Multi-site operators love a league table. Which building has the lowest falls, the fewest staff injuries, the best occupancy? The temptation is to pull a column of numbers, sort it and send a message to the bottom three. That approach often does more harm than good. A 40-bed building and a 140-bed building will not have the same raw count of anything. A building with many short-stay rehabilitation residents faces different risks than a long-term memory care community.
Fair comparison requires adjusting for size and acuity first. Otherwise you may reward luck, punish the buildings that take the hardest cases, and teach leaders to distrust the data.
Imagine one building reports 12 falls in a quarter and another reports 5. Without context, the first looks worse. But if the first has three times as many residents, its rate may actually be lower. Counts reflect size. Rates reflect performance.
The most basic fix is to divide by a measure of exposure.
Resident days are the sum of the number of residents in the building each day. They are the standard denominator for events such as falls, infections or pressure injuries, usually expressed per 1,000 resident days.
Staff hours or full-time equivalents work for workplace injuries and turnover.
Admissions or discharges suit measures tied to transitions, such as readmissions.
Beds or census suit simple comparisons such as call volume or help requests.
Choose a denominator that matches what actually creates risk or workload, and use it consistently across all buildings.
Rates alone still treat all residents as equal. Acuity describes how much care residents need and how complex their conditions are. A building with more residents who need heavy assistance, have cognitive impairment or have complex clinical needs will naturally see more of some events.
Ways to account for it, from simple to advanced:
Group buildings or units by type. Compare short-stay units with short-stay units, and memory care with memory care.
Stratify. Calculate rates separately for residents at different levels of need, such as high, medium and low, then compare within each level.
Use a case-mix indicator that your organization already tracks, and divide or adjust accordingly.
Apply risk adjustment with a statistician or analyst if you need more formal comparisons.
Start with grouping and stratification. They are easy to explain and often enough to reveal the real story.
A small building with a handful of events can swing wildly from one quarter to the next. One extra incident might move it from best to worst.
Use rolling periods, such as 6 or 12 months, rather than single months.
Show the actual counts next to the rates, so readers see how few events are behind a percentage.
Avoid ranking buildings whose results differ only slightly.
Consider showing ranges or bands, such as "within expected range," "above" and "below," rather than exact ranks.
Comparison is meaningless if buildings count differently. One may record every minor fall while another records only injuries. Define each measure in writing, train staff, and spot-check entries. Confirm that data arrives from the same source and the same time period for every building.
Show each building's trend against its own history. Improvement matters as much as rank.
Compare to the group average and to a target, with clear labels.
Pair numbers with context, such as census changes, staffing disruptions or new admission types.
Highlight practices from buildings that do well, and invite them to share what they do.
Use benchmarking to ask questions rather than assign blame. "What is working at your building?" produces better results than "why are you last?"
No adjustment is perfect. Data quality varies and some differences cannot be explained by numbers alone. Treat a benchmark as a starting point for conversation and review, not a verdict.
Reliable benchmarking depends on clean, consistent data pulled from your clinical and operational systems. UnityCare IT helps healthcare organizations connect those sources and build fair, understandable dashboards. If your current comparisons spark more arguments than action, normalization is the best place to start.
An outsourced IT department with proactive maintenance and one number to call.
Call or text: 405-285-3845
New customers: start@unitycareit.com
Existing customers: support@unitycareit.com
Address: UnityCare Technologies, 2524 N Broadway Ste 554, PMB 947974, Edmond, Oklahoma 73034-4172