This article is initially available on our podcast. Click here to listen.
After I finished the last podcast, it occurred to me that I should probably explain some things a little bit better. The concept of expected value is something we touched on in the last podcast. I suggested this is something that we should try to implement and focus on in healthcare.
Analyzing patient care
The concept being when a patient walks through the door, you should have some idea of what you anticipate in terms of revenue from that patient. Further, not only in general in terms of average reimbursement per patient, patient encounter, or something like that, but even getting much more granular and more specific. For instance, if the patient has made an appointment (they have knee pain, or they’re coming in for PT, or whatever that might be), it should be pretty easy to analyze.
With the known reason for setting an appointment or the patient presented to the ER or whatever it might be, you should have a pretty good idea of what the average CPT codes are. It would help if you also understood procedural codes that are combined with that. Determine the right combination to use. Also, know what the reimbursement is for that insurance, for those codes, and more. But I didn’t explain what the expected value was.
The coin toss
Let’s think about this from gambling. Think like a coin toss. Let’s say, for example, you’re going to bet $1 on a coin toss, and you’re going to pick heads. Today, we’re going to pick tails. We want to be contrarians, so we’re going to pick tails. Your odds of tails, forgetting for a moment the fact that it isn’t precisely 50/50 (that’s a side note). Assuming odds are 50/50 that the outcome is heads or tails, if you have a 50% probability of hitting tails and being successful and you bet $1, your expected outcome is 50 cents. You’ve bet $1. Thus, you have 50/50 odds.
On the other hand, you have a 50% chance. And the concept of expected value here is that you take the probability times the outcome, which gives you an expected value. So the result would be to win $1, and the probability is 50%. You multiply 50% times dollar, and you get 50 cents.
Flip or flop
Consider a fun side note. It isn’t 50/50 when flipping a coin. I bet you don’t know this unless you’re, again, really into statistics and wonky like me. So your odds are not 50/50 when flipping a coin. So if you’re ever in a situation in which you want to win a coin toss, you have the option of selecting. There are two different ways to play this. If somebody is tossing the coin, you want to pick the side that is already up. So if heads are up, you choose heads. If tails are up, you like tails.
The reason why is because it doesn’t flip exactly around a horizontal axis. It’s impossible to flip that exactly. So what it does is precesses around a vertical axis as well, which means that it doesn’t flip completely randomly over itself. So it’s not 50/50 odds. It’s more likely to land on the side that was already up.
There’s another way to do it, too. If you don’t have the choice when they’re flipping, and you’re flipping, and they pick heads, then you can turn heads down, the side they chose starting down, and they’ll have lower odds. You’ll be more likely to win. Now, is this cheating? I don’t know. All I know is that the probability isn’t 50/50 and that matters. It’s good to know things regardless of what you want to do with that information.
Okay, back to the expected value. That’s an explanation of the expected value. In this case, when it comes to insurance reimbursements, we have broken this down in some of our analyses where we look. For example, “Okay, you should be making $100 for that procedural code from that insurance company, but you only get paid 67% of the time from that insurance company for that code.”
So when you are doing average reimbursement, you are effectively doing an expected value for that procedural code, for that payer, or, again, whatever level of granularity you take it to. You’ve effectively done a scheduled value calculation when you do the average, assuming you do it in that fashion, including the zeros and more.
That’s the expected value. Again, I think it’s essential to be using that and to try to work that into all healthcare providers because, as we made a point in the other podcast, you can do a lot with that information and improve profitability.