Vicente Martín Moreno ^1* María Inmaculada Martínez Sanz ¹ Amanda Martín Fernández ² Sara Guerra Maroto ¹ Eva Sevillano Fuentes ¹ Elena Pérez Rico ¹ Irene Sánchez González ¹ Miriam Fernández Gallardo ¹ Julia Herranz Hernando ¹ María Palma Benítez Calderón ¹ Laura Calderón Jiménez ¹ Elena Sánchez Rodríguez ¹ Miguel Recuero Vázquez ¹ Helena Alonso Samperiz ¹ Irene León Saiz ¹ Juana Marcos Guerra ¹

• ¹ Madrid Health Service, Madrid, Madrid, Spain
• ² Polibea Concierto, Madrid, Spain

Introduction: Functional dependence for the performance of basic activities of daily living (ADLs) is associated with increased mortality, analyzing in this study whether the Barthel index and its activities discriminate long-term mortality risk, and whether changes in this index are necessary to adapt it as a method for detecting mortality risk.Methods: Longitudinal study, carried out at the Orcasitas Health Center, Madrid (Spain), on the functional dependent population (Barthel ≤60). Was included 127 persons, with a mean age of 86 years (78.7% women and 21.3% men). Functional capacity was assessed using the Barthel index and this index and each item it contains were analyzed as a test in relation to survival at three years, using tools that evaluate precision, discrimination and calibration. Date of death was obtained from the Health System.Results: Greater dependency to perform chair-to-bed transfers was associated with an increased risk of mortality (HR 2.957; CI 1.678-5.211). Also, individuals with severe (HR 0.492; CI 0.290-0.865) and moderate (HR 0.574; CI 0.355-0.927) ADL-dependence had a reduced risk of mortality when more independent in chair-to-bed transfers. Among people with moderate ADL-dependence, this percentage was 48%. Using dependenceindependence for chair-to-bed transfer as a screening test for mortality, the test showed high sensitivity (0.91) and specificity (0.83), a positive likelihood ratio of 5.45, and a negative likelihood ratio of 0.11. The area under the ROC curve was 0.814 (CI 0.658-0.970; p=0.001), with a ꭕ2= 0.235; p=0.889, according to Hosmer-Lemeshow test. The concordance C index was 0.814. According to Nagelkerke's R 2 , the model explained 53.1% of the variance in survival. As a screening test, "chair-to-bed transfer" was superior to Barthel index.Conclusions: ADL-dependence for chair-to-bed transfers is an independent risk factor for mortality for any level of dependency. Therefore, a new classification of the Barthel index is proposed, in which "being totally dependent or requiring great assistance to perform chair-to-bed transfers" is considered severe dependence, even when the total score obtained via the Barthel Index is ≥40. We propose its use as a screening test, in parallel to the Barthel index. The Barthel Index could have blind spots to discriminate mortality risk.