GIF of chest x-ray

Making a clinical diagnosis of whooping cough

Tips for health professionals

There are two ways. Hearing a good going paroxysm, and taking the right history. Not many people nowadays know what whooping cough sounds like. Once you have heard it you will understand how easy it is to recognise. It is a tune you remember, just like you recognise the National Anthem. If you don’t know what it sounds like or want to hear it again it is all on the symptoms page. Only about 50% get any whooping. For making a clinical diagnosis of whooping cough, whooping is irrelevant.

However, you almost never hear it in the patients you diagnose with it because they never (almost never) cough when you are there. This is the key diagnostic point. It is the hours of no coughing, broken by occasional paroxysms that wrack the body and make the patient feel and look as if they are choking that are virtually pathognomonic of whooping cough.

The lungs are clear in uncomplicated whooping cough.

The diagnosis is made retrospectively, and the more retro the better. If you suspect whooping cough, simply see the patient again in 3 weeks (clinical circumstances permitting of course). They will be essentially the same then and for another 3 weeks too, in all probability!

In practice, a diagnosis is made by taking the right history. But as usual, you can’t take the right history unless you suspect the diagnosis in the first place! Later I will describe the clues that are given in what patients are likely to spontaneously offer.Let me try to describe a paroxysm in words. It may help you find the right questions to ask.

The cough usually takes you by surprise. 

The patient gets a tickling sensation in the trachea and starts to cough a succession of expirations that are so intense  he/she is unable to stop to take any air in. So they keep on coughing until there is no air left to expel. By this time their chest feels as if it is being crushed, the face is congested, salivating, eyes watering and then the nausea starts so there is retching or vomiting. The patient is desperate to take a breath in but there is something stopping it. They have lost the ability. They feel panicky. People around them are staring, not knowing what to do. Then suddenly the muscles relax and they take a big lungfull of air, there is a stridulous sound and those around  relax. But the patient doesn’t because the whole cycle then repeats itself a couple of times. Then they relax because they know it is over for a good few hours and  can get on with their life.

By the time you are in a position to make a clinical diagnosis the patient will feel fine apart from the coughing. This helps to distinguish it from other respiratory infections.

Most patients with it will be over 10 years old so can give a history for themselves. Parents and other witnesses can usually give a better description of the paroxysms. Patients themselves are likely to understate it because by the time they see you they have realised they are not actually going to die in one of these paroxysms, although they did think so initially.

You are very unlikely to see it in a baby before immunization is complete, but it happens and if missed the consequences can be serious. The whole point of the immunization program is to protect young infants by limiting their exposure. If they are exposed they can by protected by erythromycin. But if young infants get it, the disease can be quite different. Coughing rapidly exhausts the baby, so the cough lessens and the apnoea increases, sometimes with cerebrotoxic phenomena in addition, so weakness, anoxia and fits may be the order of the day. It is important to be aware of this because the horror scenario is a health care worker with pertussis in a maternity or neonatal unit. It does happen, and with it being more common in adults, will happen again.

So what makes you suspect whooping cough from the initial account from the patient?

In a recent study in the Netherlands 57% of patients with whooping cough had suggested the diagnosis themselves. They are likely to apologise for troubling you because although they have had a bad cough for a long time, and they don’t feel ill.
They are likely to have been sent by a partner or parent because they are more worried than the patient. Workmates can also complain that they should see a doctor.
Somebody may have said they think they could have whooping cough. These patients usually get laughed at by their doctor.
They are usually infrequent attenders, because it affects average healthy people.
It will have been going on for perhaps 4 to 6 weeks without getting better.
They talk about the cough suddenly catching them unexpectedly and getting an attack of coughing for no reason.
They are likely to explain the severity by the fact that they retch or vomit at the end.
They express fear that the cough may do them some damage (severity only implied).
They have “never had a cough like this before”. This is highly relevant in an infrequent attender.
“I have asthma but this cough is quite different and my asthma is OK right now”. (Asthmatics are more susceptible to whooping cough, but when they get it their asthma seems less troublesome for months afterwards. The former is evidence based, the latter is my anecdotal observation).

What questions can be asked that help to diagnose it?

Is it a choking cough?
Do you go for hours without any cough at all? (“Yes” strengthens possibility)
Does coughing make you feel sick? (A very positive response is usual in whooping cough).
Have you ever had a cough like this before? (“No” strengthens possibility)
Do you know anybody else with a similar cough? This only makes sense to somebody with whooping cough. Often the answer is “yes” because it is usually caught from somebody you are in close contact with, and they say so-and-so had (past tense) the same cough.
Has anybody suggested it could be whooping cough? (“Yes” is surprisingly frequent).
Does it frighten people who who see you cough? (In whooping cough the usual answer is yes. The patient will usually rush to somewhere unobserved to avoid other people’s unwelcome comments. This is probably why adults (as opposed to children) rarely seem to pass it on outside the family.

Working definition

I have used the same one for 30 years and have no reason to change it. An almost exclusively paroxysmal cough for a minimum of 3 weeks. Here are the caveats!
‘Paroxysm’ means the sort of paroxysm you get with whooping cough. Although this definition is circular, it is meaningful, because until you have heard whooping cough you may call other severe bouts of coughing paroxysmal. But we are not discussing semantics here, we are talking about recognising whooping cough.
‘Almost exclusive’ is important because the pattern is never pure. Whooping cough starts as a dry cough and a sore throat usually, and becomes paroxysmal over a couple of weeks commonly. In this phase tickly coughing without paroxysms is usual. So by the time the cough has been paroxysmal for 3 weeks the whole thing has probably been going for 5. Also, whooping cough frequently seems to be caught while you are suffering a viral respiratory infection, so this can complicate the initial history coughwise. Secondary infection is also quite common and may produce an added cough that is not paroxysmal. But the underlying paroxysm pattern remains.

Bits and pieces

The number of paroxysms depends on duration and severity. The more severe cases will say they get 20 a day at its worst. It may remain that frequent for at least 2 weeks then start getting less frequent until by 8 to 12 weeks there are just 1 or 2 a day. Occasionally the number of paroxysms can be 50 a day.
Some mild cases just have perhaps 3 paroxysms a day. Many such cases probably never seek medical advice.
They are as frequent by day as by night.
The severity of each paroxysm seems to remain much the same.
There is no evidence that a carrier state exists.
Subclinical cases are possibly quite common but there is no evidence that they are significant transmitters.
In my experience (744 cases in 40 years) there are very few cases where the diagnosis is equivocal. (But only with long hindsight. At the time you are trying to decide if it is or it isn’t, when they first present, most are equivocal!). To be sure of the diagnosis, much follow up is required. Then there is the blood test to confirm it.

Other causes of paroxysmal coughing

The same sort of paroxysms occur in many respiratory illnesses. I have certainly made a clinical diagnosis of whooping cough when it has turned out to be Mycoplasma pneumoniae, Pneumocystis carinii and lung cancer. Many have no identifiable cause and can go on for years. Sometimes oesophageal reflux can do it and certainly asthma can. So can presumed viral respiratory infections. Paroxysms are not special. Exclusively paroxysmal coughing with temporary paralysis of inspiration generally is.


This page has been reviewed and updated on 20 March 2023