TIPs Home

Diagnostic testing for HIV

HIV presents in a myriad of ways. How can a GP respond to the clues?

Retrospective reviews of GP records after new HIV diagnoses show that patients present in primary care with HIV-related conditions, but the significance is often missed.

The difficulty is that many of them are conditions commonly encountered in general practice and commonly viewed as harmless. These conditions may also respond to treatment.

The trusted ‘watch and wait’ approach is therefore not appropriate when HIV is a possible underlying cause. The clinician needs to be able to recognise, and then respond to, conditions associated with HIV.

Key information

Better late than never?

The diagnosis of patients with HIV indicator conditions is virtually always, by definition, late.  But you can avoid it becoming too late by taking a proactive approach to patients presenting with HIV-associated conditions.

As the graph shows, after the earliest weeks of infection, a patient will typically remain asymptomatic for several years. Then, as the number of CD4 cells (T helper cells – a type of white cell) falls, they become prone to the infections, cancers and other HIV indicator conditions.
The significance of late diagnosis of HIV is covered in The case for HIV testing and you may wish to encourage your colleagues to look at this. You could also arrange a clinical meeting and use The case for HIV testing team quiz or the HIV indicator conditions: a quiz for your team (link to quiz), and arrange a discussion. This could be supported by the use of one of our slidesets.

Recognise which conditions are HIV-associated

How did the quiz go for your clinical colleagues?

It is important for GPs to familiarise themselves with those conditions associated with HIV.  For a comprehensive list of HIV indicator conditions see UK National Guidelines for HIV Testing (2008) (Table 1)

The risk of underlying HIV is higher if:

- these conditions are unusually severe (eg multidermatomal shingles)

- they are difficult to treat (severe recurrent vaginal candida or seborrhoeic dermatitis)

- the patient has more than one HIV indicator condition eg CINII and/or a lower respiratory tract infection and/or unexplained thrombocytopaenia (so do review the records of patients presenting with an HIV indicator condition).
The slideset, Diagnostic testing for HIV: The Symptomatic Patient can be used to teach your clinical team about presentations of HIV.

For further information on the clinical presentation of HIV in primary care see the HIV in primary care  booklet.

The e-GP e-learning session (GPS 3.08_21 HIV Indicator Conditions) offers interactive education for GPs and practice nurses registered to use e-GP, and covers the topic in more depth than the slideset.

Strategies to seek supporting clinical evidence of HIV

When a patient presents with an HIV indicator condition, it is worth taking these steps to assess further.

You need to know which conditions are HIV-associated and you need to be able to recognise them.

i Ask the patient if they have weight loss, sweats or diarrhoea

ii Examine the patient’s mouth, skin and nodes

iii Review the medical record for the last 2-3 years looking for additional HIV-associated conditions

For more information see the HIV in Primary Care booklet (link to PDF)

Raising the subject with the symptomatic patient

The prospect of raising the subject of HIV with a patient presenting with, for example, hard to treat seborrhoeic dermatitis, may feel challenging - or you may even feel it is inappropriate. However there is a strong clinical imperative to overcome any barriers we may have to discussing HIV with our patients.
Ideas for verbal strategies with symptomatic patients:

The great majority of patients I see with recurrent vaginal thrush have no particular reason for it. However, rarely, it may indicate diabetes – I would recommend a test for this if you agree? Even more rarely, it might indicate HIV – have you ever had a test for this?… Have you ever wondered if you could be at risk? ….Could I ask you some questions to check?
Occasionally when I see someone with [this rash], it is because their immune system is not working well. One uncommon cause of this can be HIV. Have you ever wondered if you could be at risk of this? Could we talk about that in more depth?
Patient response:

Doctor are you trying to tell me I have HIV?

Range of replies:

No I am not. However it is important that I think of rare causes of conditions as well as common ones.

It is my job not to miss HIV so I do raise this with many of my patients!

HIV is uncommon, but I don’t know if any individual patient might have it - so it is always best to talk it through to consider testing.

HIV is so treatable these days I find myself asking a lot of patients about it – the most damaging HIV is undiagnosed HIV.

Evaluation of risk

a) The process of differential diagnosis

The patient with a single condition

If an otherwise well patient has a single HIV-associated clinical condition (perhaps shingles) they are, on average, unlikely to have HIV. However, if you knew that they had a specific risk for HIV it would make a difference to whether you would offer a test. For example If you found they had had a year long relationship with someone from a country with a very high prevalence of HIV 3 years ago, this would be important information. Perhaps your male patient has had sex with three other men in the course of the last 10 years, sometimes unprotected. Perhaps your female patient from Southern Europe used to inject drugs. Therefore introducing the subject (see above) and conducting a quick risk assessment for HIV strongly supports the process of differential diagnosis. Remember, if the patient then asks for an HIV test always agree to it, whatever the risk assessment revealed.

Think carefully before ever dissuading a patient from having an HIV test.

The patient with more than one HIV-associated condition

If the clinical picture (examination of the patient and review of the history / record) provides other possible markers for HIV, then an HIV test should always be offered whatever the apparent risk. However discussing risk of HIV is a helpful prelude to offering a test as we discuss next. 

b) Good quality communication with the patient

Assessing the risk that a symptomatic patient might have HIV helps them understand that you are being objective rather than being judgmental or jumping to conclusions.

In addition, the process of assessing risk of HIV informs and educates patients about how HIV is and isn’t transmitted. If a patient is found to be at no apparent risk (even if HIV testing is still agreed on) it reinforces sexual health promotion messages if this is acknowledged ‘from what you have told me you have been good at protecting your sexual health’.

Finally, risk assessment helps that minority of patients already fearful that they might have HIV to be open with you and share their concerns.

Offering HIV tests to symptomatic patients

By this stage the offering of the test should be straightforward – a risk assessment has been done and forms a good foundation for informed consent to testing. 

Consider: Are there other blood tests to do that are relevant to this case?

Confirm the patient understands the benefits of testing

Confirm how the patient will obtain the results

Less worrying clinical picture:

‘So we have agreed that although you have no apparent risk for HIV, you would like the test so we can rule it out as a cause?’

More worrying clinical picture:

So as I have said HIV is readily treatable these days, and we have agreed you would be much better off knowing if you have it.’ 

See also Offering tests, giving results

  Primary HIV infection: the diagnostic jackpot for GPs

Primary HIV infection - a flu-like illness – occurs just a few weeks after a patient has been infected with HIV.  I learned about this in a course I went on, and I remember thinking that it would be pretty difficult to pick up. However I was impressed by the advantages of diagnosing HIV at this earliest of all stages, with the benefits to the patient (as well as the potential to stop, or limit, transmission). Because our practice serves an area with a high prevalence of HIV I ran a clinical meeting on the subject. We included all our doctors and nurses because any of the clinical team can be involved in assessing sore throat and flu-like illnesses on the phone or face to face.

We agreed that we would raise the subject of HIV whenever possible in patients with such illnesses, but that we would always try to consider HIV when arranging tests for possible glandular fever.

We discussed how one might raise the subject of HIV when faced with a patient with such an illness and agreed on some phrases. My favourite is:

'There are a number of viruses that may cause a sore throat illness such as yours, and one of the commonest is flu. However a bit less commonly it can be caused by glandular fever. Have you heard of this? Shall we do a test? Even more uncommonly a cause can be HIV. Do you think you could have been at risk of this in the last couple of months? Could I ask you some questions to assess your risk? Would you like a test?'

Just a few months after we had this meeting my registrar saw a man aged about 35 who had a bad sore throat, myalgia and fever. He had been unwell for a week. My registrar managed to raise the subject of HIV and take a history to assess risk. She established her patient had had unprotected sex 2-3 weeks before with another man who he had just met in a club. He was surprised when she mentioned HIV but she helped him understand the value of testing. The HIV test showed him to be HIV antigen positive and antibody negative: ie primary HIV infection was confirmed. Her patient is doing extremely well and is relieved his infection was diagnosed so early.

My registrar got a big boost of confidence – I asked her to present the case to the team. One of our nurses pointed out that the consultation – including the sexual history – might equally well have been conducted on the phone. We reflected that, whatever our fears, our patients will often react reasonably calmly to a discussion about a difficult topic.

Changing practice:
the challenges

There is a clinical need to normalise HIV testing, alongside all our other investigations. However many GPs and practice nurses remain unsure about conducting HIV tests themselves.  

Despite a strong clinical case for testing, barriers still exist.

Click on any of the challenges below to find ways to meet them. Which challenges are relevant to your practice?

You will find other challenges, and ideas for overcoming them, throughout the HIV TIPS website. Or see Overcoming the challenges.

But I wouldn't know what to tell the family members

+ See more

If you diagnose HIV you should refer the patient to the appropriate specialist. The specialist team will address partner notification (contact tracing), including family members.

Have a look at The case for HIV testing to make sure you are aware of the benefits of testing for the patient. 

– See less

I am not very confident about which conditions are associated with HIV, so I think I might miss it

+ See more

GPs should be familiar with the conditions associated with HIV and then have strategies for raising the subject of HIV and assessing the need for a test. For more information on this see Diagnostic testing for HIV.

For a list of conditions associated with HIV infection, click here. You can also find more information in the UK National Guidelines for HIV testing.

– See less

I don't want to scare my patient when their symptoms probably aren't HIV-related

+ See more

At first it does feel daunting to raise the subject of HIV with, for example, a patient who has a lower respiratory tract infection or pneumonia. However most patients understand that doctors need to consider rare and/or serious causes for problems.

If we can’t change our practice in this respect, we risk missing HIV.

For help with tried and tested verbal strategies see Diagnostic testing for HIV which is all about symptomatic patients.

– See less

There isn't much HIV around here

+ See more

You may work in an area with a low or medium prevalence of HIV.

The catch for these areas is that HIV is less likely to be thought of as a cause for symptoms, and so HIV is more likely to be diagnosed at a late – even life-threatening – stage.You can check the prevalence of HIV in your area by visiting Public Health England's Sexual and Reproductive Health Profiles (select the indicator Prevalence of diagnosed HIV infection per 1.000 persons aged 15 to 59). You can also try doing our audit. You'll need to sign up to the TIPS site first. (It only takes a minute!).The section Diagnostic testing for HIV will help you and your colleagues correctly identify the conditions associated with HIV infection.

– See less