescription refills have run out.
(2) Telling a physician to consider prescribing lisinopril and giving them sample bottles that say lisinopril. [Because, of course, they would know the chemical name for Zestril.]
But, this happens all the time. Telling a person that wants all the facts a lot of qualitative information will fall on deaf ears. Providing a person with lots of options when their looking for an expert opinion will frustrate them. One way to frame this is based on personality type. (Of course, that information isn’t sitting in a database somewhere for us to tap into.)
The reality is that people are different. As you think about your healthcare process, try to be the patient. As one of my bosses used to say, give it to your grandmother and see what she thinks. Can she understand it? Can she make sense of the process?
It’s not easy finding the right amount of onion to use in your recipe, but it is important to continue trying to improve.
Enabling Healthy Decisions 
Actually, data on adult health predispositions (akin to personality type) is available and could save tons of money if used. That’s my work. My work on health predispositions and health priority patterns identifies the learning style of a person (passive versus active) as well as their propensity to seek care and prescribes the optimum communication tactics (depth of information, repetitions, mode of delivery). And it works. The approach is a holstic model of health predispositions towards family, prevention, exercise, learning, trust, locus of control, and care seeking. It links them to disease prevalence, risk factors, and costs. My work describes the “shape” of adult predispositions to seek care and shows that they are as much a risk factor as any disease condition. Here’s an example:
“I only go to the doctor if I am really sick or injured.”
Agreement or disagreement with that statement explains between $500 and $1,500 of medical costs per person per year irrespective of health state or disease condition. For every 1,000 members a health plan has who disagree with the statement, they will cost the health plan between $500,000 and $1.5 million MORE than a 1,000 members who disagree. The cost differentials show up across Rx use, use of physicians and specialists, and hospital care. These costs are not driven by poor health or disease. They are costs driven by a person’s “trigger point” of when they decide its time to seek care. Those are dollars spent on treating the wrong thing!
This is one of the reasons why the “consumer-driven health” model is flawed as a solution to health care cost reduction. The CDH model is an economic model that mainly shifts costs from the employer to the employee. It may lower the overall trigger point of seeking care, but the folks who have a low trigger point will still be low in comparison to those who have a high trigger point. That’s an easily identified dynamic that can be addressed to reduce the spending of unnecessary health care dollars.