Theoretical Effect Size of Limiting Acetaminophen Exposure
Some of the large epidemiological studies, particularly the ISAAC studies, have calculated the population attributable risk for asthma symptoms associated with current acetaminophen exposure, being 41% in adolescents and 30% in children aged 6–7 years. These figures represent approximately the percentage of the prevalence of asthma symptoms that could be attributable to acetaminophen exposure.
To make it clearer; simpler and more straightforward calculations can be done, from a simple-minded practicing pediatrician's point of view, of the putative effect of avoiding acetaminophen exposure in the general pediatric population on wheezing prevalence, taking into account an OR of 1.5 against acetaminophen exposure, a wheezing prevalence of 15%, and a current exposure to acetaminophen rate of 50%. This simple calculation can show that the prevalence in the exposed versus nonexposed groups must be 17.58 versus 12.42%, respectively to comply with the aforementioned epidemiological data (OR), leaving a decrease of 5.16 percentage points as a maximum room for improvement with a population-based intervention to limit acetaminophen exposure. This means 2.58 cases of childhood asthma avoided per 100 children of the intervened population ([17.58–12.42].0.5) or a number needed to treat of 38 to avoid the development of a single childhood wheezing disorder case. These theoretical calculations depend on asthma prevalence and acetaminophen exposure rate in the particular population studied, being more favorable with higher exposure rates and asthma prevalence. For instance, with a prevalence of 30% and an exposure rate of 90%, the room for improvement would be from 30.79 to 22.77%, that is, 7.2 percentage points, 7.1 cases per 100 children and a number needed to treat of 14.
These figures do not seem to be very impressive at the individual level, but at the population level offer a great opportunity to decrease the burden of pediatric wheezing disorders and the only one, with tobacco smoke exposure, that has been clearly identified to date. It also shows how large appropriate clinical trials need to be to have enough statistical power to be able to detect such an effect size.