Posted on 9. December 2012 · Posted in discharge letter, research

Hospital Discharge-Letters – Room for Improvement?

Results of a recent study in which 200 hospital disc

harge letters were analysed using the CLAS app (Cork Letter Writing Assessment Scale) found significant deficits in many

important areas.

The letters were collected from four family practices (from a variety of secondary and tertiary care hospitals). The letters were anonymised and analysed using the CLAS dataset and scoring system, and were given a total score out of 50. A global rating was also given for each letter. Data was entered into an Excel spreadsheet and  statistical analysis carried out using SPSS.


Diagnostic test results were missing from 52% of letters, investigation results that were pending were missing from 93%, follow up plan missing in 26%, bleep number of doctor writing the letter missing in 78.5% of letters. While 86% of letters contained a list of diagnoses, only1.5% of letters highlighted a new diagnosis.

The consultant’s speciality was missing in 43% of letters, and only 2 letters (1%) contained a problem list at the beginning of the letter.

While 67.5% of letters included a list of discharge medications, only 4.4% alerted that a medication had been stopped. Doses were not written correctly in formal units in 41% of letters. Only 3.5% of letters highlighted medications that had been commenced in hospital. While 73% of letters contained the doctor’s signature, the signature was not legible in 40% of these.

Other findings included:

  • 24% did not contain the ward name
  • 73%  did not have details of presenting complaint
  • 65% did not have details of the patient’s past history
  • 87.5% did not include the patient’s clinical findings on examination
  • 52% did not mention abnormal results
  • Only 22% of letters contained the title of the letter-writer

The mean score of computer generated letters was 30.37 (p=.001) vs. regular handwritten letters 26.51 (p=.003).

Results of this study carried out at University College Cork, Ireland would suggest that the standard of hospital discharge letters is suboptimal. The discharge letter is a very important area in the transfer of patient care from hospital to community. Good Handover is reliant on receiving a timely discharge letter th

at contains all the key information a doctor needs to know when taking over patient care.

Workshops on writing a discharge letter have already been initiated at University College Cork and students receive instruction on the CLAS scale.

The study shows the need to improve the quality of hospital di

scharge letters at Handover. The use of checklists and Handover mobile applications has much to offer in this area.

Clearly, the development and adaptation of electronic discharge letter datasets would be an important step in improving the quality of hospital discharge letters with the hope of decreasing medical error at Handover.


About the Author

Bridget Maher