LETTER TO THE EDITOR
Reliable Change Formula Query: Temkin et al. reply
NANCY R. TEMKIN,
1,2
ROBERT K. HEATON,
3
IGOR GRANT,
3,4
and SUREYYA S. DIKMEN
1,5,6
1
Department of Neurological Surgery, University of Washington, Seattle, Washington
2
Department of Biostatistics, University of Washington, Seattle, Washington
3
Department of Psychiatry, University of California, San Diego, La Jolla, California
4
VA San Diego Healthcare System, San Diego, California
5
Department of Rehabilitation Medicine, University of Washington, Seattle, Washington
6
Department of Psychiatry and Behavioral Sciences, University of Washington, Seattle, Washington
Hinton-Bayre (2000) raises a point that may occur to many
readers who are familiar with the Reliable Change Index
(RCI). In our previous paper comparing four models for
detecting significant change in neuropsychological perfor-
mance (Temkin et al., 1999), we used a formula for calcu-
lating S
diff
, the measure of variability for the test–retest
difference, that differs from the one Hinton-Bayre has seen
employed in other studies of the RCI. In fact, there are two
ways of calculating S
diff
—a direct method and an approxi-
mate method. As stated by Jacobson and Truax (1991, p. 14),
the direct method is to compute “the standard error of
the difference between the two test scores” or equivalently
% ~ s
1
2
1 s
2
2
2 2 s
1
s
2
r
xx
' ! where s
i
is the standard deviation at
time i and r
xx
' is the test–retest correlation or reliability co-
efficient. Jacobson and Truax also provide a formula for the
approximation of S
diff
when one does not have access to
retest data on the population of interest, but does have a
test–retest reliability coefficient and an estimate of the cross-
sectional standard deviation, i.e., the standard deviation at a
single point in time. This approximation assumes that the
standard deviations at Time 1 and Time 2 are equal, which
may be close to true in many cases. Since we had the lon-
gitudinal data to directly calculate the standard error of the
difference between scores at Time 1 and Time 2, we used
the direct method. Which method is preferable? When the
needed data are available, it is the one we used. It computes
the variability of the difference based on the actual test–
retest differences and avoids making the assumption that
the variability at Times 1 and 2 are the same. Finally, it should
be noted that, in our study, the results obtained by the two
methods are quite similar and the differences between them
are of questionable importance (Hinton-Bayre, 2000,
Table 1).
REFERENCES
Hinton-Bayre, A. (2000). Reliable Change formula query. Journal
of the International Neuropsychological Society, 6, 362–363.
Jacobson, N.S. & Truax, P. (1991). Clinical significance: A statis-
tical approach to defining meaningful change in psychother-
apy research. Journal of Consulting and Clinical Psychology,
59, 12–19.
Temkin, N.R., Heaton, R.K., Grant, I., & Dikmen, S.S. (1999).
Detecting significant change in neuropsychological test perfor-
mance: A comparison of four models. Journal of the Inter-
national Neuropsychological Society, 5, 357–369.
Reprint requests to: Nancy R. Temkin, Department of Neurological Sur-
gery (Box 359924), University of Washington, Seattle, WA 98104-2499.
E-mail: temkin@biostat.washington.edu
Journal of the International Neuropsychological Society (2000), 6, 364.
Copyright © 2000 INS. Published by Cambridge University Press. Printed in the USA.
364
https://www.cambridge.org/core/product/91FF19178BAE79C706ECA0816212AC57
Downloaded from https://www.cambridge.org/core. IP address: 54.70.40.11, on 26 Apr 2018 at 17:08:26, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms.