When trying to understand a particular issue — whether related to cost, quality, or the patient experience — hospital administrators and health care policymakers can face a deluge of different data and data sources. This data, however, will almost always be condensed with a convenient summary metric, namely, an average. Average length of stay, average boarding time, average time to discharge, average daily census: averages are inescapable.

In its March report to Congress, the Medicare Payment Advisory Commission (MedPAC) recommended changes to the payment rates for inpatient hospitals, in part based on a nationwide average occupancy rate of 62.5%. MedPAC cited this low occupancy rate and a trend of declining average occupancy rates as evidence of excess inpatient capacity. In other words, the United States, on average, has too many hospital beds.

But averages can be viciously deceptive. As the bromide goes, you can drown in a river that’s on average only 6 inches deep; the average depth will mask the fact that part of the river is one inch deep and another contains a torrential current dipping down 20 feet. Although an average provides a convenient snapshot of the river, the distribution or variation in depth might make one wary of attempting to swim across.

At the policymaking level, using average occupancy rates to judge whether there are too many or too few hospital beds can quickly become a thorny issue. Rural hospitals often have much lower occupancy rates than hospitals in urban areas, for example. Reducing the number of hospital beds or closing hospitals altogether, however, causes an entirely different set of problems related to equity and access, where patients may need to travel long distances to receive care.

Conversely, hospitals operating at or close to their bed occupancy capacity may have trouble accommodating emergency admissions, leading to emergency department boarding. High average occupancy rates, per se, may not even reflect a hospital’s need for new resources. Despite anticipating the need to expand its capacity due to perceived high occupancy rates, Cincinnati Children’s Hospital improved its efficiency and productivity using patient flow methods instead, boosting occupancy rates without initiating a new capital project.

The persistent problem of variation rears its head in many places at the hospital level, too. Average occupancy rates can vary daily, since more people are admitted on weekdays than on weekends; monthly, if there are seasonal effects, such as “snowbirds” migrating to Florida during the winter; and even yearly, due to macroeconomic or external factors.

Preparing for averages at the hospital level raises similar issues to those raised at the policy level. Staffing to average daily census can leave providers overworked if census fluxes high, but eat into budgets if census fluxes low, as hospitals pay for unneeded staff. And, in worst-case scenarios, emergencies such as natural disasters or the closure of a nearby hospital could overwhelm an organization that only prepare for averages.

Variation is an inescapable fact of life, one that can have significant consequences on the quality of care hospitals provide in addition to the quality of life experienced by its physicians and nurses. Averages won’t go away anytime soon; they’re too ubiquitous and too convenient to disappear overnight. But with the ongoing changes in payment policy, hospitals should begin to acknowledge variation and integrate the appropriate assumptions into their planning processes to secure the health of their patients, the morale of their workforce, and the future of their finances.

Use analytics, machine learning, and simulation technology to better manage patient flow. Learn more.

Author

 

Subscribe to the Hospital IQ Newsletter now to receive our latest content like this right in your inbox!