THE validity of the emergency department waiting time performance data published on the MyHospitals website has been called into question by a study that shows hospitals with higher proportions of urgent cases are disadvantaged by the reporting methods.
The study of 158 Australian emergency departments (EDs), published in the latest MJA, presented an analysis of waiting times reported on the website according to the proportion of patients in each of five triage categories — resuscitation, emergency, urgent, semi-urgent and non-urgent. (1)
Correlating the data this way showed that hospitals with a higher proportion of patients in the “emergency” category had poorer waiting time performance, indicating that performance data was biased in favour of EDs that reported fewer urgent patients.
The researchers wrote that one explanation for this bias was that patients allocated to higher urgency categories often needed to be admitted to hospital, so an ED managing a higher proportion of urgent patients might experience “access block”.
It may also be easier to treat patients within the allocated time frame when that time frame is longer — so EDs with a predominance of patients within the less urgent categories are more likely to meet the targets.
“If true, then hospitals do not face a level playing field when being assessed on ED performance, and EDs with higher proportions of more urgent patients are disadvantaged under the current reporting system”, they wrote.
The researchers also suggested that “undertriaging” — assigning patients to a less urgent triage category than is appropriate — might be occurring. “If there are EDs that routinely allocate ‘true’ emergency patients to the urgent category, their performance scores would be inflated”, they wrote, adding that they had no evidence of this practice.
Triaging expert Professor Drew Richardson, of the Australian National University’s medical school, agreed with the thrust of the findings.
He said triage was an “imperfect process carried out by human beings” designed to help hospitals deal with patient surges, but was now used for many other purposes.
“It has now been extended for things like casemix, funding and performance, and an awful lot of hospitals are seeing only half of their patients within triage time”, said Professor Richardson, who released figures last week showing that about one-third of patients under treatment in EDs, on average, were there waiting for inpatient beds. (2)
“People are waiting 8 hours in emergency departments to be seen at the moment, we know this is happening and that’s a failure of access, not a failure of triage.”
Professor Richardson said “triage footprinting” — evaluating if a patient had been triaged appropriately by looking at the eventual diagnosis — as well as looking at admission rates and mortality rates by triage category, would provide more accurate information about a hospital’s performance.
While triage footprinting was difficult to do, mortality and admission rates by triage category were not, he said.
“It can be done and the [Australasian College for Emergency Medicine] would recommend it be done … as they have done for a decade”, he said.
The way in which hospitals were categorised was also flawed and prevented the comparison of hospitals that were alike. “Hospital peer classifications are inadequately sensitive to properly compare groups”, Professor Richardson said.
The MJA study concluded that performance bias could be ameliorated by adjusting performance scores for variation in hospital and patient characteristics.
“Our findings suggest that, while on average such adjustments would be modest in size, they could have a substantive impact on hospital rankings”, the researchers wrote. “For example, of the 10 hospitals with the highest performance scores for the emergency triage category, only six would remain in the top 10 when adjusted performance scores were used.”
The MyHospitals website draws its data from the Australian Institute of Health and Welfare. MJA InSight approached the AIHW, but a spokeswoman for the organisation said it did not wish to comment on the research findings.
– Nicole Mackee
Posted 17 September 2012
Sorry, there are no polls available at the moment.