Create Better Peer Groups with New PINC AI™ Benchmarking Methodology

Key takeaways:

• The use of hospital peer groups for comparative analysis is well established in the healthcare industry.
• Peers should be chosen based on criteria that explains a significant amount of the variation in a particular performance metric.
• PINC AI’s new benchmarking methodology leverages data on thousands of hospital characteristics and with artificial intelligence (AI) identifies the most relevant peer groups for a particular measure.

For more on this topic:

• Talk to an expert about how benchmarking methodology impacts PINC AI™ solutions.

There is a growing need for healthcare organizations to focus on improving health outcomes while simultaneously cutting costs. Many healthcare organizations utilize benchmarking to ensure their value-based care goals are on track.

Benchmarking is the process of comparing an organization’s performance to a common standard, such as those established by the Centers for Medicare and Medicaid Services (CMS) and other regulatory agencies. Benchmarking is used in healthcare to improve efficiency, quality of care, patient safety and satisfaction.

Peer Groups Provide the Foundation for Effective Benchmarking.

Insightful benchmarking requires careful selection of peer groups. A peer group is a collection of organizations that share similar characteristics and interests among one another. Peer groups are useful because they allow providers to compare themselves to others to identify performance outliers. With this being said, many healthcare organizations turn to PINC AI™ for its benchmarking capabilities. The technology and services platform is a gold standard in benchmarking and analytics thanks to an alliance of more than 4,400 hospitals and health systems in the U.S and a clinical database representing 45 percent of the nation’s hospital discharges.

Benchmarking to assess performance, find performance gaps and pinpoint areas for improvement is important for organizations embarking upon a quality improvement journey. The challenge is that peer groups often choose comparators using superficial characteristics, such as teaching status, size, location, etc. Not only do these designations sometimes slice and dice groups so finely that they yield a very limited peer group, but they fail to take into account whether these types of demographic characteristics are the ones that truly matter for performance.

To counter this potentially undesirable characteristic of peer groups, PINC AI’s data science team developed a new, patented benchmark methodology that augments experience and intuition with statistical feature selection, leveraging an extensive database of hospital characteristics. The process is built on a three-step process:

1) Identification of the specific characteristics that explain variation in the performance metric of interest.

2) Statistical clustering techniques identify peers that are similar in those ways that matter to performance.

3) Assessments of peer group quality to ensure meaningful insights versus a random sample of hospitals.

This new benchmarking methodology utilizes AI for peer selection and to better measure management and data-driven decision making for improvement.

Peer group benchmarking is a complex process. Here’s how to get it right.

The use of hospital peer groups for comparative analysis is well established in the healthcare industry. Peers should be chosen based on characteristics that account for a large portion of the variation in a given performance metric. This innovative approach, based on hundreds of thousands of PINC AI™ data points about hospital characteristics, simplifies the process of choosing criteria for peer selection and adds consistency and transparency to the selection of peer group themselves.

AI-driven peer grouping methodology ensures that hospitals compare themselves to peers that are similarly situated in ways that matter for a particular performance metric. It further mitigates potential bias (unintentional or otherwise) that can creep into the process of choosing peers. The data science team driving innovation within the PINC AI™ technologies is using AI to determine true “like” characteristics and letting data tell the story for those hospitals seeking performance drivers through comparisons with true top performers.

Managing Labor Benchmarking Metrics with PINC AI’s OperationsAdvisor®.

PINC AI’s OperationsAdvisor provides a powerful platform for organizations to identify and manage appropriate staffing levels across their organization, using productivity and benchmarking metrics. OperationsAdvisor is one of the largest healthcare operations databases in the country, containing benchmarks for over 675 subscribing facilities across the U.S.

Selecting a peer group can be difficult because there’s no one-size-fits-all approach that applies to all healthcare organizations. However, with the help of PINC AI™ Margin Improvement, healthcare organizations won’t miss out on the vast amount of notable and actionable insights that could help them improve.

The benchmarking module in OperationsAdvisor includes departmental characteristics to provide a more holistic view of operations. These characteristic profiles have been utilized to develop a patent-pending “likeness” scale to identify peer departments that are most similar. Custom peer groups can be created to increase benchmark buy-in from operating managers.

With OperationsAdvisor, users can also get an in-depth look at top performers to learn what makes them unique which could help explain the differences in performance and help them improve. At the end of the day, peer groups are an excellent tool for determining where an organization stands in respect to their competition. With this information at their disposal, healthcare organizations can better set their expectations and create realistic goals.

The insights you need to stay ahead in healthcare: Subscribe to Premier’s Power Rankings newsletter and get our experts’ original content delivered to your inbox once a month.