PEM Source

Your source for all things Pediatric Emergency Medicine


Question: Biostats

(Click the link to comment and to vote – voting not working through email, sorry!)

You design a trial comparing two different techniques for draining skin abscesses: standard I&D with packing vs. loop drainage. Your chosen outcome is the proportion of patients that require a second drainage procedure. You expect to enroll 100 patients in each group, and expect an average of 10% to require a second drainage procedure.

What is the best statistical test to use?
ScholarlyScholarly Activity

pemsou5_wp • April 10, 2018

Previous Post

Next Post


  1. Kelly April 10, 2018 - 7:41 am Reply

    B) Chi-square test
    The data will be categorical (yes or no requiring a second drainage procedure), and you will be comparing proportions. Also, you expect at least 5 patients in each cell of your 2 x 2 table, so you do not need to use the Fisher’s exact test. A paired t test would be used to compare normally distributed means between two study groups that are paired. ANOVA is used for comparing the means of 3 or more groups. Linear regression is for multiple regression analysis for a continuous outcome.

Leave a Reply

Your email address will not be published / Required fields are marked *