University of Surrey

Test tubes in the lab Research in the ATI Dance Research

End-digit preference in blood pressure recordings of patients with ischaemic heart disease in primary care.

de Lusignan, S, Belsey, J, Hague, N and Dzregah, B (2004) End-digit preference in blood pressure recordings of patients with ischaemic heart disease in primary care. J Hum Hypertens, 18 (4). pp. 261-265.

[img] Text
Restricted to Repository staff only
Available under License : See the attached licence file.

Download (124kB)
[img] Text (licence)
Restricted to Repository staff only

Download (33kB)


End-digit preference describes the disproportionate selection of specific end digits. The rounding of figures might lead to either an under- or over-recording of blood pressure (BP) and a lack of accuracy and reliability in treatment decisions. A total of 85 000 BP values taken from computerised general practice records of ischaemic heart disease patients in England between 2001 and 2003 were examined. Zero preference accounts for 64% of systolic and 59% of diastolic readings, compared with an expected frequency of 10% (P<0.000001). Even numbers are more frequently seen than odd numbers. In all, 64% of nonzero systolic recordings and 65% of diastolic recordings ended in even numbers, compared with expected proportions of 44% (P<0.0001). Among the nonzero even numbers, eight is the most frequently observed: 28% of systolic and 31% of diastolic recordings compared with an expected proportion of 25% (P<0.0001). Among the five nonzero odd numbers, five is the most frequently observed end digit, representing 59% systolic and 62% of diastolic compared with an expected level of 20% (P<0.00001). English general practice displays marked end-digit preference. This is strongly for the end-digit zero. However, there is more use of other end-digits, notably 8 and 5. This bias potentially carries important treatment consequences for this high-risk population.

Item Type: Article
Authors :
de Lusignan, S
Belsey, J
Hague, N
Dzregah, B
Date : April 2004
DOI : 10.1038/sj.jhh.1001663
Uncontrolled Keywords : Blood Pressure, Blood Pressure Determination, Diastole, England, Humans, Myocardial Ischemia, Observer Variation, Primary Health Care, Reproducibility of Results, Signal Processing, Computer-Assisted, Systole
Depositing User : Symplectic Elements
Date Deposited : 28 Mar 2017 15:04
Last Modified : 31 Oct 2017 14:23

Actions (login required)

View Item View Item


Downloads per month over past year

Information about this web site

© The University of Surrey, Guildford, Surrey, GU2 7XH, United Kingdom.
+44 (0)1483 300800