VORP WAR

I had an interesting conversation with one of our physician leaders this past week.  We were both looking at some medical staff turn-over data (which, incidentally, showed that there is no significant physician turn-over at our organization), and he asked whether we should look at "VORP".  My first response was "What in the world does VORP have to do with medical staff turn-over and hospitals?"  However, as I started to think more about his question, I wondered to myself, "Why wouldn't we look at VORP?"

Unless you are a baseball statistics geek like me, you may not have heard about "VORP".  "VORP" is an acronym for a statistic called "Value over replacement player" and is very similar to another statistic (perhaps more well known) called "WAR," or "Wins above replacement player."  The idea is that you can determine how valuable an individual player's contributions are to the team's overall win-loss record by assessing how valuable he is compared to an otherwise average professional baseball player.  Both of these statistics have grown in popularity with the growing popularity of "Sabermetrics" in baseball.  The term "Sabermetrics" is in turn derived from another acronym (SABR) for the Society of American Baseball Research and was first used by the professional sports writer, Bill James.  You may be more familiar with Sabermetrics and the story of how the General Manager of the Oakland Athletics, Billy Beane, put together a winning team based upon these statistics (this particular story was the subject of the book, Moneyball, by Michael Lewis which was made into a popular movie starring Brad Pitt as Billy Beane).

The sport of baseball lends itself very well to statistics - even the most casual baseball fans are familiar with the commonly used statistics, such as Batting Average, Runs Batted In (RBI), and Earned Run Average (ERA).  Sabermetrics takes these relatively easy to understand and commonly used statistics one step further (okay, maybe two steps further) to incorporate "Big Data" techniques and advanced statistics and mathematical modeling into statistics such as VORP, WAR, and OPS (On Base % + Slugging %).  Sabermetrics really seemed to take off when General Manager Theo Epstein reportedly used them to build a Boston Red Sox team that won the World Series twice in five years, ending an 86 year World Series drought for the team (ending the "Curse of the Bambino").  Epstein would go in to put together a Chicago Cubs team that would win the World Series in 2016 for the first time in 108 years!

So there is no doubt that "Big Data" techniques and advanced statistical modeling methods such as Sabermetrics are here to stay, at least in regards to baseball.  But back to my original question - could something like "VORP" ever be used in health care?  Should we be using more advanced statistical methods to more objectively measure the effectiveness and/or productivity of our medical staffs?  The answer is clearly (at least in my opinion) "yes!"  Productivity measurements have traditionally used the amount of revenue generated by an individual physician based upon billing (specifically, something called RVU's, or work Relative Value Units are used to assess physician productivity).  There are, in fact, benchmark data for almost every medical specialty, that are stratified according to the region of the United States, among other factors.  Unfortunately, while the RVU method is commonly used, it is not very good.

The RVU's that an individual physician generates in any given day contributes to an incentive to see a large number of patients and/or provide a large number of services ("fee for volume" as opposed to "fee for value").  Have you ever complained that you didn't spend a lot of time with your physician?  Physicians are not incentivized to spend long periods of time with any one patient, largely as a result of how they are reimbursed for care and how they are being measured from a productivity standpoint.  In addition, RVU's says nothing about how productive an individual physician is in regards to time spent on teaching medical students or residents or time spent conducting research or writing academic papers.  In other words, perhaps we are incentivizing physicians for the wrong things!

So, what would VORP look like in medicine?  Ideally, we would first start with a set of objective measures for delivering safe, affordable, and excellent care that generates high quality outcomes (both in terms of quality of life and survival, as well as patient satisfaction).  These objective measures would hopefully reward physicians who spend all the time that an individual patient requires at the bedside.  Physicians are in a great position to identify ways to improve care using improvement science - in many cases, quality improvement is not factored into RVU's or other rudimentary measures of academic productivity.  Our medical version of VORP would need to consider how much time an individual physician contributed to improving health care delivery through quality improvement.  Moreover, physicians who were good teachers (for patients/families, nursing and allied health staff, medical students and residents) would be rewarded with a measure that took these important aspects of medicine into account.  Finally, physicians would also be measured (where appropriate) on how many academic papers that they published or for generating quality research that contributes to improvements in care.  Unfortunately, until these kinds of activities, all of which are both important and necessary to the future of medicine, are measured objectively, it would be hard, if not impossible, to come up with a medical version of VORP.

While I was certainly intrigued by my colleague's suggestion last week, I just don't think we have progressed to the point where something like "VORP" or "WAR" could be used to determine how important an individual physician's contributions are to any one particular institution. I would be ecstatic, and I think we will have come a long way, when we use something other than RVU's as a measure of physician productivity.