Send letter to journal:
Re: Dissociation of mortality at high levels of overcrowding (the death plateau)

I read Gilligans et als article with interest but was very surprised initially at their lack of association of "boarding" with increased mortality (1). I note in the discussion that they compared their findings with Richardsons and stated that they did not find a similar outcome (2). I would suggest that there are a number of reasons why this study did not find similar findings to Richardson or Sprivulis who found strong correlations between admission during overcrowded periods and death rates (2,3).

Both those studies did a detailed analysis that compared all patients admitted during periods of overcrowding (measured on a rising scale) versus patients admitted during normal access periods. They did this using short periods and direct measures of overcrowding so that data was not muddied by days when part of the day may have had low levels of overcrowding versus other periods of intense crowding. In addition they adjusted all their findings for all potential confounders that could be thought of e.g. time of day, day of week,public holiday, seasonal variation,co-morbidity, age, etc and found that these factors did not reduce the strong association of overcrowding versus mortality.

What seems clear from this studies data is that overcrowding was so intense and overwhelming that there seemed to be very few periods during this study when by any criteria the department would not be deemed severely overcrowded (1). Therefore they were unable to compare periods of overcrowding versus no overcrowding. Because this was not really measured or discussed it is difficult to know how important an issue this is. However there may well be a plateau effect in ED overcrowding v mortality in that most of the late deaths are due to delayed /poor initial care exacerbated by care away from a home ward after finally being admitted. The degree to which your ED having 100% or 150% of cubicles tied up with admitted patients at 9am impacts on this probable late effect may well be relatively minor. There is also probably a point past which the ED is so overcrowded that care cannot be made much more dysfunctional by additional overload. These features of severe overcrowding need to be looked at further. It is likely that past a certain point of dysfunction there is only so much more you can do to make the patients outcome worse!

Perhaps this paper has discovered the plateau of death. However I would very much doubt that it has found that the intolerable level of overcrwding they describe is not having a serious effect on their patients outcomes.

1) P Gilligan, S Winder, I Singh, V Gupta, P O Kelly, and D Hegarty The Boarders in the Emergency Department (BED) study Emerg Med J 2008; 25: 265-269

2) Richardson DB. Increase in patient mortality at 10 days associated with emergency department overcrowding. Med J Aust 2006; 184: 213–16