Caveats concerning comparisons of change rates obtained with five methods of identifying significant client changes: comment on Speer and Greenbaum (1995).

Various methods have been proposed to identify clients who have improved, have not changed, or have deteriorated in psychotherapy. D. D. Speer and P. E. Greenbaum (1995) compared empirically determined change rates (proportions of clients identified as improved, unchanged, or deteriorated) obtained with 5 of these methods, tested the hypothesis that 1 method would be more sensitive than the other 4 in detecting significant changes, and recommended 2 of the 5 methods. This comment draws attention to problems related to the meaning of their comparisons of change rates and to some implications of their recommendations. These problems include misinterpretation of formulas; use of methods with questionable standard errors; confounding of effects of methods with effects of norms, data sets, and risks of misclassifying patients who do not change; absence of valid external criteria of client changes; and excessive reliance on the Jacobson-Truax method for evaluation of 2-wave data.