Length of Stay Trending Up: What Next? – Part I


by Dr. Arun Mohan MD, MBA

Do the math and it’s easy to understand why hospital medicine grew so quickly.  Multiply the 14-17% reduction in length of stay achieved by hospitalists by the number of admissions a hospitalist does annually and the value proposition becomes clear.  However hospitals continue to look for opportunities to improve efficiency  as cost pressures change, and they often turn to hospitalists to help them figure it out.  So what can hospitals do?

As a matter of background, length of stay is important to hospitals for a combination of clinical and, of course, financial reasons.  Clinically, there is a large body of evidence which demonstrates an association between adverse events and length of stay.  It’s hard to know whether complications lead to long lengths of stay or vice versa, but we all know that bad things can happen in the hospital and that more is not always better.

Financially, hospitals are generally paid a fixed payment per admission, the amount of which is determined by the diagnosis related group (DRG) into which the patient is classified.  By reducing a day in the hospital, the hospital not only saves on variable costs, such as labor and supplies, but also is able to accept a new admission and associated payment.

At this point, a common concern among physicians is that reducing length of stay increases the risk of readmission and worsens outcomes.  Fortunately, there’s very little evidence that’s the case.  Most recently, fellow SHM member Peter Kaboli looked  at nearly 14 years of data from the VA and showed that both average length of stay and the number of readmissions fell.  And while there was a very modest increase in readmissions for both hospitals with lower and higher lengths of stay, it wasn’t clinically relevant.

And that makes sense.

On one end of the spectrum you can imagine discharging a patient as soon as you see her, essentially doing nothing, and the person being readmitted.  On the other end of the spectrum, you can imagine a patient who is really sick, has a long length of stay, and where it doesn’t matter what you do – the person will be readmitted.  Let’s also keep in mind that most of the “extra” time that a patient stays in the hospital isn’t necessarily spent on education or care coordination.

Although average length of stay continues to decline nationally, there is substantial geographic variation.  After adjusting for race and sex, Medicare patients admitted with acute myocardial infarction spend nearly 1.4 days more in the hospital in low-performing areas of the country.  Patients with heart failure and pneumonia spend nearly 1.2 more days.  And the list doesn’t end there.

So aside from patient factors, what explains this variation?

Unfortunately, the short answer is that we don’t really know.  Some of it is due to the physician, some to the hospital, and some to the community.  But there are some lessons from the literature that may be instructive and that can help us to develop interventions that work.

In Part II, I’ll lay out five key principles that hospitalists can use to get length of stay back on target.


Dr. Arun MohanArun Mohan, MD, MBA is Chief Medical Officer for Hospital Medicine with ApolloMD, a physician-owned, multi-specialty practice that partners with more than 85 hospitals nationally.  He was previously Medical Director of Care Coordination at Emory University Hospital and Associate Vice-Chair for IT in the Department of Medicine at Emory University School of Medicine.

Dr. Mohan will be presenting at Hospital Medicine 2014 as part of the new “Bending the Cost Curve” track on Wednesday, March 26th at 4:15 p.m.  He is co-presenting with SHM’s President Dr. Eric Howell MD, SFHM.  Their talk is entitled, “Interventions to Reduce Length of Stay in Hospitalized Patients.”

1 Comment

  1. […] In my last post, I described substantial geographic variation in length of stay. Although it’s not exactly clear what explains the differences, here are 5 key principles that hospitalists should keep in mind: […]

Leave a Comment