This site is intended for health professionals only

Managing the seriously ill patient in the community

It feels like we are living in some surreal parallel universe. The way we work has changed beyond recognition, with most of us now seeing patients on video call and the waiting room looking like a ghost town. Things are no different in hospital, where decisions about ITU admission are following increasingly tight criteria, and discharge thresholds are lowering.

As a result, we are facing a significant number of unwell patients.

There are probably two categories: patients who have been discharged earlier than would be usual, for instance after emergency surgery; second, elderly and frail patients who may have less support than usual from other community services, or who may be unwell with Covid-19 symptoms and are past the point where we would usually admit to hospital.

Patients discharged early

Patients who are discharged early after a medical or surgical admission could create many problems depending on what they were admitted for. A patient with an acute kidney injury, for example, would usually need close monitoring of their renal function.

For patients with cellulitis you can use the Eron score

The frequency of blood tests will often be specified in the discharge summary – but this might cause problems if, as is the case in my area, the local hospital has cancelled all but the most urgent GP bloods. District nurses may be able to step in if the patient is unable to leave the house, but they will also be short-staffed. I have found that during this pandemic my local consultants have been incredibly accessible on the phone for advice, so investigate whether this is possible in your area. Discussing the patient with the consultant under whom they were admitted can give you the confidence to know how closely they need monitoring, and also a degree of medicolegal protection.

Infective admissions

Patients who have been sent home earlier than usual after an infective admission (such as pneumonia or cellulitis) may remain nominally under the care of their consultant with a hospital-at-home team visiting to give IV antibiotics, or they may come straight back to you. If they start to deteriorate, you will have to decide whether to keep them in the community or send them back to hospital where they run the risk of catching Covid-19 on top of the condition they were originally admitted for.

These decisions will often be made using your experience and ‘Spidey sense’. I personally use the insomnia test as my final arbiter – if I do this (or don’t do this), will it keep me awake tonight? For many clinical presentations there are also more objective measurements to help make the decision, and protect you medicolegally.

For patients with cellulitis you can use the Eron score, which runs from I to IV. Class I is an otherwise healthy patient who is not systemically unwell and can stay at home with oral antibiotics. Patients in class II are either systemically unwell or systemically well but have co-morbidities; this is the group where you would consider intravenous antibiotics. Patients who are in groups III and IV are significantly unwell with tachycardia, confusion or full-blown sepsis and need re-admission, unless a decision has been made to palliate.1

For treating pneumonia the CRB-65 score is reasonably well known,2 though of course the balance of risks and benefits from admitting or staying at home is not the same as in normal times and this will need to be discussed. Consider whether some of your patients who regularly get cellulitis, particularly those with lymphoedema, should have antibiotics for longer than a week or should be on prophylactic antibiotics to prevent another event – some guidelines3,4 recommend this for those with more than two episodes a year, though the evidence is weak. Patients discharged early after an infarct may need significant supervision of their medication, for example titrating up doses of β-blockers and ACE inhibitors – if you can get the renal function and blood pressure checked by the district nurses this can usually be done remotely.

Discharged after surgery

Patients discharged early after surgery present their own challenges, both physical and psychological. If they presented acutely and were found to have a malignancy (for example a bowel obstruction from a previously undiagnosed cancer), they may be lacking some of the usual support from the hospital team and may seek this from you. It is very difficult when we are asked questions that are way outside our knowledge and it is always best to be honest if you don’t know. If you think the patient might have slipped through the follow-up net, chase this with the hospital and consider putting them in touch with Macmillan or local cancer charities.

The range of physical issues that can occur after surgery is probably too wide to discuss in a single article, though some are mentioned in the box below. It might include infection, issues with drains, stitch removal or a lack of orthopaedic provision such as crutches or Aircasts.

Common complications of surgery


• The Wells score gives one point for three days of immobility or surgery requiring general or regional anaesthesia, so most post-operative patients have a positive Wells score.

• Joint infection must be considered after orthopaedic surgery.

• If access to a Doppler is delayed, use the BNF to work out the dose of low-molecular weight heparin; you will need to know the patient’s weight (see page 24 for more details).

Post-operative pain

• Go back to first principles – analgesia should be regular rather than on demand; patients often take pride in not using it until desperate but this approach should be discouraged.

• The pain ladder, developed for cancer pain, can be useful.13

• Consider the type of surgery – could pain be a collection deep in the pelvis, compartment syndrome or incipient bowel obstruction? Much as we are trying to keep people out of hospital, some complications have to go back in and delay can cause harm.


• Many patients are on oral antibiotics after surgery, so if you suspect infection check if they are already taking something.

• For those on warfarin, antibiotics may upset the INR – consider how to access INR testing or whether patients should be changed to a DOAC.

• If you have access to hospital results, check to see if swabs taken towards the end of the admission are back.


• The pathogenesis of this may not be the same as for community-acquired pneumonia – think aspiration, pseudomonas, klebsiella and Staphylococcus aureus.14

• Consider whether patients with respiratory complication after surgery would be helped by domiciliary physiotherapy.

• If the patient needs IV antibiotics, could this be done at home?


• ‘First do no harm’ is our maxim, but don’t forget iatrogenesis as a cause of confusion. Is the patient uraemic, either from dehydration or because of a surgical complication such as ureteric damage? Are they confused because they are taking the opiates given to them by the hospital as well as the ones they had at home?

• Remember that true delirium fluctuates, being better in the morning and worse at night.15 If you have excluded reversible causes and the confusion doesn’t fluctuate and is progressive, have you unmasked a longer term cognitive impairment?

It is important to observe the fine line between taking more responsibility in a time of crisis while not acting outside your area of competency. If you’re in doubt and unable to sort the patient out, go back to your first principles. Take a good history, use your communication skills to identify ideas, concerns and expectations, examine the patient thoroughly (while wearing appropriate protective equipment), formulate a differential diagnosis and make a plan. Bounce ideas off your colleagues – we are all in the same boat. Remember your surgical sieve from medical school – if struggling for an answer, this systematic approach may help. My favourite for its simplicity is VINDICATE – vascular, infection, neoplasm, degenerative/drugs, iatrogenic/intoxication, congenital, autoimmune, trauma, endocrine/metabolic. Some of these are less likely after surgery, but this sort of structured thinking reduces the risk of missing something.

Elderly and frail patients

Communication is key

The combination of immobility and the possibility of nosocomial infection means that, even in normal times, we try to avoid admission whenever possible.

It’s really important to use our communication skills early on and to make sure that the patient has a good understanding of why we are even more reluctant to admit them – that is, if we can find a hospital in the first place. Including any relatives in the discussion is also sensible; be ready to introduce the topic of advanced care planning and thresholds of care. We are used to doing this for patients with a known life-limiting illness but now that admission carries even more risks for a frail elderly patient, it is reasonable to discuss the pros and cons of admission versus keeping comfortable in the community. Make sure that any escalation of care or DNR decisions are communicated to the out-of-hours team by whatever shared database you have available.

Treating constipation

We should be proactive about trying to prevent ailments such as constipation, dehydration or a UTI that could lead to a frail elderly patient being admitted. If doing a ward round at a care home, remind the staff that they should encourage their residents to drink plenty of fluids, to avoid dehydration and reduce the risk of UTI.

Remember if treating someone for constipation to consider the cause when choosing a laxative. Opioid-induced constipation should be treated with a stimulant laxative (such as senna) and if a second agent is needed, an osmotic laxative (such as lactulose) or a surface wetting laxative such as docusate. Bulk-forming laxatives can make things worse because they stimulate peristalsis and can cause colic and occasionally obstruction.5,6 Conversely, for constipation with no obvious cause, lifestyle changes are important (such as increased fluid and fibre), after which a bulk-forming laxative (such as ispaghula) would be first line, assuming there is an adequate fluid intake, with an osmotic laxative being second line. Patients and their carers should be clearly told that once regular laxatives are started, they should be weaned off after two to four weeks of regular bowel movements. Stopping suddenly or too soon can increase the risk of faecal impaction.1

Acute delirium symptoms

These patients can be difficult to investigate in primary care and often end up in hospital, but at the moment that may not be an option and we may have to investigate more in primary care. If you are making any clinical decision, such as whether to send to hospital, and your decision is affected by the pandemic, remember to document that in the notes, in case of future legal action.

Bulk-forming laxatives can cause colic and occasionally obstruction

When considering the causes of delirium, the mnemonic PINCH ME is useful – Pain, Infection, poor Nutrition, Constipation, Hydration, Medication, Environment change. Public Health England7 advises that we don’t attribute delirium with no urinary symptoms to a UTI without actively considering these other causes, and that we don’t dip the urine in those aged over 65 as it adds nothing to the decision-making process – over half of this age group will have asymptomatic bacteriuria which does not need treating.

Managing the risk of falls

Falls are another problem that often land frail elderly patients in A&E; around 30% of those aged over 65 will have at least one fall per year.8 With routine outpatients largely being shut or done by phone only, we are unlikely to have access to full falls clinics. Community services such as occupational therapy that can assess for dangers in the home may be stretched or not operating.

For normal times, NICE suggests that we identify two groups of patients who are at risk of falls; those aged 65 or over who have already fallen in the last year, and those who have not yet fallen but are at risk of doing so. This risk might be due to impaired mobility, cognitive or visual impairment, iatrogenic causes, alcohol misuse or general frailty. However, it may be that those who are falling are the ones who will gain the most benefit from our limited time, and those who are just at risk may only need the most basic risk factor management. The tests to assess frailty8 are generally done when actually with the patient, and of course this group should be at home and not seeing us. Similarly, interventions we make for falls require a level of multidisciplinary intervention that isn’t possible right now.

But there are simple things we can do. If you happen to be visiting and notice obvious trip hazards, mention it to the patient. If you are worried about a patient, ask for permission to discuss it with their family – they may be able to intervene more than you can. Consider whether blood pressure control should be relaxed a bit, if they really need that amitryptiline or if you could wean the benzodiazepine that you have reluctantly continued for many years, having inherited it from a previous GP. If you really feel a patient isn’t safe at home, and they don’t have the capacity to make the decision, an adult safeguarding referral may be appropriate. Social services should still be functioning in the normal way (though may be affected by staff illness) and our frail patients don’t stop being vulnerable because of a pandemic. Quite the opposite.

Shielded patients

Another aspect of caring for the frail elderly is that many will be in the group who have been advised to shield – not leave the house for 12 weeks.9 They may have no local friends or relatives, no internet access for an online shop (and who can get a delivery slot anyway these days?) and only a small fridge with little room to stock up, this can pose a significant dilemma. While this is a social problem rather than a medical one, none of us want to see our patients go without food for fear of leaving the house. Most councils now have an official phone number or website where you can request support for these patients and many local groups are springing up. Caution has to be exercised for groups with no official mandate, but in my experience volunteers will often be happy to show you their DBS certificate, and hopefully before too long they will become officially accredited under the NHS volunteers scheme.10 Consider giving one person in the surgery the job of keeping an up-to-date list of what is available; the Covid aid website may also be of help.11

And don’t forget – look after yourself and your colleagues and stay safe.

Dr Toni Hazell is a GP in north London


1 NICE CKS. Cellulitis – acute. 2019.!diagnosisSub

2 NICE CKS. Chest infections – adult. 2019.!diagnosisSub

3 Primary care dermatology society. Cellulitis, erysipelas, and necrotising fasciitis. 2018.

4 British Lymphology Society. Consensus Document on the Management of Cellulitis in Lymphoedema. 2016.

5 NICE CKS. Constipation in adults. 2019.!scenario

6 NICE CKS. Palliative care – constipation. 2016.!scenario:1

7 Public Health England. Diagnosis of urinary tract infections: quick reference tool for primary care. 2018.

8 NICE CKS. Falls – risk assessment. 2019.!backgroundSub:1

9 Public Health England. Guidance on shielding and protecting people defined on medical grounds as extremely vulnerable from COVID-19. 2020.

10 NHS Volunteers.

11 Covid aid volunteers scheme.

12 NICE CKS. Deep vein thrombosis. 2019.!scenarioRecommendation

13 Pain ladder.

14 Chughtai M, Gwam C, Mohamed N et al. The Epidemiology and Risk Factors for Postoperative Pneumonia. J Clin Med Res 2017;9:466-75

15 Royal College of Anaesthetists. Post-operative confusion. 2017.


Visit Pulse Reference for details on 140 symptoms, including easily searchable symptoms and categories, offering you a free platform to check symptoms and receive potential diagnoses during consultations.