Interpretation of electrocardiograms by doctors

BMJ 1994; 309 doi: (Published 10 December 1994) Cite this as: BMJ 1994;309:1551
  1. Hugh Montgomery, research fellowa,
  2. Steven Hunter, senior registrarb,
  3. Susie Morris, registrara,
  4. R Naunton-Morgan, registrarc,
  5. R M Marshall, registrard
  1. a Hatter Institute for Cardiovascular Studies, Division of Cardiology, University College London Hospitals NHS Trust, London WC1E 6DB
  2. b Department of Cardiothoracic Surgery, Royal Postgraduate Medical School, Hammersmith Hospital, London W12 0HS
  3. c Department of Vascular Surgery, St George's Hospital, London SW17 0RE
  4. d Oldchurch Hospital, Romford, Essex RM7 0BE
  1. Correspondence to: Dr Montgomery.
  • Accepted 9 September 1994

After marking examination papers of medical students, one of us (HM) found that only one in 50 students correctly identified the PR and QT intervals of the electrocardiogram. As a cardiothoracic surgeon was also unable correctly to identify these intervals we investigated whether such knowledge was lacking in doctors in general.

Subjects, methods, and results

An illustration of a standard electrocardiogram (8 cm × 3 cm) was presented on a single sheet of A4 paper. One hundred and fifty eight doctors of diverse specialty and seniority (table) were asked to mark the PR and QT intervals with arrows. These intervals are defined in a standard way internationally and are discussed in various text books.*RF 1-5* If a candidate was able successfully to mark the PR interval and to state what time interval the smallest square on an electrocardiogram represents when recording at standard speed (25 mm/s), he or she was deemed capable of defining PR duration. The answers were unprompted and anonymous; respondents were supervised to ensure neither consultation with colleagues nor reference to written texts. They were asked not to mention the existence of the survey to colleagues. All answers were collected in a single morning.

Specialty and grade of the 158 doctors completing questionnaire on PR and QT interval of electrocardiogram

View this table:

Overall, 117 of the 158 doctors completing the questionnaire (74%) did not have sufficient knowledge to measure a PR interval, with 107 (68%) unable to define the PR interval and 65 (41%) unaware of the time interval represented by a small square on an electrocardiogram at standard recording speed. The ability to measure PR interval was poorest among house officers (19 out of 20 failed) and best among consultants (although 17 out of 29 (59%) still failed). Seniority did not, however, consistently correlate with a correct response. Thus 18 (78%) medical senior house officers, six (50%) medical consultants, 12 (80%) medical registrars, and all 10 medical senior registrars could not assess a PR interval. Furthermore, this lack of knowledge was not confined to any one specialty: five out of 23 cardiologists (22%), 21 out of 28 anaesthetists (75%), eight out of 10 doctors in accident and emergency medicine (80%), and all 10 general surgeons failed correctly to identify the PR interval. Worryingly, this lack of knowledge applied to all of the cardiothoracic surgeons, to doctors practising in accident and emergency medicine (eight of the 10, including both consultants), to practising physicians (as above), and even to cardiology registrars (four out of 13 (31%)). One of the five consultant cardiologists defined the PR interval incorrectly.

The duration of the QT interval is also an important basic measure, particularly in assessing patients with a predisposition to arrhythmia, including those who have taken overdoses of drugs such as tricyclic antidepressants. It was thus alarming to find that 120 (76%) of those questioned defined the QT interval incorrectly, including all 10 doctors in accident and emergency posts. Also faring badly were cardiologists (one of the three senior house officers, 10 of the 13 registrars, one of the two senior registrars, and one of the five consultants were incorrect), cardiothoracic surgeons (three of the four registrars and all three consultants were wrong), and physicians (15 of the 23 senior house officers (65%), 10 of the 13 registrars (77%), seven of the 10 senior registrars (70%), and 10 of the 12 consultants (83%) were wrong.

Overall, if defining the PR and QT intervals and the duration of one small square on the electrocardiogram were together needed to pass an examination 90% (142/158) of those questioned would have failed.


Elementary knowledge of the electrocardiogram was badly lacking in this study. Indeed, two of the 10 medical senior registrars and a consultant cardiothoracic surgeon labelled the S wave a Q wave, and 17% of our sample measured the PR interval from the middle of the P wave and 11% from its end. Bundle branch block must also confuse the 47% who measured the PR interval to the peak of the R wave.

Being able to define a PR interval correctly may make little difference to medical practice, but a lengthening PR interval may herald serious disease. How can recorded interpretations of electrocardiograms be compared if individual doctors use their own criteria to define its basic features?

Questionnaires were not necessarily completed by staff at the hospitals where we work.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract