When I first began my journey with outcomes back in 2000, I had a very simplistic impression: patient comes in and receives services and a final outcome is captured. Patients were equated to widgets (like a manufacturing company). *Something* happened during rehabilitation - it was all about me and what I chose to do during the episode of care. How far from the truth that was! What I have grown to realize is that there are many factors that contribute to the final outcome. I'd like to delve into them a little bit because they are very important to consider when it comes to value based purchasing.
The reason it is important to identify factors that affect clinical outcomes is because when a patient walks in your doors to receive services, the whole episode of care is more than just the intervention and the outcome. All sorts of interactions and variables come into play that affect the final outcome. As we move into value base purchasing, we need to understand the impact of the interactions and variables on the final outcome. The interactions and variables need to be controlled by risk adjustment. Risk adjustment is often viewed as leveling the playing field and allowing for apple to apple comparisons. In my opinion, what risk adjustment should do is take the patient and me out of the equation to allow a better assessment of the treatment and the final outcome. When you look at the diagram above, it makes total sense why the risk adjustment process can typically only explain 12-20% of the variance at best. Not every single identified variable can or will be included in the risk adjustment process... and as you can see, random events don't even have any variable listed (the unknown factors).
1. The Missing Factor - Clinician Factors: I truly believe there are clinician factors that affect outcomes. There have to be. Larry Benz has done some work in the area of empathy. Linda Resnik has done work in the area of expert. Some of the Evidence in Motion team have done work on residencies and fellowships. Think of some of the most non-evidence based individuals who provide "care" to patients. They achieve results: results that are gained mainly due to their charisma, patient belief systems and the relationship created. For the moment, many systems do not consider these factors. Until these factors are considered and controlled for, we have an awesome opportunity to tip the scales in favor of positive outcomes by doing all we can to maintain excellent clinician factors that positively affect outcomes.
2. Patient Factors: Patients definitely affect outcomes. Look at all the variables shown in the image! When it comes to assessing outcomes, this is the area that most work has been focused. All good systems will risk adjust as many of these variables as possible. If you use FOTO, you can easily see that quite a few of the patient factors are part of the risk adjustment algorithm. FOTO currently risk adjusts for gender, age, duration of symptoms, surgical history, medical comorbidities, payer type, severity and fear avoidance beliefs.
3. Treatment Factors: This is the area that we really want to be the focus when it comes to the final outcome. Treatment would focus on who delivered the care along with the provided care. It would also encompass the engagement of the patient. What treatment gives the best outcome? In order to know, all the other factors have to be accounted for, right? In other words, if we can identify and control for all the other factors involved in an episode of care, the target focus could be on the treatment and the outcome.
4. Random Factors: I'm not sure how random is random. Maybe things like how the patient chose your services or whether you serve coffee and doughnuts have an effect? Maybe the patient's impression of your company logo has an effect? Maybe the credibility of a word of mouth referral has an impact? Maybe the influence your clinic has in the community has an impact? What about the color scheme in your clinic or the odor in your clinic? From a statistical perspective, I'm obviously not perceiving "random" very accurately. I'm interpreting "random" like the young adults in my household would perceive it.
As the payment model tips toward value based purchasing, there will be more science involved in determining better and better risk adjustment algorithms. The goal, as much as possible, will be to really create a treatment/intervention = outcome sort of an equation.
If you are looking for a product that currently has the most sophisticated risk adjustment process known to date, look no further. Please talk to Judy Holder about how FOTO can help.