I remember the first time our CEO announced to our board of directors that we had received the Top 100 Hospital designation. I believe the award was “owned” by Solucient at the time. It actually came as a shock because we were not expecting it. I was VPMA back then. We were just vaguely aware of the hospital performance measures it used.

emergency hospital performance

The CEO and board were very proud of the designation. We all were. It was a welcome recognition of all of the hard work we had been doing.

Our executive team met on our usual Wednesday operations meeting shortly thereafter. We reviewed the data presented with the award, and determined that several factors had come together to help us achieve it.

We had started an aggressive length of stay initiative about 3 years prior. In addition, we had implemented a nurse documentation and coding program.

Finally, we had implemented a quality measurement tool from CareScience (now owned by Premier and called QualityAdvisor) several years earlier. And we were collaborating with our affiliate hospital, Rush University Medical Center, to work on quality projects together. We had seen some impressive improvements in mortality and complication rates, and reductions in length of stay as a result of some of the projects.

The metrics being considered have changed significantly since then. When this all started, present on admission (POA) coding had not been implemented, so it was very difficult to distinguish co-morbidities from complications when analyzing closed charts. And CMS core measures were not yet being tracked. A lot has changed.

In July of this year, I wrote a post that described the steps for achieving Truven's Top 100 Hospital status. In it, I listed the metrics that Truven used to rank almost 3,000 U.S. hospitals using its balanced scorecard. The list of hospital performance measures had changed little over the previous decade.

When I reviewed the methodology for the 2016 report, I found several significant changes in the performance measures. The notes also mentioned that there are more changes to come.

Top Hospital Performance Measures

Here is the list of measures used to rank hospitals in the most recent release of the Top 100 list:

1. Risk-adjusted mortality index

This is a measure of survival in your hospital compared to other hospitals.

2. Risk-adjusted complications index

This is a measure of “error-free” care.

3. Core measures mean percent

This item has been used in the past. It is based on hospital use of care guidelines from CMS. But the core measures change over time. In 2015, Truven used core measures for heart attack (AMI), heart failure (HF), pneumonia, and surgical care. In 2016, Truven switched to core measures for stroke care and blood clot prevention.

4. 30-day risk-adjusted mortality rate for AMI, HF, COPD, pneumonia and stroke

This is unchanged from 2015, except for expansion to include COPD and stroke.

5. 30-day risk-adjusted readmission rate for AMI, HF, COPD, pneumonia, hip/knee arthroplasty and stroke

This is unchanged from 2015, except for expansion to include COPD and stroke.

6. Severity-adjusted average length of stay (ALOS)

This is a measure of efficiency and indirectly of quality. Patients with complications remain in the hospital longer. And those with extended stays tend to experience complications such as falls and infections.

7. Mean emergency department (ED) throughput

This is a new hospital performance measure. It is calculated by averaging the mean throughput times for door to admission, door to discharge, and time to receipt of analgesics for fracture care.

8. Case mix-/wage-adjusted expense per discharge

This is a measure of cost efficiency.

9. Medicare spend per beneficiary (MSPB) Index

This is a somewhat new measure. It is adjusted for various factors. It is used to compare the cost of care from 3 days prior, to 30 days after, the index admission.

10. Adjusted operating profit margin

The most basic financial metric, operating margin is a measure of management's ability to operate within current fiscal restraints.

11. HCAHPS score

The  Hospital Consumer Assessment of Healthcare Providers and Systems is a patient satisfaction survey required by CMS. The Truven measure is based on the patient overall hospital rating question only.

Note the following changes:

  • The patient safety index was removed. According to Truven, it eliminated this measure because of “concerns that many of the metrics reflect inaccurate coding and documentation rather than adverse patient safety incidents.”
  • Truven changed the way it applies the POA (present on admission) indicator. It accepted all principal diagnoses as POA. And it accepted any secondary diagnosis as POA across the board if more than 50% of hospitals coded the diagnosis as POA.
  • The mortality risk model excludes all records from mortality measures in which ICD-9 code v49.86 is present on admission (POA).

Other Considerations

My observations for optimizing hospital performance under this balanced scorecard are as follows:

  1. As usual, risk-adjustment is a big deal. Risk-adjustment is what determines how actual hospital performance looks when compared to expected performance. The more complete the risk-adjustment, the better your hospital performance will look. And missing appropriate risk-adjustment information will make your outcomes look crappy.
  2. Risk-adjustment depends on capturing secondary diagnoses, especially those present on admission. So, it is critical that physicians include all treated and historical diagnoses on the admitting history. It is also helpful for nurse documentation specialists to identify any missed diagnoses by reviewing old records and other information.
  3. There are many inclusion and exclusion criteria for the performance measures. These should be well understood and identified when present. The best example is the accurate identification of DNR status. The v49.86 code must be present in the record so that terminally ill patients are NOT included in mortality calculations by Truven.
  4. It is crucial that someone on your team understands the methodology. Then you can educate other team members about changes in the measures and issues described above.

dashboard hospital performance

Why I Like Truven's Top 100 Balanced Scorecard

Some physicians do not like rankings like these. They believe it promotes gaming the system, rather than improving care.

But, if we don't have validated, published quality metrics, how will we know if we are underperforming? And how will our patients know? And what incentive will we have to change, if it is not demonstration of our performance gaps?

I also like the idea that this scorecard includes quality, satisfaction and financial performance. What better way to demonstrate to our stakeholders that we are doing a good job, than by demonstrating that our care is good, our length of stay is low, our patient experience is positive and our financial performance is superior?

Finally, I like the fact that of the current measures, the Chief Medical Officer should have direct oversight of the first five or six, and will influence the others.

Next Steps

Take a minute to scan through the methodology section of Truven Top Hospitals 2016 if you're interested in learning more.

Contact Truven Health Analytics to find out how your hospital balanced scorecard looks.

For more of my thoughts on healthcare and leadership Subscribe here.

Please help me out by taking a short survey:  Survey Page

And feel free to email me directly at john.jurica.md@gmail.com with any questions about addressing Top 100 Hospital performance measures.

See you in the next post!