Service evaluation of 'GP at Door' of accident and emergency services in Eastern England.

AIM

We describe activity, outcomes, and benefits after streaming low urgency attenders to eneral practice services at oor of ccident and mergency departments (GDAE).

BACKGROUND

Many attendances to A&Es are for non-urgent health problems that could be better met by primary care rather than urgent care clinicians. It is valuable to monitor service activity, outcomes, service user demographics, and potential benefits when primary care is co-located with A&E departments.

METHODS

As a service evaluation, we describe and analyse GDAE users, reasons for presentation, wait times, outcomes, and co-located A&E wait times at two hospitals in eastern England. Distributions of outcomes, wait times, reasons for attendance, deprivation, and age groups were compared for GDAE and usual A&E attenders at each site using Pearson chi-square tests and accelerated time failure modelling. Performance in a four-hour key performance indicator was descriptively compared for co-located and similar emergency departments.

FINDINGS

Each GDAE saw about 1025 patients per month. Wait times for usual accident and emergency (A&E) care are relatively short at only one site. Reattendances were common (about 11% of unique patients), 75% of GDAE attenders were seen within 1 hour of arrival, 7% of patients initially allocated to GDAE were referred back to A&E for further investigations, and 59% of GDAE patients were treated and discharged with no further treatment or referral required. Pain, injury, infection, or feeling generally unwell each comprised > 10% of primary reasons for attendance. At James Paget University Hospital, 4.3%, and at Queen Elizabeth Hospital, 16.1% of GDAE attendances led to referral to specialist health services. GDAE attenders were younger and more socially deprived than attenders to co-located A&Es. Patients were seen quickly at both GDAE sites, but there were differences in counts of specialist referrals and wait times. Process evaluation could illuminate reasons for differences between study sites.