select search filters
briefings
roundups & rapid reactions
before the headlines
Fiona fox's blog
Reaction to the conference abstract from the European Respiratory Society International Congress that shows evidence of association between early life paracetamol use and decreased lung function in adolescence.
Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:
“This research study looks potentially interesting, but it’s really too early to say whether its findings are important.
“First, this is a conference presentation, and all that is available so far is a brief abstract of what the researcher may say, and a press release. Conference talks are not generally reviewed in full by scientific peers, so that quality check hasn’t yet occurred. Details of exactly how the study was conducted aren’t yet available. The way that the results are dealt with statistically is important in studies of this kind, but I haven’t been able to examine these. Statistical issues do sometimes arise, and I just can’t check. The conference abstract describes results that are different from those described in the press release, presumably because the abstract was written well in advance of the conference before the data were fully investigated, so we really can’t yet be sure exactly what is being presented. It would be much better to wait until the full report on this research is reviewed and published before coming to any conclusions about exactly what it implies.
“But, taking the press release at face value for now, it’s important to bear in mind that a study of this kind, which simply recorded whether children took paracetamol, and later recorded whether they had asthma and tested some of their genes, just can’t show whether any link between paracetamol use and asthma is causal. The press release is, rightly, very careful to point this out. Perhaps, instead of paracetamol use in children causing asthma, something else makes it more likely that an infant is given asthma and independently makes it more likely that he or she will develop asthma. Professor Brusselle, who is quoted in the press release, suggests one possibility, that lung infections in early childhood, caused by viruses, might increase the chance that a child is given paracetamol and might also increase their chance of asthma. The fact that the relationship between paracetamol and asthma was only recorded in children with a particular genetic make-up doesn’t rule out such an explanation at all – a confounding factor such as viral infections may operate differently in people with different genetics. The research also found some evidence of a relationship between paracetamol use in early childhood and lung function when the children were 18, in children with a certain genetic make-up, but again the research provides no basis to conclude that the paracetamol is causing the changes in lung function, and anyway the press release explicitly says that it is not known whether the relationship has any clinical importance.”
Prof Stephen Evans, Professor of Pharmacoepidemiology, London School of Hygiene & Tropical Medicine, said:
“The press release mentioned asthma but the abstract simply described impairment of lung function. This emphasises that the report must not be over-interpreted on the basis of a press release without a validated scientific paper.
“The author says “Importantly, the observed association between paracetamol use in infancy and the increased risk of asthma in adolescents, especially in subjects with dysfunctional GST genetic variants, does not provide proof of a causal relationship. Indeed, the association could be due to confounding factors”.
“We are not sure that the children who used paracetamol are similar to those who didn’t in various characteristics, including the likelihood of developing asthma in the future, when they took the drug.
“Parents who suffer from asthma themselves (perhaps also unknowingly having a genetic predisposition to asthma) may prefer to use paracetamol themselves because other drugs can make asthma worse. This point is not noted by the authors. However, it is equally true that the findings cannot be assumed to be due to other factors. Given the previously-known risks of excess paracetamol, minimising the use of paracetamol is reasonable – though not on the basis of this study alone as it has not yet been through the usual checks and balances of science.”
Dr June Raine, Director of MHRA’s Vigilance and Risk Management of Medicines Division, said:
“Paracetamol is safe and effective for treating pain and fever for a range of conditions when used correctly. People are advised to consult their doctor if their symptoms continue.
“All medicines, including those bought over-the-counter, must be used responsibly and as advised in the information provided with the medicine.
“The safety of all medicines is of paramount importance and under constant review. Our priority is to ensure that the benefits of medicines outweigh the risks. Any new information on the efficacy or safety of medicines will be carefully reviewed and action will be taken if required.”
Prof John Henderson, Paediatric Respiratory Medicine Programme Director, University of Bristol, said:
“Paracetamol, also known as acetaminophen, is a commonly used pain-killer in adults and children that can be purchased over the counter. It is also frequently used by parents for the treatment of fever in children, but its use commonly extends to a variety of common childhood ailments.
“This study adds to a body of literature suggesting an association between frequent use of paracetamol in pregnant women or young children with asthma, one of the commonest chronic respiratory disorders of children worldwide. The association between asthma and paracetamol use was first described in adults and subsequently in the offspring of mothers who took frequent paracetamol during pregnancy, and in children who had frequent paracetamol as infants, in much larger studies than the present one, such as the Children of the 90s study (ALSPAC) in Bristol and the Norwegian Mother and Child Cohort Study; these were population-based birth cohorts that included thousands of children.
“However, because paracetamol can be bought over the counter, it is difficult to monitor an individual’s true intake of paracetamol and most studies, including this one, must rely on what people tell them. These reports may be inaccurate for a variety of reasons, including ability to recall past intake. This was addressed in the present study by a nurse telephoning every four weeks, but could still be influenced by people’s individual reporting and healthcare behaviour. For example, people have different tolerance to symptoms such as pain or feeling unwell, and may recourse to taking medicine at differing thresholds habits. Paracetamol is also commonly used for ill-defined symptoms in infancy, such as teething discomfort. Therefore, frequent paracetamol use compared with infrequent or none, is more likely in people with certain characteristics that might themselves be associated with a greater risk of asthma (confounding), the reason for which the paracetamol is taken could be linked with asthma. For example, a link between migraine and asthma has previously been hypothesised (confounding by indication), or paracetamol could be given to treat symptoms that are the early indications of developing asthma in young children (reverse causation). These are all common problems that limit the use of observational data to draw conclusions about what exposures might cause disease. However, this is an area in which conducting a randomised trial that would produce better evidence is challenging at best; other drugs used to treat pain and fever in children may also be associated with asthma in children.
“The authors of this study have attempted to shore up their association by looking at interactions between some genetic variants in their participants and paracetamol use. A family of genes called glutathione-s-transferase (GST) genes code for substances that help to combat the effects of toxins that can injure body tissue through oxidation; these include products of paracetamol metabolism by the body, which are responsible for its toxic effects, the most well-known of which is liver damage from paracetamol overdose. One of the GSTs, GSTP1, is abundant in the lungs and is part of a system guarding against oxidant injury from a variety of sources. Children with a specific variant of the GSTP1 gene, that has previously been linked with asthma in children, were more likely to have asthma if they had frequent paracetamol in infancy compared with children with the more common variant.
“This finding is interesting. The Bristol-based ALSPAC study, which had shown an association between frequent prenatal paracetamol use and asthma in children, also considered variants of GST genes in both mothers and children and showed some interactions but not with the GSTP1 variant reported here, although the present study focused on direct infant rather than maternal intake.
“These observations do not provide conclusive evidence for a causal effect of paracetamol on asthma risk. It is interesting to speculate whether therapeutic doses of paracetamol, rather than the toxic effects of overdose, could overwhelm a sophisticated antioxidant defence system and lead to pulmonary injury/inflammation that persists as asthma. This study found weak evidence of an association with reduced lung function but we know that reduced lung function in infancy is associated with a higher chance of developing asthma in childhood and with early wheezing illnesses that might themselves be treated with paracetamol.
“The association between paracetamol intake and asthma has been known about for more than 20 years but we appear to be no closer to conclusive evidence on which to base advice to pregnant women and mothers of young children about the risks of asthma associated with frequent paracetamol use. Newer methods of analysing large populations to look for evidence of casual associations of exposures and disease outcomes, such as those being pioneered by George Davey Smith and colleagues in the MRC Integrative Epidemiology Unit at the University of Bristol, may help to substantiate whether or not this association is cause and effect, but it may be that the only solution to this dilemma is to devise a clinical trial, such as that proposed and piloted by a group of New Zealand researchers in 2016.”
Prof Neil Pearce, Professor of Epidemiology and Biostatistics, London School of Hygiene and Tropical Medicine, said:
“The idea that paracetamol use early in life may increase the risk of developing asthma has been around for about 20 years. It has been extraordinarily difficult to prove or disprove.
“The problem is that children are not given paracetamol early in life for no reason. They are often given it because they have respiratory infection. It may be the infection which increases the risk of asthma, not the paracetamol. The picture is further complicated because these children are often also given antibiotics, which is also a possible risk factor for developing asthma.
“Many studies have found a positive association between early life paracetamol use, but many studies have not. Furthermore, the positive associations have often reduced, or disappeared, when the data has been adjusted for other exposures such as infections or antibiotic use.
“It remains a very difficult problem to solve.
“From the limited information available in the press release and abstract it’s hard to judge how good this study is, and it has not been peer-reviewed or published. It is a small study, and it is in a select group (children with at least one family member with asthma or a related disease). Also, the main outcome measure appears to be lung function and not asthma.
“Assuming the data does stand up, this study adds one important finding, namely that the association between early life paracetamol use and subsequent asthma risk occurred in children with a particular genetic variant (the GSTM1 null genotype). This perhaps provides one more piece of evidence in support of the idea that the association between early life paracetamol use and asthma may be causal, but there is a long way to go before we can be more certain about this.”
Prof Alastair Sutcliffe, Professor of General Paediatrics, UCL, said:
“This is the best conference press release I have ever read as it is measured and cautious, very good indeed.
“In the press release the authors report an association, and it was also cautiously reported, but we can’t know whether it was the paracetamol that caused the asthma. It appears plausible on the basis of the known effects of paracetamol, and this study is an attempt to explain the already suggested association between paracetamol consumption and asthma observed in other previous research.
“The abstract however does not report anyassociation between paracetamol and asthma. As I understand it, the authors provided more up-to-date data for the press release after the abstract had been submitted, so the abstract is older. From the very limited information available it’s hard to be able to judge how robust the science is here.
“This is early work and has not yet been through peer review or been published. The study does not identify a causative relationship but inference by association.
“The authors refer to the potential confounding that children with respiratory complaints are more likely to be given paracetamol – this means it’s possible the relationship with asthma is not due to the paracetamol itself but to the condition the paracetamol was treating. But we can’t tell from this study.
“The lead brand of paracetamol for infants (which occupies 80% of the market) is called Calpol. It is heavily advertised and is available over the counter. This study is strong in the sense it was prospective and used gene associations to try tease out why some children appear to be more likely to develop asthma than others after earlier exposure to paracetamol. I congratulate this researcher for what sounds like a well performed study with appropriately cautious outcomes.
“In general, the improper uses of paracetamol are potentially multiple. It is also already well known to be toxic in excess to the liver, kidneys, heart and lungs. Parents sometimes give paracetamol just for a fever, which is understandable but the drug paracetamol is an excellent and safe pain reliever and should be given for that purpose alone. My advice is generally if a child has a low fever, say less than 38.5 degrees C, paracetamol should not be given due to the potential longer terms risks, and if it is given it should be given for short time frames and strictly according to the bottle. There is no advantage to any particular brand so parents could buy the cheapest one if they wanted. Over the years use of paracetamol in infants has gone up despite children generally being weller and weller. Paracetamol treats symptoms rather than the cause of illness.”
Declared interests
Prof Kevin McConway: “I am a member of the SMC advisory committee.”
Prof Stephen Evans: “No conflicts of interest.”
Dr June Raine: “No conflicts of interest to declare.”
Prof John Henderson: “Author of observational studies of prenatal paracetamol and asthma in children, former member of Executive Committee of Avon Longitudinal Study of Parents and Children (ALSPAC; Children of the 90s), staff member University of Bristol and affiliated to MRC Integrative Epidemiology Unit.”
Prof Neil Pearce: “None.”
Prof Alastair Sutcliffe: “I was invited to an advisory board by Johnson and Johnson the makers of Calpol for which I received a fee, but it did not evidently change my views on this subject.”