Avoiding hospital admissions this winter – the challenge for general practice

Nada Khan is an Exeter-based GP and an NIHR Academic Clinical Lecturer in General Practice at the University of Exeter. She is also an Associate Editor at the BJGP

Autumn can feel like the calm before the storm of winter in healthcare.  Winter pressures impact across primary and secondary healthcare. In anticipation, NHS England has sent a letter to all Integrated Care Boards (ICBs) and NHS trusts asking them to take steps to ‘maintain patient safety and experience’ amongst significant strain in urgent and emergency care pathways.¹

The letter asks primary and community services to proactively identify and manage people with complex needs and long-term conditions to optimise care before winter and actively avoid hospital admission.  The letter suggests that ‘alternatives to hospital attendance and admission’ should be provided, in particular to people with complex health needs, frail older people, children and those with mental health issues who are ‘better serviced with a community response’ outside of hospital.¹  This might leave some GPs scratching their heads, wondering where the capacity to proactively identify and manage these at-risk patients is going to come from. What are the alternatives to hospital attendance and admission, and what kinds of admissions could, or should be avoided?

What is an avoidable hospital admission?

We need to be careful in terminology here between ‘unplanned’, ‘inappropriate’ and ‘avoidable’ hospital admissions.  These terms are used interchangeably but mean different things.  An interesting discourse analysis of the concepts ‘avoidable’ and ‘inappropriate’ hospital admissions starts by pointing out that policy makers sometimes wrongly assume that there is an optimum level of hospital admissions, and that reducing admissions or referrals improve health care delivery or efficiency.²  Defining an avoidable admission involves much more than just the decision a GP makes with a patient sat in front of them.  An admission to hospital can be avoided if a different course of action was taken at any stage along an illness trajectory, starting from primary prevention through to secondary prevention and later decisions in primary or secondary care.

But what is an ‘inappropriate’ hospital admission?  There’s no standardised definition of ‘inappropriateness’, which then makes it difficult to understand and evaluate the problem.³  A hospital admission might not be deemed appropriate, as some patients might need respite or home care instead of a hospital admission, but to avoid admissions in this group of patients, GPs and patients need access to alternative social care services.  

Unplanned hospital admissions

One way to think about hospital admissions is to focus on unplanned admissions, when someone is admitted to hospital urgently and unexpectedly, as potentially preventable.  The Nuffield Trust looked at groups of conditions to determine trends in potentially preventable emergency hospital admissions.⁴  They looked at ambulatory care sensitive (ACS) conditions where community and person-centred care could prevent an admission, and urgent care sensitive conditions like COPD where a patient could have an acute exacerbation, but that the system should try and treat and manage at home without an admission where possible.  Their trend analysis shows that hospital admission rates for both groups of conditions have remained fairly stable since 2008.  Admissions for urgent care sensitive conditions are often necessary, but the Nuffield Trust suggests that a higher rate might suggest that some admissions could be avoided.

Researchers are increasingly using routine data to try to identify those at highest risk of admission.  I recently talked to Dr Jet Klunder, a GP trainee and researcher based in the Netherlands, for the BJGP podcast about a paper she has recently published in the BJGP focussing on a prediction model to predict unplanned admissions to hospital amongst older adults.⁵  People with a previous hospital admission in the last year, those taking more than five medications, and with a history of COPD or heart failure were at highest risk of an unplanned admission.  This work is at an early stage and needs validation in another dataset, but Jet’s plan is that this kind of prediction model might act like a flag in general practice electronic health records to indicate which patients are at highest risk of a hospital admission to guide proactive management plans.  Models like this might help with identifying people at risk of hospital admission, but the next step is knowing what to do.

Preventing hospital admissions

The NHS England letter to ICBs suggests that GPs should first act to proactively identify and then manage patients to actively avoid hospital admissions.  It’s currently unclear how to achieve this in practice as the evidence for community or primary care interventions to reduce hospital admission is mixed.  While telemedicine and self-management might help reduce unplanned admissions in some populations, the majority of interventions don’t work in the general population.⁶  And many of these interventions involve a much wider community and secondary care team alongside GPs.

A randomised control trial of using a comprehensive geriatric assessment hospital at home intervention involved a nurse practitioner working alongside a geriatrician and other multidisciplinary team members including occupational therapists and social care to provide an alternative to hospital for the participants, but didn’t show any difference in people living at home after six months, though the service didn’t lead to worse outcomes compared to a hospital admission.⁷  Looking more widely, a systematic review of interventions looking at decreasing emergency department use and hospital admissions in older people showed that community-interventions that include a comprehensive geriatric assessment, and multidisciplinary teams involving a geriatrician were more likely to reduce acute care use.⁸  Home visits from dual or interdisciplinary teams, including nurses, GPs or geriatricians had the best outcomes, but these were not a quick fix, with most of these interventions running over 6 months to a three year period.  But buyer beware.  Richard Holland has spoken about his trial of a pharmacist-led home medication review after hospital admission amongst patients aged over 80, which found that the actually intervention increased hospital admission and didn’t improve quality of life or mortality in the intervention group.⁹

An alternative to admitting patients is the use of virtual wards, which provide hospital care at home through remote monitoring often under the care of a hospital team.  Virtual ward initiatives are highly varied, and working out whether different operational models lead to cost-savings, fewer admissions or a better patient experience is complex.¹⁰

The NHS England letter suggests that GPs should target people with complex and long-term conditions for admission avoidance.  While this might seem like a good place to start when trying to reduce unplanned admissions, Martin Roland has pointed out that focussing interventions just on people at higher risk of hospital admission is problematic as high risk patients don’t account for most admissions.¹¹  To reduce overall numbers admissions, the shift needs to focus from just the smaller proportion of high risk patients to the wider population, but it’s unclear if these wider interventions are effective, cost-effective, or safe.

Back to another cornerstone of general practice – continuity of care.  Relational continuity of care is associated with fewer hospital admission for ambulatory sensitive conditions, or those taken to be manageable in general practice.¹²  Amongst patients with dementia, those with the highest level of relational continuity had a lower risk of emergency admission to hospital.¹³  Could better continuity of care in general practice help reduce hospital admissions, and how does this work?  Continuity might lead to better management of long-term conditions, or a better working relationship between GPs and patients with complex multimorbidity who understand each other and patient preferences better.  We need to understand more about the reasons behind these associations, and explore whether continuity reduces hospital admission safely and appropriately.

Finally, and importantly, it’s important to keep in mind that some hospital admissions are necessary, so fewer hospital admissions is not always better, or safer.¹¹   Telling people that they could, or should avoid going to hospital, or suggesting that a hospital admission is inappropriate could not only discourage people from help-seeking, but could also result in delayed diagnoses or serious and preventable complications.²

Final thoughts

As we gear up for winter, can we, and should we be identifying and trying to proactively manage patients to prevent unplanned hospital admissions?  Framing hospital admissions as inappropriate or avoidable comes with its own problems, as does knowing who to target or what to do, and when.  In terms of capacity, I’d be hard pressed to find a GP who feels that we have the time or resource to deliver this ask from NHS England amidst increasing workload pressures in practice.  Perhaps this is a role that members of the Additional Roles Reimbursement Scheme (ARRS) could fulfil by going out and completing a comprehensive geriatric assessment, taking into account the views of the patient and their carers, and flagging at risk patients for a further review from other members of the multidisciplinary or community response team. An increasing political impetus to shift care from hospital to community settings means that we may increasingly see these asks to keep patients out of hospital, but the challenge will be making sure we have the funding and infrastructure to do so safely.

References

1. Winter and H2 priorities: NHS England; 2024 [Available from: https://www.england.nhs.uk/long-read/winter-and-h2-priorities/.
2. Clubbs Coldron B, MacRury S, Coates V, Khamis A. Redefining avoidable and inappropriate admissions. Public Health. 2022;202:66-73.
3. Thwaites R, Glasby J, le Mesurier N, Littlechild R. Room for one more? A review of the literature on ‘inappropriate’ admissions to hospital for older people in the English NHS. Health Soc Care Community. 2017;25(1):1-10.
4. Potentially preventable emergency admissions: Nuffield Trust; 2024 [Available from: https://www.nuffieldtrust.org.uk/resource/potentially-preventable-emergency-hospital-admissions.
5. Klunder JH, Heymans MW, van der Heide I, Verheij RA, Maarsingh OR, van Hout HP, et al. Predicting unplanned admissions to hospital in older adults using routinely recorded general practice data: development and validation of a prediction model. Br J Gen Pract. 2024;74(746):e628-e36.
6. Purdy SP, S.; Huntley, A.; Thomas, R.L.; Mann, M.; Huws, D.W.; Brindle, P.; Elwyn, G. Interventions to reduce unplanned hospital admissions: a series of systematic reviews. 2012.
7. Shepperd S, Ellis G, Schiff R, Stott DJ, Young J. Is Comprehensive Geriatric Assessment Admission Avoidance Hospital at Home an Alternative to Hospital Admission for Older Persons? Ann Intern Med. 2021;174(11):1633-4.
8. Pritchard C, Ness A, Symonds N, Siarkowski M, Broadfoot M, McBrien KA, et al. Effectiveness of hospital avoidance interventions among elderly patients: A systematic review. CJEM. 2020;22(4):504-13.
9. Holland R, Lenaghan E, Harvey I, Smith R, Shepstone L, Lipp A, et al. Does home based medication review keep older people out of hospital? The HOMER randomised controlled trial. BMJ. 2005;330(7486):293.
10. Co M. Virtual wards: no place like home? : The Health Foundation 2024 [Available from: https://www.health.org.uk/news-and-comment/blogs/virtual-wards-no-place-like-home.
11. Roland M, Abel G. Reducing emergency admissions: are we on the right track? BMJ. 2012;345:e6017.
12. Barker I, Steventon A, Deeny SR. Association between continuity of care in general practice and hospital admissions for ambulatory care sensitive conditions: cross sectional study of routinely collected, person level data. BMJ. 2017;356:j84.
13. Delgado J, Evans PH, Gray DP, Sidaway-Lee K, Allan L, Clare L, et al. Continuity of GP care for patients with dementia: impact on prescribing and the health of patients. Br J Gen Pract. 2022;72(715):e91-e8.

Featured image by Fabrice Villard on Unsplash