Identifying frailty: do the Frailty Index and Groningen Frailty Indicator cover different clinical perspectives? a cross-sectional study

BMC Family Practice

Research Article

René JC Eijkemans¹  Irene Drubbel²  Mattijs E Numans²  Nienke Bleijenberg²  Guido Kranenburg²  Niek J de Wit²  Marieke J Schuurmans³ 

[1] Department of Biostatistics, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Universiteitsweg 100, 3584, Utrecht, CG, The Netherlands;Department of General Practice, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Universiteitsweg 100, 3584, Utrecht, CG, The Netherlands;Department of Rehabilitation, Nursing Science and Sports, University Medical Center Utrecht, Heidelberglaan 100, 3584, Utrecht, CX, The Netherlands;

关键词: Frailty;    Primary care;    Frailty index;    Groningen frailty indicator;    Proactive care;   

DOI  :  10.1186/1471-2296-14-64

 received in 2013-01-18, accepted in 2013-04-27,  发布年份 2013

来源: Springer

【 摘 要 】

BackgroundEarly identification of frailty is important for proactive primary care. Currently, however, there is no consensus on which measure to use. Therefore, we examined whether a Frailty Index (FI), based on ICPC-coded primary care data, and the Groningen Frailty Indicator (GFI) questionnaire identify the same older people as frail.MethodsWe conducted a cross-sectional, observational study of 1,580 patients aged ≥ 60 years in a Dutch primary care center. Patients received a GFI questionnaire and were surveyed on their baseline characteristics. Frailty-screening software calculated their FI score. The GFI and FI scores were compared as continuous and dichotomised measures.ResultsFI data were available for 1549 patients (98%). 663 patients (42%) returned their GFI questionnaire. Complete GFI and FI scores were available for 638 patients (40.4%), mean age 73.4 years, 52.8% female. There was a positive correlation between the GFI and the FI (Pearson’s correlation coefficient 0.544). Using dichotomised scores, 84.3% of patients with a low FI score also had a low GFI score. In patients with a high FI score, 55.1% also had a high GFI score. A continuous FI score accurately predicted a dichotomised GFI score (AUC 0.78, 95% CI 0.74 to 0.82). Being widowed or divorced was an independent predictor of both a high GFI score in patients with a low FI score, and a high FI score in patients with a low GFI score.ConclusionsThe FI and the GFI moderately overlap in identifying frailty in community-dwelling older patients. To provide optimal proactive primary care, we suggest an initial FI screening in routine healthcare data, followed by a GFI questionnaire for patients with a high FI score or otherwise at high risk as the preferred two-step frailty screening process in primary care.

【 授权许可】

CC BY   
© Drubbel et al.; licensee BioMed Central Ltd. 2013