Purpose: To examine the extent and type of influence of response shift (RS) (reconceptualization, reprioritization, recalibration) on the assessment of changes in health-related quality of life (HRQoL) after the inpatient rehabilitation of patients with chronic back pain and chronic ischaemic heart disease.
Methods: RS was assessed using the then-test and structural equation modelling (SEM) approaches. HRQoL was recorded in a questionnaire study at the start and end of rehabilitation using disease-specific instruments. Global items were also developed to collect data on HRQoL at the start and end of rehabilitation, as well as for the retrospective evaluation of HRQoL at the beginning of rehabilitation via the then-test.
Results: There are small recalibration effects in the then-test in both diseases in nearly every item which may underestimate the actual effects of change. Using the SEM approach, uniform and non-uniform recalibration effects appear in a few of the scales that cause true change scores to be under- and over-estimated.
Conclusion: RS reflects a phenomenon that should be included when assessing HRQoL to allow a more valid interpretation of treatment effects. The results of SEM approach, however, suggest that RS does not lead to fundamentally different results in rehabilitation effects.