PEM Source

Your source for all things Pediatric Emergency Medicine

1

Question: Study design

You are conducting a study to compare the efficacy of a new bronchodilator against standard albuterol therapy in patients with acute asthma exacerbations. To reduce the possibility of selection bias in your study, the key element in your study design is:

A. Blinding study participants so that they do not know which treatment has been selected for them, and do not change their subjective assessment of improvement in asthma symptoms

B. Enrolling sufficient numbers of study participants to ensure an accurate estimate of the difference in treatment effects

C. Randomizing study participants to ensure that the two groups studied are equivalent in potential confounding factors

D. Only enrolling study participants > 2 years old, to avoid selecting bronchiolitis patients instead of asthma patients

Also, if you’re interested in the Peds ID Antibiotics Question of the Week, you can find it here

ScholarlyScholarly Activity

pemsou5_wp • June 20, 2017


Previous Post

Next Post

Comments

  1. Kelly June 27, 2017 - 8:16 am Reply

    C. Randomizing study participants
    Selection bias is selecting treatment groups in such a way that they are fundamentally different (eg subconsciously choosing sicker patients for the albuterol group), making the study comparison biased toward one or the other treatment. The best way to reduce selection bias is through randomization such that investigators or clinicians or patients are completely unable to influence to which study group a patient is assigned. Blinding study participants helps reduce the chance that their knowledge of what treatment they are receiving affects their reported outcome (ascertainment bias, or if patient is reporting an outcome, reporting bias). Enrolling sufficient numbers to ensure accurate estimates gives a study adequate power to draw conclusions. Enrolling bronchiolitis patients into an asthma study would be misclassification bias.

Leave a Reply to Kelly Cancel reply

Your email address will not be published / Required fields are marked *