Hospital admission rates vary widely across England's emergency departments

Neither local case mix, primary care provision, nor geography explain the large differences in hospital admission rates, ranging from less than 10% to more than 40%, from emergency departments across England, finds research published online in Emergency Medicine Journal.

The findings imply that locally tailored strategies, which take account of local resources, such as urgent care facilities and transport options, may more effectively reduce unnecessary admissions than uniform national policies, say the researchers.

The growth in hospital admissions from the emergency department has become an international concern due to their impact on costs, hospital crowding, and potential readmissions, explain the researchers.

But rather than focusing on identifying predictive factors, which many previously published studies have done, the researchers wanted to understand the variation in these admission rates once local area characteristics had been accounted for.

They used national anonymised data on emergency care and admissions for hospitals across England for the year 2024-5 to measure the variation in the likelihood of a patient being admitted from an emergency department.

They adjusted for local case mix, attendance (method of arrival; primary reason for seeking emergency care), geography, and general practice characteristics (access; workforce availability/1000 patients; quality of care; availability of alternative urgent care facilities).

After exclusions due to missing patient and primary care provider information, the data analysis was based on 15,201,063 emergency department attendances.

Around 1 in 4 (26%) of these attendances resulted in admission to hospital. Three quarters (74%) of attendances occurred during weekdays, and around a quarter (26.5%) of patients arrived by ambulance; 14% lived in the most socioeconomically deprived districts.

Admission rates varied widely, ranging from rates of less than 10% to more than 40%. To estimate the probability of a patient being admitted, the researchers included a set of 3 variables describing the characteristics of patients' attendance, their residential area, and their GP practice.

Unsurprisingly, the probabilities of hospital admission varied significantly, influenced by patient demographics, socioeconomic status, illness severity, and access to the emergency department.

And older adults, men and those living in areas of high deprivation had higher probabilities of being admitted after an emergency department attendance.

Patients arriving by ambulance or during periods when GP practices were closed, as well as patients living near to their GP practice were also more likely to be admitted.

GP practice characteristics, such as awareness of extended hours, staffing levels, and the prevalence of long term conditions affected admission rates too.

But the 3 sets of variables accounted for only around a fifth (22%) of the variation in admission rates.

And even after extensive adjustment for local hospital case mix, substantial unexplained variation in hospital admission rates remained, ranging from 2% to 42%.

Areas like North East and North West London and the South West still had the highest rates, while areas such as Lincolnshire, Yorkshire, and the North East had around average rates.

This is an observational study, and as such, can't establish cause and effect. And the researchers note that any study drawing on administrative data is susceptible to completeness, accuracy, and coding issues.

Unobserved factors, such as differences in hospital culture or in health service configuration might explain some of the residual differences in admissions, they add.

Nevertheless, the findings have policy implications, they suggest.

"A policy implication is that a uniform 'one-size-fits-all' national approach to managing emergency admissions is unlikely to be effective due to the underlying local and hospital-specific factors driving the observed variations," they write.

"A tailored strategy may be required, which takes regional and institutional factors into account, and is directed at local resource availability, patient needs, and hospital capabilities.

"Although we acknowledge that unobserved case mix measures and local factors remain, it is unlikely that additional measures would eliminate such large variations in [emergency department] admissions," they conclude.

Source:

BMJ Group

Journal reference: