This very complex study should become a model for study by epidemiologists, statisticians, and clinicians looking to understand study design and reporting. I found at least 6 separate items for comment but had to settle on just 2.
Source: Sheehan WJ, Mauger DT, Paul IM, et al. Acetaminophen versus ibuprofen in young children with mild persistent asthma. N Engl J Med. 2016;375(7):619-630; doi:10.1056/NEJMoa1515990. See AAP Grand Rounds commentary by Dr. Mike Dubik (subscription required).
I'll also be interested to see if today's (November 8) number of views of this blog is the lowest in history, at least in the US. I hope everyone is planning to or has voted. Maybe with that and with following the election results, there's a bit less time for all of us to pursue our routines like keeping up with the medical literature.
First, and very briefly, this randomized, controlled, double-blind prospective study puts to rest (at least for the near future) concerns that had been raised about acetaminophen increasing the risk of asthma-related complications in children. The study design is elegant, particularly trying to navigate the hurdles of studying a chronic disease population over time. It pairs nicely with a posting from last month, a case control study showing an association of ibuprofen use in childhood respiratory disease with development of empyema. The current study dispels that notion; acetaminophen seemed no more prone to do this than did the comparator drug, ibuprofen.
Evidence pyramid example, by Bud Wiedermann. (If you don't remember what GOBSAT means, Google it!)Prior observational studies in asthma have suggested that acetaminophen use for febrile illnesses might be associated with higher rates of asthma exacerbations. Remember, however, one of the somewhat controversial points in evidence-based medicine, the evidence pyramid. The pyramid is frequently misunderstood (ergo the controversy), but all it says is that some study designs (e.g. randomized controlled trials), if performed well, are more likely to stand up to scrutiny years later than are other design types such as case-control studies. Stated differently, a well-performed case-control study is more likely to be proven incorrect in subsequent studies than is a well-performed randomized controlled trial. Remember, case-control studies can suggest an association, but not cause and effect. Randomized controlled trials are much better to answer causality questions.
A second point I'd like to mention is use of the term "factorially linked trial" in the current study. One way to look at this is performing 2 trials for the price of one; here, the same group of children also were being studied to determine best approaches to asthma treatment. Needless to say, this approach can complicate the picture, and requires careful assessment for bias and confounding factors. I think the authors did a good job of addressing those issues.
As pointed out by the study authors, the accompanying editorial, and Dr. Dubik's commentary, we still don't know whether both acetaminophen and ibuprofen can cause asthma exacerbations; this study didn't include a placebo group. It's not likely we'll see a study using placebo for fever control in young children, so don't hold your breath waiting for that question to be answered.
Medicine has been a nice retreat for me in this long US election process. We have lots of controversies in medicine, leading to much debate, but at the end of the day I think we all try to look at the evidence and come to the best conclusion to match that evidence with our patients' individual needs. Maybe politicians could learn something from evidence-based medicine!
