Paediatrics is particularly prone to the pitfalls of overdiagnosis and overtreatment. Although this is a problem, the reasons for overdiagnosis are actually good ones:
- The tendency for symptoms to resolve with time (e.g. vomiting and crying in infancy)
- In childhood there is usually an illness that can’t be treated which mimics those that should be taken more seriously (e.g. viral urticaria leading to overdiagnosis of antibiotic allergy)
When there are no good tests available to tell between two possibilities, we sometimes give a therapeutic trial to help answer the question. That is a strategy which will lead to misdiagnosis if symptoms improve despite our treatment rather than because of it.
With therapeutic trials, it is often best to challenge the assumption that it was the treatment that worked. The two best examples that I can think of are childhood asthma and cow’s milk protein allergy in infants.
Let me give you a case to illustrate what I mean:
A 3 month old has been treated unsuccessfully for symptoms of gastro-oesophageal reflux disease (GORD). A clinician suspects non-IgE Cow’s Milk Protein Allergy (CMPA) because first and second line treatment for GORD has been unsuccessful and because they notice that the baby has quite significant eczema. (Click here to see a guide to diagnosing feeding problems in this age group) The clinician decides to trial an extensively hydrolysed feed. Over the next few weeks, the child’s symptoms of being unsettled and bringing back feeds improve considerably. The eczema is responding to topical treatment.
In this situation, it is easy to assume that the change of milk was what made the difference. Often, this is simply confirmation bias. Colic, reflux and other symptoms of infancy have a tendency to self-resolve. Of course the treatment may have been what worked but at this point in time, we genuinely have no idea.
This is the time to stop the hydrolysed formula and reintroduce a standard formula. (Only do this for Non-IgE CMPA. IgE CMPA is the kind that has urticaria and wheeze etc. The children with this type of allergy need to be referred to an allergoligist.) If the original symptoms of being unsettled and vomiting lots return in the next couple of weeks, the diagnosis is now more robust. If the child remains well despite a return to standard formula, you have undiagnosed a thing. Marvellous.
The second clinical scenario is the 7 year old with a nuisance cough. The cough has been there for somewhere around 2-3 months. There are no associated symptoms such as wheeze or altered exercise tolerance, but the cough is waking the family up at night. The chest is clear on examination.
So, what is the likely diagnosis? Surprisingly, in research land, coughs like this turn out to be caused by pertussis infection more often than asthma or reflux disease. (1,2) It seems that although the pertussis vaccination is successful, infection is still relatively common. Instead of causing a more significant respiratory illness, what we see in vaccinated children is often just the cough that lasts 100 days. There are other, similarly benign reasons for chronic cough in children. Also, there are plenty of significant pathological causes of chronic cough that are not asthma.
Diagnosing ‘cough variant asthma’ is possibly the biggest reason for the current debate about overdiagnosis of asthma in children, fuelled by an article in the BJGP earlier this year. (3) Many children in the UK are prescribed inhaled steroids for chronic cough symptoms. If they get better, this is taken as evidence that they had asthma, but there are other possible reasons for this resolution of symptoms. The evidence suggests that the most likely thing is that the cough has resolved with time rather than with treatment.
This is therefore another opportunity to undiagnose a thing. As well as stopping inhaled steroids after (Snelson makes up a number quickly…) three months it is probably a good idea to get some sort of objective assessment before, during and after the therapeutic trial. Peak flows are great if you can get the child to do these well. In many cases a symptom score (4) is more achievable. If the only complaint was cough, then a symptom diary is all that is required.
If when you stop the steroids, the child’s cough is still resolved, you have a winner. Your New Year's resolution is fulfilled. Of course, once you start, undiagnosing an become a bit addictive. If you find it becomes a problem, why not join a gym instead?
Edward Snelson
Diagnosectomist
@sailordoctor
Disclaimer: My New Year's resolution is to get a better disclaimer.
- Marchmont et al, Evaluation and Outcome of Young Children With Chronic Cough, Chest Journal, May 2006, Vol 129, No. 5
- Wang et al, Whooping cough in school age children presenting with persistent cough in UK primary care after introduction of the preschool pertussis booster vaccination: prospective cohort study, BMJ, 2014;348:g3668
- Looijmans-van den Akker et Al, Overdiagnosis of asthma in children in primary care: a retrospective analysis, BJGP, 1 March 2016
- Asthma.com, Child Asthma Control Test
