Avoid Benchmarking's Bear Traps!

On our website you'll find a discussion of online "real time" benchmarking. IMPACT!demonstrates just how far benchmarking has come from the "good old days" of Monitrend. (Editor's note: If you don't remember Monitrend, the author believes you are far too young to hold a responsible healthcare management position). Although that ancient and noble first effort at benchmarking sometimes gets a bad rap today, it really deserves a tip of the hat. Monitrend was the first serious effort to provide the field with benchmarking data using computer technology.

In any case, the occasion of IMPACT's introduction provides a perfect excuse to revisit benchmarking's basic foundations and review its successes (and occasional miscues) as a management tool.

What is a benchmark?

The beginning is always a good place to start any endeavor, so let us first address the most fundamental of all benchmarking questions which is, "What is a benchmark?"

According to The American Heritage Dictionary of the English Language (© 2000 Houghton Mifflin Company), a benchmark is "a standard by which something can be measured or judged." In the case of most hospital benchmarking efforts, that standard is derived by observing the performance of other hospitals in whichever area is under study.

The distinction between a "benchmark" and a "management goal" is an important one. A benchmark comes from outside the organization. In the case of labor productivity benchmarking, for example, the benchmark is based upon the reported staffing performance of other hospitals. A management goal, on the other hand, is established by the hospital.

What can benchmarking tell us?

Some management teams have misunderstood the question that benchmarking can answer. That question is not, "How should Department "X" at my hospital be staffed?" It is obvious upon reflection that a staffing benchmark generated by an aggregation of departments in other hospitals can't say much about how a department in your hospital should be staffed. In fact, assuming peer groups are selected appropriately, the only question that benchmarking can answer is, "What is the range of observed staffing in reasonably similar departments in reasonably similar hospitals?" The implications for management action that arise from the differences between these two questions are profound.

Department "X" in any hospital is a system. It is axiomatic that any system performs exactly in the way it is built to perform. It cannot be otherwise. How the system you call Department "X" in your hospital should be staffed is a function of hundreds of interconnected elements. Now it may well be the case that there are "broken" elements within Department "X" that prevent optimum performance. However, if Department "X" is operating with a broken element, that broken element is as much a part of how Department "X" is built at the moment as is any other element.

Benchmarking is a powerful management tool that has helped many hospitals achieve significant cost reductions (millions of dollars annually in some cases) while facilitating improvements in quality, physician and patient satisfaction, employee relations and community image. As with any powerful resource, its misuse can bring devastating consequences.

One of the most common serious mistakes that hospitals make is to impose benchmark-based department productivity standards without knowing whether the established standards will really allow work to be done at acceptable quality levels. When a department's performance varies significantly from valid benchmarking comparisons, it is imperative that the reasons for that variance be identified and resolved before a potentially crippling standard is imposed. Benchmarking is an indispensable, even vital, management resource but it is the beginning of an improvement process, not the end of one.

How often should benchmarks be updated?

That depends upon who is going to use the benchmarks and for what purposes.

At the department manager level, this question helps drive home the reason for the distinction made earlier between external benchmarks and department productivity standards which are internal management goals. Assuming reasonably stable department workloads, a systematic comparison of management's productivity standards against external benchmarks should be accomplished at regularly scheduled intervals, say twice a year or so. Conversely, if department workloads are volatile or trending sharply in one direction or another, the comparison should be done more frequently to assess whether management's productivity standards need to be changed. While it is true that observed staffing practices change over time because of new technology, increased medical knowledge, regulatory changes, and a host of other factors, bombarding department heads with constantly changing benchmarks may unintentionally discourage them from making good use of data and actually work against effective cost management.

Power users, on the other hand, require more frequent access to current information. For example, Decision Support staff need the most current data available when putting together "what if" scenarios for budgeting, acquisition and consolidation planning, and new service development. In a similar vein, because the relationship between workload and staffing requirements is not linear, Human Resource planners need to apply the most current activity-based benchmarks against historic utilization swings to develop core staffing plans that project the best possible mix of full-time and part-time staff given the range of anticipated workload.

FJB