POS1492-HPR PSYCHOMETRIC PROPERTIES OF THE FIBROMYALGIA IMPACT QUESTIONNAIRE – REVISED IN FIBROMYALGIA AND CHRONIC WIDESPREAD PAIN: A RASCH ANALYSIS

Abstract

BackgroundThe Fibromyalgia Impact Questionnaire - Revised (FIQR) is one of the most commonly used self-rating instruments developed to quantify the severity of disease burden and impact of disease on functional ability and social participation in patients with fibromyalgia (FM) [1]. The FIQR consists of 21 individual items categorized within three different domains; ‘function’, ‘overall impact’ and ‘symptoms’ [2].The FIQR that has been translated into several languages, has mainly been validated based on classical test theory (CCT) - except for the Italian version [3]. The psychometric properties of the Italian version of the FIQR was evaluated by Salafi et al. based on item response theory using Rasch analysis [3]. The benefit of using Rasch measurement models, is that it allows for a detailed analyses of an instrument’s rating scale properties and aspects of validity, including fit of individual scale items to a unidimensional model [4]. Although the FIQR demonstrated adequate psychometric properties in the original studies based on CCT, the Rasch analysis in the Italian study revealed problems with the rating scale properties [2,3].The FIQR is a frequently used questionnaire when assessing patients with FM in Denmark, but the Danish version of the FIQR has not yet been validated.ObjectivesThe aim of this study was to evaluate the psychometric properties of the Danish version of the Fibromyalgia Impact Questionnaire Revised (FIQR), when used to quantify the severity of disease burden in a Danish population of patients with Chronic widespread pain (CWP) including Fibromyalgia (FM).MethodsA total of 924 participants diagnosed with CWP and/or FM completed an electronic version of the FIQR via touchscreens in the clinic at referral for specialist care. Data was collected from January 1st, 2018 to September 3rd, 2020. Rasch measurement methods were applied.ResultsRating scale analysis suggested multiple threshold disordering in the 0 to 10-category rating scale. A principal component analysis suggested assessment of a multidimensional construct. Thus, the Rasch analysis of the full FIQR was discontinued. Instead Rasch analyses were performed on the two subscales: ‘function’ and ‘symptoms’. By collapsing the rating scale to a 0 to 4-category scale, the remaining threshold disordering of both subscale was solved. Only the symptom subscale indicated a multidimensionality. There was underfitting misfit of item 21 and overfitting misfit of item 12. No significant Differential Item Functioning was found defined by sex, ethnicity, or education.ConclusionThe FIQR should be considered as an instrument consisting of three separate subscales representing ‘function’, ‘overall impact’ and ‘symptoms’. We recommend calculating and reporting on both a 0 to 10- and a 0 to 4-category scale. Also, if using the total FIQR score as an outcome measure, this should be done with caution, until revision of the rating scale.