As I discussed in my first blog, starting an Observation Unit (Obs Unit) at our hospital has been rewarding, and has been a learning experience. I would now like to share some of the results and consequences. First, we set out and achieved our goals, which were:
- Decreasing the readmission rate.
- Increasing efficiency and creating a release valve for a busy inner city ED.
- Decreasing one day stays (they went from the high 30% range to 20%).
- Decreasing the average length of stay.
The new Medicare penalty for a 30 day readmission applies to readmissions for heart attacks, heart failures, and pneumonias. As part of this new policy, hospitals will be penalized if patients with the same diagnosis are readmitted within 30 days. By definition, an Obs status equals an outpatient visit, thus an Obs status is a way for the hospital to avoid potential penalties. In our experience, we gained efficiency but lost potential revenue and became victims of our own success.
A primary focus for our hospital in creating the Obs Unit was to decrease one day inpatient stays. Most of the “borderline patients”-- patients who were in between an easy inpatient diagnosis and Obs status-- who deteriorated in the Obs were converted to acute inpatient status after 24 hours. We found that 20% of our Obs patients were being converted to acute inpatient status every month. We went back and reviewed some of these borderline patients, and we found that a large percentage of these patients actually met the Interqual Criteria for acute admission from the beginning. This meant that they should have been admitted as an acute inpatient and not have gone to the Obs at all. The end result was that the hospital lost payment for a day of inpatient treatment for each patient in this 20%, resulting in an overall loss of tens of thousands of dollars of revenue.
Some of our borderline patients were heart failure patients--acute on chronic exacerbation, atrial fibrillations with rapid ventricular response needing drips for a short time, syncope with orthostatic changes, complex pneumonias in young people, most of TIA, and possible but not acute CVAs. Also included were most chest pains even including high risks/ high TIMI scores, etc. We found that these and other borderline patients got good care in a shorter amount of time and were very satisfied, since they ended up having more face time with the physicians and nurses in the Obs Unit than on the medical floors. By taking care of these borderline patients, our hospitalist developed a high tolerance to the severity of illness in the Obs Unit and streamlined the discharge process. On the other hand, the hospital ended up with the loss of revenue for many of these patients because they did meet the acute inpatient Interqual criteria.
After two years of having an Obs Unit, we have learned that since there is not a national benchmark for the percentage of Obs patients, maybe our goal of 20% on one day stays was too drastic. If patients meet the Interqual criteria of acute inpatient admission, it can be OK to admit them even if they stay just 24 hours, as long as there is documentation to prove the necessity. It will be interesting to see how each hospital plans for the payment changes that the ACA and CMS are bringing and how they deal with the readmission penalties.