Assessing acute illness

Most symptoms never get anywhere near a healthcare professional – although given how busy general practice is, that may be hard to believe – but when patients do present it is important to be able to recognise signs of acute illness

The term ‘symptom iceberg’ was coined by Professor David Hannay in 1979 to refer to the phenomenon that, just as the majority of the ice in an iceberg is under the surface, so, with healthcare, the majority of symptoms are submerged from professional view.¹

A survey of UK general practice patients between the ages of 18 and 60 (i.e. not the age group you might expect to have the most trouble) found that over a 2 week period each person had an average of 3.66 symptoms from a list of 25, with a range of 0 to 22.2 Another survey by the same authors, also among the 18 to 60 age group, found that in about half of cases the symptom sufferer did nothing at all; 35% self-treated (in 70% of cases with an over-the-counter medication) and only 12% consulted a healthcare professional.³ If you think you are busy now, imagine what would happen if all patients presented with all their symptoms, or if there was even a slight shift in patient tolerance. It makes you realise what a precarious footing primary care is on.

So what makes that minority of patients with symptoms seek professional healthcare? Every symptom brings attendant anxiety: how much you worry about a symptom will determine the lengths to which you will go to find out what is happening and to seek a resolution. So every patient has at least two diagnoses – the symptom and the worry it causes.

Cecil Helman, an anthropologist and also a medical doctor working in South Africa, where there is a rich tradition of people getting healthcare from all manner of healers, suggested reasons why patients consult with a healthcare practitioner.⁴ (See Box 1.)

TAKING THE CALL

An initial layer of assessment is done by your reception colleagues. The majority of requests to primary care are made over the phone. Patients are increasingly booking online, and a few traditionalists will stand in the queue and make their appointments over the reception desk. But in most instances first contact is with someone over the phone, which, in most cases, will be a member of the reception team.

There are some symptoms which need immediate hospital care as the chance of serious illness, and illness which may be fatal and need urgent care, is very high. Many years ago it was normal to treat heart attacks in the community, and even when intensive care units became available there was an argument that all the fuss and tubes resulted in a worse prognosis. Now, on the other hand, any general practice which did not have a system for the immediate admission to hospital of people with cardiac-sounding chest pain would be deemed derelict in its duty of care as these days there is so much more that secondary care can contribute.

In practical terms, this means that if a receptionist taking a call becomes aware that a patient is suffering from a number of key symptoms the correct response is not to book a triage phone-call or an appointment, but to ring for an ambulance. Box 2 has suggestions from my own practice. In most cases practice receptionists are not clinically trained and some may be reluctant to take on this responsibility, but they really are the front-line troops and there may well be situations where a window of care opportunity will be missed if there is a delay.

Some patients are reluctant to divulge their symptoms over the phone and some encouraging and explanatory phrases may have to be used. General practice nurses (GPNs) may be called upon to help identify the practice list of ‘alert’ words and phrases, and train reception colleagues in their use.

YOU ARE NOT ALONE

A feature of the current NHS (and a feature not shared by all western healthcare services), is that secondary care is accessed through primary care – you have to see a GP before you can see a consultant. This, of course, is only partially true. Although there are over 340 million consultations in general practice a year, over 20 million patients a year bypass primary care and refer themselves to A&E departments.^5,6

Sending a patient who you think may be seriously ill straight to A&E is only one option. The primary care team is composed of people with differing skills and responsibilities. Should a sticky clinical situation arise, it is always appropriate to pick up a phone or bang on a door for advice. You may (or may not) be surprised to learn that your GP colleagues do not know everything, and they may in turn make a call to the hospital on-call teams to find out what to do next. It may be that the situation can be managed in the community, but an acute hospital admission or an urgent or routine outpatient referral may also turn out to be the best approach.

ACUTE PRESENTATIONS

It is suggested that acute presentations in primary care can usefully be divided into four types.⁷

1. Acute minor illness

2. Acute major illness

3. Acute presentation of existing major illness

4. Acute presentation of new chronic illness.

1. Acute minor illness

This is the stuff that will get better by itself or which needs only minor supportive intervention. Examples here would be coughs and colds, earache, diarrhoea and joint aches and pains. A GPN faced with such presentations will require knowledge of the conditions and their prognosis (‘it should be better in two weeks’), the treatments recommended or available, and situations where it may be a good idea to involve another colleague, such as a GP.

My practice operates a nurse triage phone service, run by GPNs and supported by a series of written clinical protocols. Same-day appointments can only be allocated after a triage contact. In each protocol there is a brief synopsis of the symptoms of the condition, advice that can be given to patients, recommended treatments (prescribed and otherwise), and when a problem should be passed to a colleague (who should see a GP, and with what urgency).

The writing of such a protocol is a learning exercise for the whole practice:

• There has to be a literature review to establish current good practice
• The proposal has to be discussed and agreed by all practice clinicians (this has the additional advantage of making sure patients do not ‘play off’ one discipline against another in the hope of different treatment)
• The proposal has to comply with practice realities (it’s not much good asking a patient to come back in four days to see if they need antibiotics if you can’t give them an appointment in four days).

There is scope for disappointment if a patient does not get what he wants (as opposed to what he needs). This can be ameliorated with good consultation skills which, in particular, attend to patient ideas, concerns and expectations. Only when these are identified can proper explanations and management advice be offered. This may seem a bit excessive when dealing with a minor and self-limiting problem but a problem is never ‘trivial’ to the person who has it. It should always be borne in mind that, in primary care, not intervening always takes more explanation than intervening. Good consultation skills and proper explanations are also good ways of avoiding complaints. Such skills do not always come naturally and investment in learning them is worthwhile for all health care professionals.

2. Acute major illness

These are the conditions which invariably need treatment. Major illness is actually quite uncommon in primary care, and because it is uncommon it is easy to miss. On the other hand all GPNs and all GPs get the majority of their preliminary training in a hospital setting, and so should be very aware of the serious conditions that tend to end up in secondary care.

Table 1 lists some of these conditions.⁷

A key question to ask is: ‘can this be handled in primary care?’ Based on the skills available in the practice, could a colleague be involved (and does he/she have time to be involved)? Are sufficient resources available in the community, and can they be accessed in a timely way? Can the family cope? Or does the condition need secondary care input, and if so how quickly (immediate admission, 2 week referral, routine referral)?

To help in your decision making you will also want to consider and assess the following:

• Is your patient unwell?

Some patients with pneumonia may cope well at home, but others will need acute admission. With experience, most GPNs will develop an instinct for patients who are severely unwell, or who are not responding to treatment as expected. Box 4 is my suggested list of features consistent with an ill patient. You can probably think of several more.

• Vital signs?

These can be useful if you are not sure how unwell your patient is. Unconsciousness, fits and delirium are readily detected from clinical observation. The Royal College of Physicians (endorsed by the Royal College of Nursing) recommends the National Early Warning Score (NEWS) based on six measurements,⁸ albeit these are mainly aimed at secondary care workers:

• Respiratory rate
• Oxygen saturation (a pulse oximeter can be bought for as little as £10 and every practice should have one)
• Temperature
• Systolic blood pressure
• Pulse rate
• Level of consciousness.

Each is allocated a ‘score’ and the total score is used to assess how ill your patient is. For full details please see reference 8. For example, a temperature over 39oC scores 2 points, as does an oxygen saturation of less than 93%. Minor abnormalities on a number of these parameters should be taken as seriously as a substantial abnormality on only one of them. In hospital it is suggested that a score of 4 should be the threshold for more intensive monitoring.⁸ Recommendations are not available for primary care, but I would suggest that a score of 2 or more should prompt a chat with a GP colleague.

• ‘Red Flags’ or alarm signs. This idea is usually applied to a possible diagnosis of cancer (of which more later) but may apply to other situations. Box 2 gives some examples of situations invariably needing urgent care, but not dependant on a confirmed diagnosis.

3. Acute presentation of existing major illness

The records will show if your patient has an ongoing condition that may be prone to acute exacerbations. Examples here would be patients with diabetes with unstable blood glucose, or someone with COPD who has a chest infection. At least it should be clear what is going on, and the need for non-primary care treatment will be determined by how unwell your patient is, and whether or not sufficient community support can be mobilised (which includes how the carers are coping).

A professional without prior knowledge or access to medical records, so without a context, may be tempted to recommend inappropriate care, e.g. an urgent hospital admission for a situation that should be manageable in the community. This is yet another argument for continuity of care and/or sharing medical records.

4. Acute presentation of new chronic illness

All chronic illness starts sometime, and this may well be a particular fear for your patient. Because of general practice’s accessibility, illnesses tend to be seen in primary care at a very early stage, often before it is clear what the diagnosis is. It is suggested that in up to 40% of cases presenting to primary care, a diagnosis is never made.⁷

In other instances a diagnosis will only become apparent as time elapses. Not infrequently patients and politicians assert that patients who become ill should be seen as soon as possible, if not immediately. However, accurate diagnosis depends on the diagnostic tools available and if an illness is seen for assessment too early in its course, the tools available may not work. Accordingly, there is probably an optimum delay before professional care is sought if a proper diagnosis is to be made. Unfortunately nobody seems to know what the length of this optimum delay should be.

IS CANCER DIFFERENT?

Special rules apply to suspected cancer in the NHS. The presenting symptoms of cancer are often rather vague and non-specific. There is a possibility that nearly any symptom might be a sign of cancer, and this possibility will certainly be uppermost in your patient’s mind. Typically cancer does not make patients acutely unwell at first presentation, but, almost invariably, cancer prognosis is better if the cancer is caught and treated at an early stage, and here lies the conumdrum.

The NHS does not do very well by European standards in identifying and treating cancers.⁹ Much of this is attributed to delay in diagnosis,¹⁰ and later stage cancers have a worse prognosis. The ‘gatekeeper’ role of primary care in the NHS is internationally praised for its ability to deliver efficient, economic and equitable care, but may nevertheless not be an ideal model when dealing with cancer.⁷ An unacceptably high proportion of cancers are only diagnosed when patients in the UK attend an A&E department with acute symptoms.¹¹

Part of the delay in diagnosis is attributable to patients who do not present their symptoms, so there may be scope to reduce the delay by increasing public awareness of the symptoms of cancer. However, symptoms – especially in the early stages of cancer (the time when diagnosis and treatment can be of most benefit) – are often non-specific and may mimic mild and self-limiting illness. An example of limited success in improved public knowledge was the ‘Blood in Pee’ campaign for urological cancer, mounted in 2013. In the pilot study, this campaign resulted in a 28% increase in urgent suspected cancer referrals.¹² However, such a programme may not be applicable to all cancers as the development of haematuria is something that is readily visible.

In 2005, NICE produced Referral guidelines for suspected cancer to assist clinicians in making referrals to secondary care. This guidance was supported by the introduction of the ‘2 week rule’ referral pathways for suspected cancer, initially for breast cancer (1999) and later extended to all cancers. Under these rules, all cases of suspected cancer should be seen by an appropriate specialist within 2 weeks of referral.

The 2005 NICE guidance was designed to set a suspicion threshold of 5%: compliance with the referral guidance should result in 5% of referred patients being diagnosed with cancer. An early study of effectiveness showed that about 10% of patients referred under the 2 week rule ended up with a diagnosis of cancer, but also that only about a third of all cancers diagnosed were as a result of 2 week rule referrals.⁷

In 2015 NICE produced revised guidance, setting the bar lower at 3% risk.¹³ Some of these revisions are due to advances in diagnostic techniques, and others indicate the shift in the risk threshold. For example, the 2005 guidance suggested that a shift in bowel habit was only relevant in the diagnosis of colorectal cancer if it was towards diarrhoea, whereas the 2015 guidance regarded any shift in bowel habit (towards diarrhoea or constipation) as important.

Screening programmes in the NHS for cancers of the bowel, breast and cervix are well established. Other screening procedures may be available in the future, but at present these are the only three that are deemed worthwhile (a view not shared by all,¹⁴). If and when better testing is available this situation may alter.

CONCLUSION

By far the majority of illness seen in primary care is minor and self-limiting. Indeed if you completely ignored the possibility of your patient needing treatment or hospital referral, you would still be correct most of the time. However, serious illness does occur, and the results of missing a diagnosis may be fatal.

With experience, most clinicians develop an instinct for when a patient is unwell. If there is doubt, the instinct can be reinforced by a scoring system such as NEWS,8 using readily available clinical measurements. It is always safer to make ten unnecessary referrals to hospital than it is to miss a patient who would benefit. Remember, it is the job of secondary care to support primary care, and not vice-versa.

Rather different rules apply to suspected cancer as the possibilities are numerous and the initial presenting symptoms may be very vague. Luckily there are some particular guidelines and specialised fast-track referral mechanisms for support. Under the current guidance, if you are doing it right less than one in thirty of the patients referred under the ‘2 week rule’ will have cancer.

REFERENCES

1. Hannay DR. The symptom iceberg: a study of community health. Routledge & Kegan Paul, 1979. ISBN 10: 0710089821 / ISBN 13: 9780710089823

2. McAteer A, Elliott AM, Hannaford PC. Ascertaining the size of the symptom iceberg in a UK-wide community-based survey. Br J Gen Pract. 2011 Jan;61(582):e1-11. doi: 10.3399/bjgp11X548910.

3. Elliott AM, McAteer A, Hannaford PC. Revisiting the symptom iceberg in today’s primary care: results from a UK population survey. BMC Family Practice 2011;12:16

https://doi.org/10.1186/1471-2296-12-16

4. Helman C. Culture, Health and Illness. 4th Edition London: Arnold, 2001

5. RCGP. Put Patients First, 2017. putpatientsfirst.rcgp.org.uk

6. King’s Fund. What’s going on in A&E? The key questions answered, 2017.

7. King’s Fund. Managing acute illness, 2010.

8. RCP. National Early Warning Score (NEWS). July 2012 https://www.rcplondon.ac.uk/projects/outputs/national-early-warning-score-news

9. Cancer Research UK. UK cancer survival trails Europe. http://www.cancerresearchuk.org/about-us/cancer-news/news-report/2017-07-19-uk-cancer-survival-trails-europe

10. King’s Fund. How to improve cancer survival, 2011

11. ITV News. Thousands diagnosed with cancer in A&E each year ‘despite repeat GP visits’ http://www.itv.com/news/2017-04-25/thousands-diagnosed-with-cancer-in-a-e-each-year-despite-repeat-gp-visits/

12. Public Health England. Be Clear on Cancer symptom awareness campaigns.

PHE Gateway number: 2014 087 May 2014

13. NICE NG12. Suspected cancer: recognition and referral, 2015 (updated July 2017).

14. Let's Get Checked. Cancer Screening.