Report comment

There is a lot here that most Emergency Physicians would agree on.
What you describe is a Bayesian phenomenon reflecting the different populations:

A GP attender has a 1% chance of needing hospital admission.
A patient who gets to my hospital ED by ambulance has a >50% chance of needing hospital admission.

GPs and Emergency Physicians are trained to work with different populations of patients.
GPs are more accurate at assessing when something doesn't need to be done (because it is true in most of their patients)
Emergency Physicians are better at detecting serious disease in patients in whom it may well be present.

What is difficult to help people understand is that just because a patient attends the ED, they are automatically part of a different population group, in which serious disease is much more common, and therefore we have to test them more carefully.

If you send a lot of GP patients to the ED, they will get more investigations than if a GP saw them and reassured them.
Conversely, if you put a GP into the majors side, and I have seen this done four times, they will fail to spot significant disease, each time with catastrophic results.

There was a very good analysis of the current situation by the Health Select Committee in July 2013
http://tinyurl.com/pqr723a
that identified a root cause as the very poor data about urgent and emergency care. Their description was that we are 'flying blind'.

The CEM is continuing to press for a complete review of the scope and content of acute service data, so that these problems can be better defined and hence resolved.

The immediate solution is that there must be GPs co-located with Emergency Departments, as the press release from multiple Colleges yesterday emphasised.

COI - Emergency / Acute Physician, CEM chair of Informatics group.