Measuring fatigue with multiple instruments in a Brazilian cohort of early rheumatoid arthritis patients

• Autor(a) principal: Diniz,Leonardo Rios
• Data de Publicação: 2017
• Outros Autores: Balsamo,Sandor, Souza,Talita Yokoy de, Muniz,Luciana Feitosa, Martins,Wagner Rodrigues, Mota,Licia Maria Henrique da
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Revista Brasileira de Reumatologia (Online)
• Texto Completo: http://old.scielo.br/scielo.php?script=sci_arttext&pid=S0482-50042017000500431
• topic: Disability; Fatigue; Rheumatoid arthritis

Abstract Objective: To assess the prevalence of fatigue in a Brazilian population with early rheumatoid arthritis using multiple instruments, and the predictors of these instruments by differents independent variables. Methods: Cross-sectional study with direct interview and medical records review. Fatigue, dependent variable, was assessed using eight instruments: Profile of Mood States (POMS), Multidimensional Assessment of Fatigue scale (MAF), Fatigue Severity Scale (FSS), Bristol Rheumatoid Arthritis Fatigue Multidimensional Questionnaire (BRAF-MDQ), Numerical Rating Scales (BRAF-NRS), Short-form Survey 36 (SF-36), Functional Assessment of Chronic Illness Therapy Fatigue Scale (FACIT-F) and Visual Analogic Scale for Fatigue (VASf). Independent variables: sociodemographic, clinical and serological, were measured using medical records and direct interview. Disability and disease activity were assessed using the Health Assessment Questionnaire (HAQ) and disease activity assessed using the Disease Activity Score 28 joints (DAS28). The scores of scales demonstrated the level of fatigue and multiple linear regression method used in statistical analysis to demonstrate prediction models. Results: A total of 80 patients was assessed, and 57 reported clinically relevant fatigue (VASf > 2), representing 71.25% prevalence point (51 women [89.5%], mean age 48.35 ± 15 years, and mean disease duration of 4.92 ± 3.8 years). Eight predictive models showed statistical significance, one for each fatigue instrument. The highest coefficient of determination (R 2) was 56% for SF-36 and the lowest (R 2 = 21%) for FSS. The HAQ was the only independent variable to predict fatigue on all instruments. Conclusion: Clinically relevant fatigue is a highly prevalent symptom and is mostly predicted by disability and age in the population assessed.