The Unintended Consequences of Observation Units

I recently worked with a 440-bed county hospital that cares for a large number of Medicare and Medicaid patients. To lower its average cost per case and reduce potential penalties for unnecessary admissions and readmissions, it created a 22-bed observation unit staffed by a top-notch hospitalist group.

For a while, the observation unit appeared to be doing a great job. Average length-of-stay for these patients was down to 16 hours, and 70 to 85 percent of patients were discharged to home. This setup allowed the hospitalists to closely manage these patients and to coordinate excellent follow-up care.

All in all, the observation unit solved many of the problems it was intended to. But now there were new problems.

Problem #1: Revenue (or lack, thereof). Many of these patients (who under the old system would have been admitted) were being very successfully managed and discharged from the observation unit. The hospital began to realize a loss of revenue to the tune of $7 million dollars a year — far more than it would have lost to any Medicare penalties imposed.

Problem #2: Medicare patients were not happy. Patients who did convert to inpatient from the observation unit and were discharged from the hospital two days later did not qualify for post-acute care, as that one day in observation did not count toward the three-day minimum requirement. Also, Medicare patients who were discharged from the observation unit were hit with large out-of-pocket copayments, because Medicare classifies observation as outpatient care (covered on a fee-for-service basis by Medicare Part B).

After an initial surge in enthusiasm for observation units, hospitals are starting to feel the sting of unintended consequences. Observation units are a well-intended concept, but under current Medicare regulations, very few hospitals have been able to make them work.

Concerns have intensified to the point where the American Hospital Association (AHA) has sued the Department of Health and Human Services to eliminate observation status altogether.

To illustrate some of the problems involved, let's consider a hypothetical patient named Grace. She's 66 years old and comes to the emergency department complaining of tightness in her chest and shortness of breath. Her initial EKG is normal — however, her cardiac enzymes are slightly elevated. The ED physician decides to put her in an observation unit so that additional testing can be performed (and to get her out of the chaotic, noisy environment of the emergency department).

Grace's tests come back normal, and within 15 hours, she's discharged to home. The hospitalist coordinates a next-day follow-up appointment with her primary care doctor.

You'd think Grace would be pleased. After all, she's resting comfortably in a bed, her care is a high priority and she's out of there in a flash. (On average, observation speeds up discharge by a whopping 38 hours.)

The hospital's also happy. By keeping Grace in observation, it cut its costs by at least half and avoided Medicare penalties for one-day or unnecessary admission.

But in reality, Grace is unhappy. Thanks to growing awareness among patients (fueled by incisive AARP and NPR reporting), she is very aware that observation care can be a bad deal for her financially. She spends her entire visit fighting to be admitted.

When she leaves, her patient satisfaction survey reflects her displeasure — despite her excellent care and clinically positive outcome.

And the cost savings the hospital achieved by keeping Grace in observation are more than offset by the loss in their reimbursement. (On average, Medicare pays $3,926 for a three-day inpatient stay versus just $1,489 for observation.)

Even Grace's ED doctor is somewhat financially dis-incentivized. He followed her in both the ED and the observation unit, but can bill Medicare only once (even though he performed two levels of service within the same 24-hour period).

Given these realities, what's a hospital to do? How do we make observation — which still makes sense from a cost and patient care standpoint — workable?

Change may be coming, albeit slowly. CMS' new "two-midnight" rule, which goes into effect in 2014, will likely renew interest in observation units as a solution for those "gray area" cases. However, until the reimbursement rules change, it's uncertain whether observation units will be a long-term solution.

One thing's for sure: hospitals and consumer groups will be keeping an eye on the coming changes, which may determine the fate of observation units.