Seminar Details

Is weekend hospitalisation associated with an additional risk of death? A prognostic model derived from over 14 million hospitalisations in the National Health Service in England in 2009/10.
Monday 27th May, 2011
Seminar Room, PCPH, Upper 3rd Floor, Royal Free Hospital
Speaker Prof. Nick Freemantle

There has been considerable interest in potential additional risks associated with being hospitalised during the weekend. It has been suggested that lower standards of care and the lack of facilities that are routinely available during the week means that hospitalised patients die unnecessarily. If weekend care is substandard and incurs additional risk the health policy implications include the potential for reorganisation to provide all services over seven days with substantial implications for working practices, costs, and efficiency.

The effects of being in hospital on different days were investigated using all data regarding hospitalisations to the NHS in England in 2009/10. A Cox survival model was used to estimate relative hazards, accounting for a range of patient level risk factors and estimating the effects of hospitalisation on different days using a time dependent covariate which identified the day of the week. The day of admission was additionally included in the model as it was posited a priori that admission at the weekend may identify patients at a higher level of risk (as lower risk patients may be expected to have postponed admission outside the weekend period. Several supportive analyses were conducted including examining all deaths (not just those in hospital), censoring the first 3 days of hospital stay – as many deaths occur close to admission which may confound the analysis (and depart from constant proportional odds).

Page last modified on 26 may 11 18:15 by Maryanne Ogbogbo