Riddle and Lingard make an important point which supports our sentiments that the implications of response shift in orthopedic clinical research are potentially profound and need to be accounted for. These authors, however, raise concern about the then-test methodology in measuring response shift.

A growing body of literature has examined the impact of response shift in patients. Schwartz et al.(1) in a recent meta analysis of response shift-related articles reported that the majority of reviewed articles (1966-2004) used the then-test design. Furthermore, it was only these studies, which were able to provide requisite data for effect size (ES) computation. Other response shift studies using individualized measures, such as the Patient-Generated Index, Schedule of the Evaluation of Individual QOL (SEIQOL), or qualitative interviews did not provide the data necessary for calculating ES.

The then-test method assumes that respondents will use their post test internal standards when providing a reevaluation rating of their baseline score. We acknowledge that the then-test design has been criticized for its susceptibility to recall bias. This has been evaluated by Visser et. al.(2) in a convergent validity study. They measured response shift by the then-test approach, anchor recalibration and Structural Equation Modeling (SEM). They showed good convergent validity between the then-test approach and SEM. The results of both of these methods were largely comparable. This suggests that the SEM and the then-test approach measure the same concept. These methods use statistically independent operations for response shift. SEM does not use retrospective data and is, therefore, not susceptible to recall bias. Based on the convergence of methods in their study and for their results, this suggests that the then-test was not affected by recall bias.

Generalizability to other then-test studies is not necessarily guaranteed. Visser et al.(2) used a testing interval of 3 months; in our study 6 months was used. Further study needs to be performed, whether the convergence of methods will still show good convergent validity at 6 months. As such, we do agree with Riddle and Lingard that additional study is required to further define the role of response shift and the potential limitations in our ability to accurately measure this.

REFERENCES:

1. Schwartz CE, Bode R, Repucci N, Becker J, Sprangers MA, Fayers PM. The clinical significance of adaptation to changing health: a meta- analysis of response shift. Qual Life Res. 2006 Nov;15(9):1533-50. Review.

2. Visser MR, Oort FJ, Sprangers MA. Methods to detect response shift in quality of life data: a convergent validity study. Qual Life Res. 2005 Apr;14(3):629-39.

To The Editor:

In the paper "Response Shift in Outcome Assessment in Patients Undergoing Total Knee Arthroplasty," Razmjou and colleagues (1) concluded that patients demonstrated a response shift and that this finding supports the need for accounting for response shift in clinical research. The implications of these findings are potentially profound. Unless authors account for response shift in randomized trials, for example, the findings should be questioned.

The authors may indeed be right because response shift appears to be real(2,3). However, we have a concern about the way in which the authors quantified response shift. The authors used the Then-test approach to measure response shift. The Then-test requires the person to complete, in this case, a WOMAC pre-operatively and at 6 months postoperatively the WOMAC is completed twice (reporting current status and recall of their preoperative status). In other words, the Then-test assumes that because more information is available to the patient following surgery, it is more valid than serial change scores for detecting real change. When using the Then-test, one must assume that patients accurately recall their actual status several months prior and then adjust their WOMAC ratings based on the newly adjusted internal standard. However, this assumption has been challenged by others(4). In addition, the implicit theory of change suggests that patients begin with their present state and infer what their initial state must have been. An inherent part of this theory is that recall of a previous state will be directly influenced by the patient's current state and the recall data will be biased and inaccurate(5).

We agree that Razmjou and colleagues examined a potentially important issue but question the use of the Then-test to adjust for response shift for clinical studies of arthroplasty. It incurs considerable burden on the patients and accuracy of recall is questionable especially as the length of time following surgery increases. Given the uncertainty in this area, additional study is essential to correctly direct changes in the way that clinical studies and trials are designed and conducted.

The authors did not receive any outside funding or grants in support of their research for or preparation of this work. Neither they nor a member of their immediate families received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, division, center, clinical practice, or other charitable or nonprofit organization with which the authors, or a member of their immediate families, are affiliated or associated.

References:

1. Razmjou H, Yee A, Ford M, Finkelstein JA. Response shift in outcome assessment in patients undergoing total knee arthroplasty. J Bone Joint Surg Am 2006;88:2590-2595.

2. Linton SJ, Melin L. The accuracy of remembering chronic pain. Pain 1982;13: 281-285

3. Postulart D, Adang EMM. Response shift and adaptation in chronically ill patients. Med Decis Making 2000;20:186-193.

4. Allison PJ, Locker D. Feine JS. Quality of life: A dynamic construct. Soc Sci Med 1997;45:221-230.

5. Ross M. Relation of implicit theories to the construction of personal histories. Psychol Rev 1989;96:341-347.