January 15th, 2015

Low-Quality Discharge Summaries May Have Consequences

The CardioExchange Editors interview Leora Horwitz about her research group’s study of the association between the quality of discharge summaries and risk for readmission for heart failure. The article is published in Circulation: Cardiovascular Quality and Outcomes.

CardioExchange Editors: Please summarize your findings for our readers.

Horwitz: We collected more than 1500 discharge summaries, from 46 hospitals around the nation, as part of a large randomized controlled trial (Telemonitoring to Improve Heart Failure Outcomes). All summaries were for patients who had been hospitalized with heart failure and who survived to discharge. We found that not a single patient’s summary met all three criteria of being prepared in a timely manner, being transmitted to the right physician, and being fully comprehensive in content. We also found that hospitals varied very widely in their average quality. For instance, at some hospitals, 98% of summaries were completed on the day of discharge; at others, none were. In the accompanying data report, we show that summaries that are transmitted to outside clinicians and that include more key content elements are associated with lower risk for rehospitalization within 30 days after discharge. This is the first study to show an association between discharge-summary quality and risk for readmission.

Editors: What are the obstacles to creating timely discharge summaries with good content?

Horwitz: As a practicing clinician, I am acutely aware of the challenges in creating high-quality discharge summaries. With ever-declining lengths of stay, patients are coming and going with extreme speed, leaving little time for clinicians to prepare paperwork. There are always competing time demands. New patients absolutely should have a full workup, history, and physical completed within 24 hours, and existing patients require a note every day. By contrast, few hospitals in my experience require that the discharge summary be completed on the day of discharge. When faced with competing requirements (one patient has a hard deadline and the other doesn’t), it is easy to focus on the one with the deadline.

To date, electronic medical record systems have offered little help. Most of the discharge summaries in our dataset were still being dictated or even being handwritten, thereby requiring clinicians to create all the content on their own. Out-of-the box discharge-summary templates on most electronic health record systems are little but computerized shell forms that leave nearly all the work to the clinician.

For instance, Yale-New Haven Hospital was only recently able to fundamentally redesign its discharge-summary template to automatically pull in laboratory studies that were pending at the time of discharge, as well as the discharge vital signs — crucial to safe transitions of care but typically omitted from summaries. The summary is now also automatically routed to the primary care physician, which eliminates another manual step commonly omitted by busy inpatient physicians.

Editors: How much effect do you think poor discharge summaries have on readmission and clinical outcomes? Do they matter? 

Horwitz: Previous studies had not been able to identify an association between discharge-summary quality and risk for readmission or other adverse outcomes. Our accompanying data report, however, shows that elements of discharge-summary quality are indeed associated with readmission risk, suggesting that better-informed outpatient clinicians may be better equipped than less-informed clinicians to care for patients after discharge.

It is possible that the hospitals whose policies encourage prompt completion of discharge summaries and transmission to outside clinicians also have high-quality medication reconciliation processes, routine post-discharge follow-up calls, and so on. That is, discharge-summary quality may be important in its own right and/or it may be a marker of high-quality transitional-care practices. Without a randomized controlled trial (which would be unethical), we cannot know definitively. Nevertheless, our results — combined with the obvious importance of physician communication — lend strong support to efforts to improve discharge-summary quality. Many institutional policies and information-technology changes could make material differences in discharge-summary quality. Busy clinicians need not be asked to solve all of this on their own.


How timely and complete are discharge summaries at your institution? Will Dr. Horwitz’s study influence the importance you ascribe to discharge summaries? 

One Response to “Low-Quality Discharge Summaries May Have Consequences”

  1. Enrique Guadiana, MD says:

    I think is very difficult to show an association between discharge-summary quality and risk for readmission. The study is not randomized, it is focused on omissions in content and was not designed to assess content accuracy. The discharge summaries were less likely to include discharge condition (30.7%), discharge volume status (16.0%), or discharge weight (15.7%) so you couldn’t take in consideration this variables and in consequence the severity and general condition of the HF patient at discharge was a mystery. This variables could easily explain the difference in the risk for readmission. The study is a good evaluation for discharge summary quality in HF patients in the USA.