Validation of two frailty questionnaires in older patients with rheumatoid arthritis; a cross-sectional study.

OBJECTIVES: Several questionnaires exist to assess frailty, a geriatric syndrome. None of these has been validated in older patients with rheumatoid arthritis (RA). Our objective was to assess aspects of validity of two frailty questionnaires: Groningen Frailty Indicator (GFI) and Geriatric 8 (G8) among RA patients. METHODS: In a cross-sectional study among patients ≥65 years information was collected on socio-demographics, disease characteristics including comorbidities and physical function and on frailty using the GFI and G8. Content validity was assessed by linking items of the GFI and G8 to the International Classification of Functioning, Disability and Health (ICF). Classic psychometric methods were used to test hypotheses on construct validity and interpretability. RESULTS: Eighty patients (74.6 years (SD 5.9); 66% female) participated. The GFI has more items on social and mental functions; the G8 more on functions of the digestive system (e.g. nutritional status). As hypothesised, correlations (r) with physical function (RGFI=0.54; RG8=0.56) and disease activity (RGFI=0.24; RG8=0.36) were moderate to weak. However, correlations with age (RGFI=0.20; RG8=0.11) or comorbidities (RGFI=0.30; RG8=0.16) were lower than expected. Instrument-specific thresholds classified 43 (54%) of participants as frail on the GFI and 44 (55%) on the G8; 33 (41%) were frail on both instruments. CONCLUSIONS: The GFI and G8 differ in content with more emphasis on nutritional status for the G8. Both instruments are insensitive to age and comorbidities. Before deciding on their usefulness in RA, their predictive validity for mortality and resource utilisation independent of disease activity and physical function should be further evaluated.