Psychometric properties and factor structure of the Spanish version of the HC-PAIRS questionnaire

Abstract

To develop a Spanish version of the Health Care Providers' Pain and Impairment Relationship Scale (HC-PAIRS) and to test its psychometric properties. A forward and backward translation methodology was used to translate the questionnaire, which was then applied to 206 participants (174 physiotherapy students and 32 family physicians). The intraclass correlation coefficient was calculated to assess test-retest reliability. Internal consistency was evaluated using Cronbach's alpha and item analysis. Construct validity was measured using Pearson correlation coefficients between HC-PAIRS and FABQ, FABQ-Phys, FABQ-Work and the responses given by participants to three clinical case scenarios. An exploratory factor analysis was carried out following the Kaiser normalization criteria and principal axis factoring with an oblique rotation (quartimax). Sensitivity to change was assessed after a teaching module. Test-retest reliability was ICC 0.50 (p<0.01) and Cronbach's alpha was 0.825. The HC-PAIRS scores correlated significantly with the scores of the FABQ and also with the recommendations for work and activity given by the participants in the three clinical case scenarios. Sensitivity to change test showed an effect size of 1.5, which is considered a large change. Factor analysis suggests that the Spanish version of HC-PAIRS measures a unidimensional construct. The Spanish version of the HC-PAIRS has proven to be a reliable, valid and sensitive instrument to assess health care providers' attitudes and beliefs about LBP. It can be used in evaluating clinical practice and in undergraduate acquisition of skills and knowledge.