Article Text

Download PDFPDF

Randomised trial comparing prototype structures for clinical letters
  1. G Wynn1,
  2. D Hindley2
  1. 1Stepping Hill Hospital, Stockport, UK
  2. 2Fairfield General Hospital, UK;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

A seven and a half minute consultation leaves a GP little time to extract the salient points from waffling clinical correspondence. Consultants’ letters to GPs often contain minimal structure, yet GPs prefer letters containing lists of key problems and management issues.1–3 We examined whether highly structured letters, using headings and bullet points, were preferred over problem/management list letters or unstructured prose.

A total of 210 GPs were each randomly allocated two of four prototype letters and asked to rate each letter on readability, structure, content, and overall feel. The GPs, and 76 consultants, were asked to rate the most important roles of letters from consultants to GPs.

Highly structured letters were significantly preferred in all aspects over letters with problem/management lists (p = 0.001–0.05) and these were preferred over prose letters (p = 0.05). There was no significant preference expressed about whether structured letters could contain a short prose summary. Consultants and GPs agreed that providing specialist advice and a management plan were vital aspects of consultants’ letters. Consultants also felt the letter was vital to document information given to the patient and to form part of the hospital records. This dichotomy may explain why consultants’ letters often do not meet GPs’ expectations.

Clear concise communication is central to patient care. Structured letters are easier to extract information from, quicker to read,2 and much preferred by GPs. Our low response rates (42% of GPs and 41% of consultants) leave open the possibility of response bias. However, our results showed high levels of significance and were in keeping with previous studies.1,2

Currently, discharge letters may be so poor that they can hinder continuity of care.4 Letters can be improved with training and prompt sheets,5 but these are expensive options. Fully structured letters provide an easy way to improve communication and reduce the likelihood of serious omission.