Monday August 16 2010
The study did not prove that paracetamol doubles asthma risk in teenagers
“Paracetamol doubles asthma risk for teenagers,” reported the Daily Mail. It said that adolescents who use the painkiller once a year can increase their risk of asthma by 50% compared with those who do not.
This newspaper story is based on very early research which on its own is fairly weak evidence for paracetamol increasing the risk of asthma. The study has a number of limitations and its design can only show an association between paracetamol and asthma, and not whether paracetamol caused the asthma. The researchers acknowledge this and call for further randomised controlled trials, a sensible conclusion given the preliminary nature of this study.
Where did the story come from?
The study was carried out by researchers from the Medical Research Institute of New Zealand and other academic and medical institutions from China, Malta and Germany. The researchers say that funding was provided by “numerous sources throughout the world”. The BUPA Foundation is listed as being the main source of funding. The study was published in the peer-reviewed medical journal: American Journal of Respiratory Critical Care.
It is an overstatement on the part of the newspapers to claim that this study provides strong evidence that paracetamol “doubles asthma risk for teenagers”. This is preliminary evidence from a weak study design that may lead to further research.
What kind of research was this?
To aim of this study was to investigate the paracetamol use and the risk of asthma and other allergic disorders in 13- and 14-year-old children from different populations worldwide.
This was a cross-sectional study design using participants from a larger study – the International Study of Asthma and Allergies in Childhood (ISAAC). It involved 322,959 adolescents from research centres across 50 countries completing written and video questionnaires that assessed a number of factors including their asthma symptoms and their use of paracetamol in the previous 12 months. This data was then analysed to see whether there was an association between taking paracetamol and developing asthma.
What did the research involve?
A large number of children were recruited into the International Study of Asthma and Allergies in Childhood (ISAAC) from schools across the world. ISAAC is a cross-sectional study of two different age groups, six- to seven-year-olds and 13- to 14-year-olds. Each group completed two written questionnaires and a video questionnaire. Only the data from the older children are presented in the research paper.
The questionnaires asked about their asthma symptoms (wheezing or whistling in the chest), symptoms of rhinoconjunctivitis (sneezing, runny or blocked nose without associated cold or flu as well as watery, itchy eyes) or eczema (itchy rash). The researchers also collected information on the severity of the asthma. The children were asked whether they had taken paracetamol at least once a month, at least once over the previous 12 months, or never.
The researchers analysed whether there was any association between paracetamol-use and asthma symptoms (or eczema or rhinoconjunctivitis for those who didn’t report wheezing in the past 12 months). They also looked more closely at the link between paracetamol-use and different severities of asthma symptoms (through further questions).
The questionnaires showed participants five scenes of “clinical asthma”, asking them to say whether or not they had ever experienced similar symptoms. The analyses of these results were adjusted for possible confounding factors, namely maternal education, maternal smoking, siblings and current fruit-and-vegetable consumption, as well as gender, region of the world, language and the country’s gross national income. Centres for which the data was less than 70% complete were excluded, leaving a total of 180,887 adolescents for the main analysis.
What were the basic results?
The results showed that using paracetamol in the previous 12 months was associated with an increased risk of currently having asthma symptoms. Compared with those who said they had not used paracetamol in the previous year, those reporting medium use (once or more times a year) were 1.43 times more likely to report asthma symptoms.
Those reporting high use (one or more times a month) were 2.51 times more likely to report asthma symptoms. In those reporting no wheeze in the previous 12 months, use of paracetamol was associated with symptoms of eczema and rhinoconjunctivitis.
How did the researchers interpret the results?
The researchers conclude that taking paracetamol may represent an important risk factor for the development or continuation of asthma, and other disorders such as eczema, in teenagers.
This is a large study, but despite its size it has several limitations that affect its interpretation. On its own it is fairly weak evidence for paracetamol increasing the risk of asthma.
The limitations include the following:
- This is a cross-sectional study, a type of study design that cannot determine causal links between exposures and outcomes as it cannot show which came first, in this case whether paracetamol use preceded symptoms of wheeze.
- This study does not establish whether the children had actually been diagnosed with asthma or whether they were suffering from other causes of wheeze.
- It is likely that in many cases children were taking paracetamol to relieve pain or fever that may be associated with an illness causing wheeze symptoms. The researchers say that “in many countries, acetaminophen (paracetamol) is marketed as the preferred analgesic and antipyretic of choice in persons with asthma”.
- The researchers have adjusted for some factors that may cause asthma, but there are a number of others that were not accounted for, including family history, recent infections, prematurity or low birth weight, or exposure to cigarette/cigar smoke as a child (the study adjusted for current maternal smoking).
- Data was missing from a number of the centres so while the original sample of adolescents available was near to 300,000, more than 100,000 of them were excluded due to missing information. It is unclear how excluding a third of the original sample may have affected the outcomes.
While the researchers cite some studies that support their findings, it is not clear whether these are from a systematic review of studies into the topic. There are likely to be other studies that haven’t found any link. The researchers acknowledge some of the weaknesses of their study and are cautious in their conclusions, saying, “It is not possible in a study of this design to determine whether the positive association observed was causal.” They call for more research using a stronger design, such as a randomised controlled trial.