Joe Zebrowitz MD, started the day talking about medical necessity and observation status. A big challenge to observation is the different rules for Medicare, Medicaid, and managed care - keeping all these straight is problematic. He presented that the typical attending physician doesn't really know what "observation status" really means. They care about how observation will impact their patients:

  • Access to care
  • Are concerned about how it may impact their reimbursement, and
  • Are oblivious to the impact of observation status on the hospital.

Ensuring that physicians are educated and supportive is key. His data, based on about 16,000 cases reviewed, shows that observation is over used on average 45%. If the ALOS is less than 24 hours, the over use is closer to 35%, and if the ALOS is over 24 hours, the over use is closer to 55%. For every patient that is put in observation inappropriately, you've wiped out $100,000 of hospital revenue.

Proper case management is dependent on both a good process and 100% review of every case. Many hospitals are putting case managers in their EDs to review cases, but many get steered into a social worker role doing complex discharge planning. Process variability is a common theme among many of the presenters at this conference. Joe offered a great approach to reducing variability.

Joe also introduced a new term to the audience, the retrospectoscope, a device that allows the user to look back in time to "improve" decisions made in the past.

Kathy Tyrrell, Case Management Director, Prince William Hospital, talked about a role at their hospital, the bed control ESD case manager. This role assesses the appropriate level of care and appropriate status. Reinforcing the previous presentation, Kathy described a great process, and less than perfect results came from process variability. As a small community hospital, she provided a very realistic approach to improving the observation process.

The last presentation that I caught, was by Chris DeFlitch, MD. He compared and contrasted clinical decision units with observation units. At his hospital, justification for the CDU was based on improved
patient flow. Hershey Medical Center was at 100% capacity, and they
could either board patients in the ED or try an observation unit. They
built an 8 bed unit (Hershey has just under 500 beds overall). Like
many other presenters, DeFlitch reported a heavy emphasis on the tight
protocols and disease specific guidelines.

They had a great foundation;
a bueautiful new space (8 beds, 2 RNs) - but RN hires didn't wan to be an
"inpatient nurese" and they had inconsistent MD buy-in. The size
of the unit and patient volume meant RNs and docs were shared with the ED - the
staff resisted adoption of the obs service. What they learned was that implementing the Clinical Decision Unit was not just a new space and procudures, but a transformational change
that they didn't really address.

Observational medicine is different
and requires a considerably different As an aside, the CDU is now also
used surge capacity, minor emergency crowding and boarders. In addition to these changes, observation patients are also allowed in other units/services. Patients must come in through the ED. If the Internal Medicine attending wants to manage observation patients, that's fine too, but there's oversight to ensure they have the appropriate hour-by-hour focus to ensure that care and disposition progresses in a timely fashion. Emergency Medicine observation patients have an ALOS that's 6 hours shorter than the other services - the important point, of course, being that ALOS is tracked and reported monthly by service. Flitch wrapped up with a great description of how they overcame the inevitable implementation problems.