GERIATRICIANS AND FUNCTIONAL OUTCOMES

Abstract

Letter to the editor comments on an article by Phelan et al. (see record 2004-16706-001). Phelan's findings are not valid; even if valid, they would not represent care provided by most geriatricians; and greater care needs to be taken with what we claim for geriatrics and the studies we undertake and publish. Selection bias plays a major role in the Phelan et al. study. The trends are noteworthy despite none of the statistical tests reaching significance. Does anyone think seeing an internist instead of a family physician raises patients' risk of functional decline? Even focusing on patients who are independent in ADLs at outset and using logistic regression does not eliminate this bias. Even if the findings by Phelan et al. were valid for their study population, they should not be generalised to geriatric care elsewhere. (PsycINFO Database Record (c) 2016 APA, all rights reserved)