Convergent validity and responsiveness of the 9-item PROMIS-29 derived low back pain impact score

Core domains recommended for assessing low back pain (LBP) include pain, function, and health-related quality-of-life. The 9-item questionnaire, derived from PROMIS-29 Impact score, was proposed by the National Institute of Health task force as a comprehensive yet concise measure of LBP impact. However, its psychometric properties have not been widely evaluated. The objectives were to evaluate convergent validity, responsiveness and minimal clinically important difference of the Impact score (total and domains) compared with established measures: Oswestry Disability Index, Numerical Rating Scale, EuroQol 5-Dimension 5-Level, the Patient-Specific Functional Scale, and the Global Perceived Effect Scale (GPE). A secondary analysis was conducted using data of WELBack RCT of 372 participants with chronic non-specific LBP. Outcomes were collected at baseline and 2-months follow-up. Convergent validity was assessed via correlations between the Impact score and legacy measures and one-tailed hypothesis testing. Responsiveness was analyzed using distribution-, construct-, and criterion-based approaches. The MCID estimate was derived using the GPE as an anchor. The questionnaire demonstrated high convergent validity, meeting 80% of predefined hypotheses. Our analyses demonstrated moderate responsiveness, with a large effect size (0.8), 60% of hypotheses were met in construct-based approach, and an acceptable area under the curve (0.73) in criterion-based evaluation. MCID was estimated at 6 points. In conclusion, the Impact score, which captures a construct of LBP impact, reflecting multiple health domains, demonstrated high convergent validity and acceptable responsiveness in patients with chronic LBP. Its brevity and psychometric robustness support its use in clinical practice and research. Perspectives: This article evaluated the convergent validity, responsiveness and minimal clinical important difference of the Impact score to evaluate low back pain impact. The results demonstrate good validity, responsiveness and an MCID of 6 which supports the use of this short and easy to apply questionnaire.