Abstract
Background and objectives:
Chronic pain impairs health-related quality of life (HRQOL). The 10-item Patient-Reported Outcomes Measurement Information System Global Health (PROMIS-10) assesses HRQOL via physical health (PH) and mental health (MH) domains. The minimal important change (MIC), the smallest within-person change perceived as important, supports interpretation of patient-reported outcome measures (PROMs). MIC thresholds for PROMIS-10 in chronic pain rehabilitation are unknown. We evaluated the responsiveness of PROMIS-10, estimated MIC thresholds for improvement and deterioration using contemporary anchor-based methods, and assessed their credibility.
Methods:
We analyzed pretreatment and post-treatment data from nine interdisciplinary pain centers in Denmark (September 2018-September 2024). Complete PH data were available for 7315 patients, and MH data for 7279, each linked to the Patient Global Impression of Change (PGIC) anchor. Responsiveness was tested against seven prespecified hypotheses using standardized response means (SRMs) and construct validity via Pearson's correlations. MIC thresholds were estimated using three anchor-based methods: mean change (MICmean), improved adjusted predictive modeling (MICiapm), and longitudinal confirmatory factor analysis (MIClcfa). We assessed anchor reliability and present state bias (PSB) and evaluated MIC credibility with the credibility instrument.
Results:
The PROMIS-10 demonstrated good responsiveness, with SRMs of 0.75 (PH) and 0.65 (MH), supporting all hypotheses. Improvement MIC thresholds for domain raw scores were: PH 0.91 (MICmean), 0.78 (MICiapm), and 0.76 (MIClcfa); MH 1.04 (MICmean), 0.67 (MICiapm), and 0.66 (MIClcfa). Deterioration thresholds were more variable: PH -0.07 (MICmean), -1.27 (MICiapm), and -1.97 (MIClcfa); MH -0.07 (MICmean), -3.41 (MICiapm), and -4.17 (MIClcfa). Anchor reliability was higher for PH than MH. PSB was acceptable for improvement (<0.50) but modestly exceeded the threshold for deterioration (>0.55). All MICs met four of five credibility criteria. The PGIC had a limited ability to differentiate levels of deterioration, thereby reducing the credibility of these thresholds.
Conclusion:
PROMIS-10 is responsive in interdisciplinary pain rehabilitation. A raw score increase of ∼1 point in PH or MH likely indicates an important improvement in patients with chronic pain. Declines of ∼1 to 2 points in PH and ∼3 to 4 points in MH may reflect important deterioration but should be interpreted cautiously due to anchor limitations and lower reliability.
Plain language summary:
Chronic pain greatly impacts physical and mental well-being. To track progress, clinicians and researchers often use short questionnaires. One such tool, the PROMIS-10, measures physical abilities, symptoms, and mental health. However, it has been unclear how much change in PROMIS-10 scores truly matters to patients with chronic pain. In this study, we analyzed data from over 7200 Danish patients with long-term chronic pain who completed PROMIS-10 before and after a rehabilitation program at an interdisciplinary pain center. Patients also rated their overall change on a seven-point scale ranging from "much better" to "much worse." By linking these self-ratings with PROMIS-10 scores, we identified the MIC, the smallest score change that most patients find meaningful.
Keywords:
Anchor-based estimation; Chronic pain; Minimal important change (MIC); PROMIS-10 Global Health; Patient-reported outcome measures (PROMs); Present state bias (PSB); Responsiveness.
Læs den videnskabelige artikel