Edinburgh Napier University

  STUDY DESIGN:
A prospective cohort.
OBJECTIVE:
The aim of this study was to compare the responsiveness of a patient specific outcome measure with the Oswestry Disability Index v2.1 (ODI) and the Roland and Morris Disability Questionnaire (RMDQ) for patients with mild to moderate subacute and chronic low back pain.
SUMMARY OF BACKGROUND DATA:
Patient specific outcome measures allow individuals to select and rate the relative importance of activities they consider of greatest relevance. However, there is a paucity of research reporting the responsiveness of these measures for patient with back pain.
METHODS:
Two hundred and one patients were assessed at baseline and 12 months using the patient specific activity questionnaire (PSAQ), the ODI, the RMDQ, and a transition rating scale. The latter was used to categorize patients into 3 groups (better, same or worse). Effect size statistics, receiver operating characteristic curves and 1-way between groups analysis of variance were used in the analysis.
RESULTS:
All instruments were able to detect improvements in back pain as measured by the transition rating scale. In this improved group effect sizes were large for the PSAQ (1.08 to 1.31) and ODI (-0.88 to -1.00) but only moderate for the RMDQ (-0.70 to -0.74). In the deteriorated group effect sizes were moderate to large for the ODI (0.61 to 1.16) and RMDQ (0.69 to 1.25), but small for the PSAQ (-0.16 to -0.26). The areas under receiver operator characteristic curves for the ODI, PSAQ, and RMDQ, respectively were: 0.75 (95% CI 0.68 to 0.82), 0.75 (95% CI 0.68 to 0.82) and 0.69 (95% CI 0.61 to 0.76).
CONCLUSION:
The PSAQ was highly responsive to change in patients who report improvement in back symptoms but unlike the ODI and RMDQ was unable to detect deterioration. Overall, the ODI was more responsive than either the PSAQ or the RMDQ.