Upper extremity function, an independent predictor of adverse events in hospitalized elderly

To examine the relationships between upper extremity function (UEF) and mortality, need for social assistance and change of residence, at discharge from hospital and at 1 month following discharge.

Observational, cohort study.

Acute Geriatric Unit of a Tertiary Teaching Hospital.

356 Consecutive patients admitted over a 6-month period. Performance of 4 UEF tasks (UEFTs) was assessed by direct observation on admission, at discharge and at 1 month after discharge: picking up a full glass, touching the scapula, cutting with a knife and unfastening a button. UEF was correlated with measures of global physical and mental functioning, namely the Barthel index, the Lawton index, Holden's FAC scale and Pfeiffer's test. The association of UEF with adverse events such as mortality, need for social assistance and change of residence was also assessed.

UEF was well-correlated with global functioning scales (p<0.001). Using multivariant models, the inability to perform 3 or 4 UEFTs on admission was an independent predictor of mortality at discharge (OR 15.2; CI 95% 5.2-44.4) and at 1 month (OR 3.3; CI 95% 1.8-6.2), of need for social assistance at discharge (OR 2.1; CI 95% 1.1-4.1) and at 1 month (OR 3.3; CI 95% 1.1-10.1), and of change of residence at discharge (OR 3.5; CI 95% 1.2-10.4).

UEF, independently of global functioning, is a predictor of adverse events in the hospitalized elderly. Its determination by direct observation may be an indirect measure of global functioning during hospitalization, avoiding potentially biased data facilitated by caregivers.