Session Information
Session Type: Abstract Submissions (ACR)
Background/Purpose: The Fibromyalgia Impact Questionnaire (FIQ) is a multidimensional instrument that encompasses many of the core domains recommended by OMERACT for evaluation in fibromyalgia (FM) clinical trials, including pain, tenderness, fatigue, global wellbeing, functioning, sleep disturbance, depression, anxiety, and stiffness. In randomized clinical studies, significant decreases in FIQ total scores were found with milnacipran (MLN) vs placebo (PBO), indicating overall improvements in FM severity with this treatment. A Principal Component Analysis (PCA) of data from MLN clinical studies was conducted to determine which FIQ items may be most relevant to FM patients and to further evaluate the effects of MLN on these specific items.
Methods: FIQ data were pooled from 3 double-blind trials in which FM patients were randomized to MLN 100 mg/d (n=1139), MLN 200 mg/d (n=837), or PBO (n=1133). For the PCA analysis, correlations were performed among all FIQ items based on changes from baseline in all patients receiving either MLN or PBO. Principal components with optimally weighted variables were then extracted to identify groups of FIQ items that accounted for most of the variance in FIQ results. Mean changes in FIQ scores were analyzed to evaluate whether items in the extracted components discriminated the effects of MLN treatment in this study population.
Results: Three independent groups of FIQ items were identified by PCA: Component 1 (“Core Symptoms”) primarily composed of FIQ items 2 (feel good), 4 (do job), 5 (pain), 6 (fatigue), 7 (rest), and 8 (stiffness); Component 2 (“Depression/Anxiety”) primarily composed of FIQ items 9 (anxiety) and 10 (depression); and Component 3 (“Physical Function”) primarily composed of FIQ items 1 (physical impairment) and 3 (work missed). Each FIQ item above represents a meaningful loading with at least a correlation of 0.4 with its component. Component 1 accounted for 37.3% of the total variance in the data, suggesting that this selected set of FIQ items has the strongest relevance in FM patients. Components 2 and 3 accounted for 17.5% and 17.2% of the total variance, respectively, also indicating clinical relevance. Mean changes in FIQ by treatment group indicated significant improvements with MLN (p<.01; both doses vs PBO) for all FIQ items in Components 1 and 2; discrimination of treatment effect was less consistent with the FIQ items in Component 3.
Conclusion: This PCA analysis of the FIQ resulted in 3 independent components of FM symptom domains. In the population of FM patients included in this analysis, improvements in FIQ scores were largely explained by core FM symptoms, including pain, fatigue, global wellbeing, and stiffness. Independent of these core symptoms, depression/anxiety, and physical functioning also explained a meaningful portion of the FIQ improvement. Additionally, treatment with MLN was associated with improvements in all of the items in these symptom groups.
Disclosure:
P. J. Mease,
Forest Laboratories,
2,
Forest Laboratories,
5,
Forest Laboratories,
8;
R. M. Bennett,
Forest Laboratories,
9;
R. H. Palmer,
Forest Laboratories,
3;
Y. Wang,
Forest Laboratories,
3.
« Back to 2012 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/identifying-core-symptom-domains-in-the-fibromyalgia-impact-questionnaire-principal-component-analysis-of-data-from-milnacipran-clinical-studies/