OBJECTIVES

To investigate the distribution of 25-question Geriatric Locomotive Function Scale (GLFS-25) scores in Japanese rheumatoid arthritis (RA) patients and evaluate relationships with clinical variables.

METHODS

Among 15,115 patients registered in the NinJa database for fiscal year 2015, 1710 with complete GLFS-25 and disease activity score-28 (DAS28) data were analyzed. Correlations between GLFS-25 score and clinical variables were assessed by Spearman coefficients. Mean GLFS-25 scores were compared among DAS28 groups (<2.6, 2.6-3.1, 3.2-5.0, ≥5.1) using the Kruskal-Wallis test. To evaluate the performance of the GLFS-25 and Health Assessment Questionnaire Disability Index (HAQ-DI) for predicting DAS28 ≥ 3.2 (moderate/high disease activity), receiver operator characteristic (ROC) curves were constructed.

RESULTS

GLFS-25 score was significantly correlated with age, disease duration, DAS28, and HAQ-DI. GLFS-25 score increased in parallel with DAS28. The proportion of patients with locomotive syndrome stage 2 also increased with DAS28. Area under the curve values for HAQ-DI and GLFS-25 score were 0.739 and 0.768, respectively. At a GLFS-25 positive cutoff score ≥16, sensitivity was 0.716 and specificity was 0.661 for predicting DAS28 ≥ 3.2.

CONCLUSION

This study documents the GLFS-25 score distribution in Japanese RA patients and demonstrates that GLFS-25 is a useful measure for evaluating functional ability in RA.