Editor
The paper by Lyons et al [1] is one of many papers (references availble
from me) that have tried to show that doctors have or have not got worse
handwiting than other healthcare professonals.
Lyons et al have attempted to use an objective means of measuring
illegibility in handwriting with the use of computer assisted handwritng
analysis software. Most other preceeding studies have relied exclusively
on human assesment of handwriting, using anything from one to four
assessors, with some even attempting to control for inter-rater
reliability [2].
Though the use of the subjective human may lead to problems in
reliability and reproducability in these studies, there is much to commend
these methods of handwritng analysis.
Firstly, a computer will rarely if ever be expected to analyse
handwriting by any healthcare professional in the real world of the NHS.
Notes are written for humans to read. Statistically significant values of
poorer handwriting by doctors is clinically and pragmatically irrelevant.
The important test is: Can anyone read it ?
Secondly, doctors are urged by the GMC [3] and other bodies (e.g. the
defence organisations and BMA) to kepp legible and
contemporaneous notes. The expectation is that both writer and other other
healthcare professional should be able to read the handwriting, not a
computer. Thus once again man is the expected end user.
Thirdly, medico-legally, a computer will not be used to asses
legibility. Even if it were to be used, can a computer ever tell us if the
handwriting is of a resonable standard or unambiguous in its
interpretation: all expectations that are required of hand written notes
[4] [5]. Man, again, is the discriminating instrument and it is against
another man that we will be judged.
Lastly, we recently conducted an in-practice audit of legibility of
hand written entries; with one fallible human assessor and a non-
standardised, unvalidated, but pragmatic criteria derived from a limited
review of the liteature (details available from me). Here tha nursing
staff had consistently more legible handwriting than the doctors. But not
all doctors had illegible hand writing, in fact a spectrum existed that
varied within and between each doctor. More shades of grey than a simpe
black and white
picture of legibility versus illegibility.
If handwriting is to be assessed efforts must be put into finding a
valid, reliable and pragmatic way of assessing handwriting that is human
based.
Geoffrey Wong
GP Registrar
Richard Halvorsen
GP Principle and Trainer
Holborn Medical Centre
64, Lambs Conduit Street
London
WC1N 3LW
)171 405 3541
Competing interests: NONE
1 Lyons R, Payne C, McCabe M, Fielder C. Legibility of doctors
handwriting: quantitative comparative study. BMJ
1998;317:863-4
2 Winslow EH, Nestor VA, Davidoff SK, Thompson PG, Borum JC.
Legibility and completenss of physicians' hand-written
medication orders. Heart & Lung 1997;26:158-64
3 Good Medical Practice. The General Medical Council. London. July
1998
4 Hoyte P. Can I see the records? Clinical notes-disclosure and
patient access?.London: The Medical Defence Union
Ltd. 1996
5 Palmer RN. Pitfalls of practice. London: The Medical Protection
Society Ltd. 1995
Rapid Response:
Computer analysis of handwriting is flawed
Editor
The paper by Lyons et al [1] is one of many papers (references availble
from me) that have tried to show that doctors have or have not got worse
handwiting than other healthcare professonals.
Lyons et al have attempted to use an objective means of measuring
illegibility in handwriting with the use of computer assisted handwritng
analysis software. Most other preceeding studies have relied exclusively
on human assesment of handwriting, using anything from one to four
assessors, with some even attempting to control for inter-rater
reliability [2].
Though the use of the subjective human may lead to problems in
reliability and reproducability in these studies, there is much to commend
these methods of handwritng analysis.
Firstly, a computer will rarely if ever be expected to analyse
handwriting by any healthcare professional in the real world of the NHS.
Notes are written for humans to read. Statistically significant values of
poorer handwriting by doctors is clinically and pragmatically irrelevant.
The important test is: Can anyone read it ?
Secondly, doctors are urged by the GMC [3] and other bodies (e.g. the
defence organisations and BMA) to kepp legible and
contemporaneous notes. The expectation is that both writer and other other
healthcare professional should be able to read the handwriting, not a
computer. Thus once again man is the expected end user.
Thirdly, medico-legally, a computer will not be used to asses
legibility. Even if it were to be used, can a computer ever tell us if the
handwriting is of a resonable standard or unambiguous in its
interpretation: all expectations that are required of hand written notes
[4] [5]. Man, again, is the discriminating instrument and it is against
another man that we will be judged.
Lastly, we recently conducted an in-practice audit of legibility of
hand written entries; with one fallible human assessor and a non-
standardised, unvalidated, but pragmatic criteria derived from a limited
review of the liteature (details available from me). Here tha nursing
staff had consistently more legible handwriting than the doctors. But not
all doctors had illegible hand writing, in fact a spectrum existed that
varied within and between each doctor. More shades of grey than a simpe
black and white
picture of legibility versus illegibility.
If handwriting is to be assessed efforts must be put into finding a
valid, reliable and pragmatic way of assessing handwriting that is human
based.
Geoffrey Wong
GP Registrar
Richard Halvorsen
GP Principle and Trainer
Holborn Medical Centre
64, Lambs Conduit Street
London
WC1N 3LW
)171 405 3541
Competing interests: NONE
1 Lyons R, Payne C, McCabe M, Fielder C. Legibility of doctors
handwriting: quantitative comparative study. BMJ
1998;317:863-4
2 Winslow EH, Nestor VA, Davidoff SK, Thompson PG, Borum JC.
Legibility and completenss of physicians' hand-written
medication orders. Heart & Lung 1997;26:158-64
3 Good Medical Practice. The General Medical Council. London. July
1998
4 Hoyte P. Can I see the records? Clinical notes-disclosure and
patient access?.London: The Medical Defence Union
Ltd. 1996
5 Palmer RN. Pitfalls of practice. London: The Medical Protection
Society Ltd. 1995
Competing interests: No competing interests