Answers are grounded in this paper and cite their sources.
In Brief Study Design. Assessment of correlation of measures of low back pain (LBP) using data pooled from 2 identical studies. Objective. To assess the relative responsiveness of and correlation between various measures of LBP, including the Roland-Morris Disability Questionnaire (RMDQ), the LBP intensity (LBPI) visual analog scale (VAS), and patient’s global assessment of response to therapy (PGART). Summary of Background Data. Several tools are available to measure pain, functional limitation, and response to therapy for LBP. Studies have shown varying degrees of responsiveness and correlation. Methods. This was a pooled subgroup analysis of patients with chronic LBP from 2 identical studies comparing etoricoxib 60 mg, 90 mg, and placebo. LBP was assessed by the time-weighted average change from baseline over 12 weeks as measured by RMDQ, LBPI VAS, and PGART. Correlation was calculated using Pearson’s correlation coefficient. Results. The correlation (r) between LBPI and RMDQ changes ranged from 0.657 and 0.703; correlations between LBPI and PGART changes ranged from 0.677 and 0.738. Cutpoints separating responders from nonresponders for all 3 measures fell near the 66.7th percentile of response and were consistent with minimal clinically significant changes identified in the literature. Conclusion. In this study, the RMDQ, LBPI VAS, and PGART showed a high degree of correlation in measuring response to therapy in LBP, suggesting clinicians may be able to simplify assessments. In this analysis of pooled data from 2 previous low back pain efficacy trials, we found that 3 common tools—the Roland-Morris Disability Questionnaire, the low back pain inventory visual analog scale, and patient global assessment of response to therapy—show a high degree of correlation with similar responsiveness.
No takes yet. Share an insight, caveat, or question.
Sheldon et al. (2008) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: