Why behavioral analytics will be crucial for hospital performance

Analyzing data can help healthcare organizations stratify providers and apply limited coaching resources where they will have the greatest impact.


To understand the difference next-generation analytics can make in improving overall physician performance, consider a football team whose offense is scoring an average 40 points per game. Even in the pass-happy NFL, that should be enough to win most of their games and make a deep run in the playoffs.

Unless, of course, the defense is giving up an average of 45 points a game. You don’t have to be a math whiz to understand why that team won’t even sniff the playoffs.



As the coach, you have two options to turn it around. One is you can try to coax another six or seven points out of your high-performing offense. That may not be so easy, however. The other option is to use all the means at your disposal to figure out where the holes are in your defense and improve it by changing the scheme or getting better players. It should be an easy choice.

While hospitals and health systems may not have the large scouting departments, sophisticated video equipment and staffing to break down every procedure “play-by-play,” one thing they do share with professional sports teams now is access to advanced analytics. The right analytics solution makes it easier to determine who the high-performers, just-getting-bys, and laggards are so the organization can invest its “coaching” time and resources where they will have the greatest impact.

Most healthcare organizations today are using analytics for basic provider profiling. They look at the efficiency of each physician in terms of quality and financial utilization. Certainly this is good information, especially since it helps identify the outliers, which enables the organization to manage risk more effectively.

What the current provider profiling doesn’t do, however, is dig deeper to help the organization determine where to best focus its coaching time and attention. As a result, that coaching tends to be spread equally across all providers.

A better approach is to use behavioral analytics measured against industry benchmarks to determine which providers are performing at the top of their games, which are under-performing, and which are in-between those two extremes, and then determine why.

Following this strategy will enable the organization to focus coaching resources where they will have the greatest impact. It will also ensure the organizations is not expending resources either where they’re not needed or where they won’t have an impact.

The issue with standard provider profiling is that it is focused on outcomes, e.g., changes in the patient’s condition, the cost of a procedure or episode, patient satisfaction scores, and so on. What it doesn’t address is how these outcomes came about, which could have a significant impact down the road.

Behavioral analytics go deeper to uncover the process being used to achieve those outcomes, such as whether providers are following the healthcare industry’s or the healthcare organization’s best practices.

For example, a provider may have a good track record for knee replacements on the surface. But when you dig deeper you find that he/she is not using the organization’s preferred components and is sending most referrals out-of-network. Taken together, there is a high probability that this individual will need considerably more coaching from the organization than another provider with similar outcomes who is following the organization’s best practices.

By taking hundreds of data points into consideration, advanced behavioral analytics can be used to assign a score to each provider. The organization can then stratify those providers to set the level of attention each provider needs to receive, helping ensure that its limited resources will be applied where they will do the most good.

High performers may only require a “touch base” now and then. This frees up resources to spend more time coaching up the “just-getting-bys” in an effort to elevate their performance to the next level.

Initially, the organization may also want to dedicate a higher level of resources to the laggards as well. If their performance does not improve over time, however, it becomes obvious that their performance will likely never reach organizational standards and dedicating resources to performance improvement is a waste of time and money.

From there, the organization can decide whether it wants to continue the relationship and live with that level of performance or replace that provider with one who will meet organizational standards.

Given that allowing some providers to fall below standards will likely lead to others following suit, replacing the laggard is generally the best option, especially as the industry continues to transition to value-based care. Just as NFL teams use analytics to upgrade their rosters through the draft and free agency in order to win championships, hospitals and health systems can use behavioral analytics to build a roster of providers to help them meet the Triple Aim.

Winning in the current healthcare environment takes more than a few individual stars. It takes an entire team dedicated toward delivering the highest quality care at the lowest reasonable cost. And while you’d be hard-pressed to find a provider who would disagree with that assessment out loud, actions speak louder than words.

Using behavioral analytics to stratify providers enables hospitals and health systems to understand their providers better and apply their limited coaching resources where they will have the greatest impact. It’s the smarter way to work.

David Hom is chief evangelist at SCIO Health Analytics

More for you

Loading data for hdm_tax_topic #better-outcomes...