Newsroom

Transparent Risk Modeling Clears the Path to Performance Improvement

06/13/17

A small, 300-bed health system in western Illinois wanted to see how it benchmarked against similar community hospitals for chronic obstructive pulmonary disease (COPD) readmission rates. The process began by working with their data team and analytics solution to create a benchmark report that would inform a performance improvement initiative if necessary.

For many hospitals needing data to inform and guide improvement initiatives, the benchmark report would have been a masked combination of academic medical center and community hospital data with risk modeling not ideally matched to the patient population served by the health system in Illinois.

For this facility, the Vizient Clinical Database (CDB) offered the option to select a specific compare group with complete data transparency and specific risk modeling for credible and insightful benchmarking. 

“Since its inception more than 30 years ago, the CDB has evolved to become the definitive analytic platform with over 110 AMCs (representing 97 percent of the nation’s AMCs), more than 160 community hospitals and over 50 heath care systems nationwide contributing data and pulling metrics to guide their performance improvement initiatives,” said Beth Godsey, senior director, data science, methodology and national imperatives at Vizient.

The CDB arms its members with transparency of methodology, giving credibility to its data. That transparency, coupled with the ability to select which hospitals are in a compare group and the visibility into more than 990 clinically specific risk models, significantly reduces the barriers between health care professionals accepting, embracing and leveraging that data.

Risk adjustment modeling is a key differentiator for the CDB and helping guide its development is the Risk Adjustment Task Force, one of several Vizient work groups who provide input to ensure the tool continues to deliver value to members. This particular group connects CDB users to some of the most sophisticated and world-renowned clinicians, physicians and statisticians in the industry.  

Recently, Vizient has started to incorporate lab results into its risk adjustment models that allow for gradations of a patient’s condition. For example, malnutrition is measured more accurately on a scale as opposed to a ‘yes’ or ‘no’ result. While this is a work in progress, the task force is providing feedback on how to capture the true clinical comorbidities of the patient or the patient’s conditions – enabling Vizient to risk adjust more precisely.

“The underlying foundation of what makes our risk models extremely valuable is that we create a sense of homogeneity around the data, because Vizient uses similar hospitals in that model-building process,” Godsey said. “Other organizations – and even CMS – put AMC and community hospital data all together, and what we’ve seen is the coefficients aren’t as effective or accurate as they could be if the data was split out separately.”

Overall, Vizient has three outcome variable models for mortality, length of stay and direct cost. Each is then applied by condition type, such as burns, heart failure and stroke. Then for each clinical condition, which is its own unique model, a unique set of variables exists – all of which are transparent to members.

As Vizient develops new risk models, the task force is reviewing and providing feedback. “We look to the task force members to answer questions such as: Do these seem clinically relevant for you in a mortality model? Do these seem relevant in the patients that stay longer? Do you feel like that they would have these kinds of clinical conditions?” said Godsey.

Equally important as the risk models themselves is the fact that all CDB members have access to the methodology and transparency into how they are constructed.

“We want CDB members to see our risk models, processes and methodologies,” Godsey said. “We have an open-door policy which means that members can come to us at any time with questions and feedback. That openness enables us to explore ideas or improvements we might not have otherwise considered.”

For more information on how the Vizient CDB gives your organization transparency that facilitates performance improvement, read the white paper, “The Power of Transparency in Clinical Analytics.” For information on how your organization can begin utilizing the CDB solution contact Beth Godsey.

Multimedia Files:

Preview image