Predicting length of hospitalisation and social factors.

SIR—The paper by Supervia et al. [1] published in Age and Ageing showed that the Barthel Index could predict the length of stay (LOS) of elderly patients in a Short Stay Unit (SSU). Strict admission criteria are applied in these units because their expected average stay is 4–5 days. Frequently, patients with perceived 'social problems' are excluded from admission [2], assuming that social needs exist prior to health needs and that they are not caused or influenced by the latter. The authors justified their study arguing that 'the lack of autonomy could be an inconvenience in accepting discharge' (p. 339). Most would agree that those words were at least inappropriate. Functional ability is not an 'inconvenience' but a key factor when dealing with the complexity of discharging elderly patients from health institutions. The exclusion of 'social problems' according to clinicians' perspectives prior to admission does not ensure the exclusion of social needs once patients are in the units. In their study, social factors were not related to functional ability scores on admission or discharge, neither were they related to LOS. This is of particular significance in any healthcare context, but specifically in Spain, a country with an underdeveloped welfare system [3]. The study concludes that the more dependent the patients, the longer their LOS, and that females stayed longer than males. To explain this supposed causality, the authors argue that 'we think that difference on LOS between men and women on AECPD group could be explained because, in our country, in general, women are better carers of their husbands than men of their wives' (p. 341). Their classic attribution problem may have been better approached by exploring some of the social factors of their sample. Were those women married or widowed (women tend to live longer than men), how old were their carers?, etc. Most importantly, in Spain, men and women have to care for their relatives with little help from social services. In countries with limited publicly funded social care, if external support is required, discharges become complicated and therefore LOS increases. In SSUs, packages of care are needed at short notice and social services departments would not be able to provide the necessary services for safe discharges in a timely manner. I would like to remind the authors of the need to develop an understanding of social factors when assessing how elderly patients' characteristics may influence LOS [4]. …