Intended for healthcare professionals

Clinical Review ABC of hypertension

Blood pressure measurement

BMJ 2001; 322 doi: (Published 28 April 2001) Cite this as: BMJ 2001;322:1043

Part II—Conventional sphygmomanometry: technique of auscultatory blood pressure measurement

  1. Gareth Beevers,
  2. Gregory Y H Lip,
  3. Eoin O'Brien

    The measurement of blood pressure in clinical practice by the century-old technique of Riva-Rocci/Korotkoff is dependent on the accurate transmission and interpretation of a signal (Korotkoff sound or pulse wave) from a subject via a device (the sphygmomanometer) to an observer. Errors in measurement can occur at each of these interactionary points of the technique, but by far the most fallible component is the observer.

    Rose classification of observer error

    • Systematic error

    • Terminal digit preference

    • Observer prejudice

    Observer error

    In 1964, Geoffrey Rose and his colleagues classified observer error into three categories.1

    Observer training techniques

    • Direct instruction by an experienced observer

    • Instruction manuals and booklets

    • Audiotapes

    • Video films

    • CD Rom presentations

    Systematic error

    This leads to both intraobserver and interobserver error. It may be caused by lack of concentration, poor hearing, confusion of auditory and visual cues, etc. The most important factor is failure to interpret the Korotkoff sounds accurately, especially for diastolic pressure.

    Terminal digit preference

    This refers to the phenomenon whereby the observer rounds off the pressure reading to a digit of his or her choosing, most often to zero. Doctors may have a 12-fold bias in favour of the terminal digit zero; this has grave implications for decisions on diagnosis and treatment, although its greatest effect is in epidemiological and research studies in which it can distort the frequency distribution curve and reduce the power of statistical tests.2

    Observer prejudice or bias

    This is the practice whereby the observer simply adjusts the pressure to meet his or her preconceived notion of what the pressure should be. It usually occurs when there has been recording of an excess of pressures below the cut-off point for hypertension and it reflects the observer's reluctance to diagnose hypertension. This is most likely to occur when an arbitrary division is applied between normal and high blood pressure, for example 140/90 mm Hg. An observer might tend to record a …

    View Full Text

    Log in

    Log in through your institution


    * For online subscription