A few weeks ago i put a question (see below) to the list
and was pleased recieve so many constructive replies. For a
detail list of the replies please feel free to e-mail me,
but for now i've just listed the references given to me. I
would like to thank everyone below, in particular Chris
Theobald who was kind enough to forward responses to a
similar question put to the list some time ago.
AFS CO., USA
The Guy's, King's and St Thomas' School of Medicine, UK
University of Edinburgh, UK
Warner Consulting, USA
Paul T Seed
Guy's, King's and St. Thomas' School of Medicine, UK
Blaise F Egan
BT Data Mining Consultancy, UK
St. George's Hospital Medical School, UK
Dr John Whittington
Mediscience Services, UK
Capital Bank, UK
Three devices were used to record measurements over 'x'
time points on 'y' number of patients. The first is blood
pressure as measured by device A, the second and third are
blood flow as measured by devices B and C. Device C has
only an arbitary scale so that a reading of 100 on patient
A may not have the same 'meaning' as 100 on patient B.
The response data for each device is in the form of the
% change from a base reading taken at time t=0.
Their question is:
Do devices 2 & 3 measure the same information? (I'm
sorry if this sounds a wee bit wooly but that's what's
they've written. Perhaps what they mean are the results
produced by the two devices 'comparable' and if so how do
you go about trying to prove it?)
Altman & Bland in "Statistician" 1983 vol 32 pp307-317
Armstrong, White & Saracci. Principles of exposure
measurement in epidemiology (chapter 4). Oxford Medical
Bland JM, Altman DG. Statistical methods for assessing
agreement between two methods of clinical measurement.
Bland JM, Altman DG. Comparing methods of measurement: why
plotting difference against standard method is misleading.
Lancet 1995; 346:1085-7
Bland, J. Martin and Altman, Douglas G. Statistical
methods for assessing agreement between two methods of
clinical measurement. Lancet, 1986;i: 307-310.
"Correlation, regression and repeated data", Bland and
Altman 1994, BMJ 308: 896
M.J. Cardone (1983). Detection and determination of error
in analytical methodology. Part I & II. J. Assoc. Off.
Anal. Chem., 66: 1257-1282 (see section 3.5: Validation of
alternative assay methods) & 1283-1294.
Dunn, G. (1992). Design and analysis of reliability
studies. Statistical Methods in Medical Research Volume 1,
No 2, pp 123-158.
D.J. Finney (1978) Statistical Method in Biological Assay,
3rd ed., Charles Griffin & co., London.
Fleiss JL. The design and analysis of clinical experiments
(chapter 1). Wiley, 1986.
WA Fuller (1987).Measurement Error Models (Wiley, 1987,
F.E. Grubbs (1973). Errors of measurement, precision,
accuracy and the statisticalcomparison of measuring
instruments. Technometrics, 15: 53-66.
RT St Laurent (1998). Evaluating agreement with a gold
standard in method comparison studies. Biometrics 54,
Lawrence I-Kuei Lin (1989) A Concordance Correlation
Coefficient to evaluate reproducibility. Biometrics, 45,
Longford, 1995, Models for Uncertainty in Educational
Testing, Springer-Verlag, Chapter 6).
"Analysis of serial measurements in medical research",
Matthews, Altman et al 1990, BMJ 300:230-235
H. Passing & W. Bablok (1983/84). A new biometrical
procedure for testing the equality of measurements from two
different analytical methods. Part I & II. J. Clin. Chem.
Clin. Biochem., 21: 709-720 & 22: 431-445.
Pitman-Morgan approach to checking whether one method has a
much higher measurement error variance than the other
(Shukla, Biometrics 29, 1973, 373-377.
'Measurement in Laboratory Medicine' (1995).
Butterworth-Heinemann, Oxford. ISBN 0 7506 2259 8
(paperback, about 20 pounds)
Sir Humphry Davy Department of Anaesthesia
Bristol Royal Infirmary
Bristol BS2 8HW
Tel. 0117 9283169
Fax. 0117 9268674
e-mail:- [log in to unmask]