Have you ever tried to drive change in your organization? If you have, then chances are that you must have encountered some resistance. Perhaps colleagues told you ‘we’ve always done it this way’ and were reluctant to consider your perspective or ideas.
This happened to me early in my career as a data analyst at an insurance company. Rather than put my head down and just follow the statistical methodology they had in place to measure marketing campaigns, I decided to use their data to show where their methodologies were delivering inaccurate outcome projections. I learned early on that an effective method for executing change is to support your proposal with proof in the form of data. If you can produce evidence as to why a change is needed, you’re more likely to experience success.
This is especially true in health care where change is often slow and difficult because evidence of outcome is king. So how do we speed up performance improvement in care while also meeting the need for evidence? Through data, or to be more exact, predictive risk modeling.
At Vizient, we devote a substantial amount of time and effort to create accurate risk adjustment models for patient outcomes. We have three outcome variable models – namely for mortality, length of stay and direct cost. Each is then applied by condition type, such as heart failure, stroke or malnutrition. Then for each clinical condition, which is its own unique model, a unique set of variables exists.
Today, we have more than 330 risk models and there are more in development. These risk models will accurately bring the hospital’s patient population on a similar level and provide guidance for where to improve their quality metrics. The risk adjustment requires a rigorous process of variable selection. This process and predictive model building ensures that our risk adjustment models offer predictive power as well as reasonable interpretability for physicians and hospitals so they can devote their limited resources to improve performance, maximize efficiency, produce favorable financial results and positively impact patient outcomes.
Expecting change to occur without the support of data is a futile exercise. But by using the power of data, hospitals and their physicians and clinicians can implement meaningful change in their operations and caregiving. I firmly believe that adopting a scientifically data-driven approach is the best way to help your hospital improve performance and make a difference in patients’ lives.
About the author. In his role as director of advanced analytics and decision sciences at Vizient, Ajmani leads a team of data scientists who address data challenges faced by member hospitals. His more than 20 years of experience in applying statistical methods to solve industry issues includes roles at such well-known companies as Eastman Kodak, 3M, Ameriprise Financial and United Health Group. Ajmani also has more than 20 years of teaching experience. Most recently, he was adjunct faculty at Northwestern University in Evanston, Illinois, where he won the Distinguished Faculty of the Year Award in 2013. He authored the book, Applied Econometric Using the SAS System.