Street Stats: The Bonferroni Correction

    NextPrevious

    Street Stats: The Bonferroni Correction

    Tl;dr: (1) Expected false positive rate for any one statistical test is generally 0.05 (aka, alpha). But this error compounds when you run multiple statistical tests. (2) Adjust your target p-value by applying the Bonferonni correction (0.05/n where n = # statistical to tests) to see if authors’ findings are truly consistent with their reported findings.

     

    Sometimes stats are really boring and awful. Sometimes they can be great and truly deepen your understanding of a paper. Hopefully this discussion on this underutilized stat will fall into the latter.

    The problem of multiple comparisons is common in literature and often unrecognized by authors and readers alike. Every statistical test inherently has a false positive rate; we generally accept alpha = 0.05 as an acceptable false positive rate (hence the obsession with p<0.05). If you run multiple tests, then the generation of a false positive becomes more and more likely. Authors generally work around this issue by proclaiming one primary study outcome and then a litany of secondary outcomes. It isn’t a horrible strategy if you force yourself to only interpret the p-values on the primary study outcome (which we suck at doing). It is a horrible strategy, mainly because the underlining math hasn’t changed at all, when interpreting secondary study outcomes.

    If you want to try to draw more meaningful conclusions from secondary study outcomes, then consider applying the Bonferroni correction. It is a simple correction of dividing your desired false positive rate (generally alpha = 0.05) by the number of outcomes tested (n): 0.05 / n. For example, if the authors ran two different statistical tests with separate outcomes, then the desired p-value should be <0.025 (or 0.05/2).

    Know that the Bonferonni correction is an overly conservative assumption, but it is easy to understand and apply quickly while reviewing a paper. Consider applying it as you try to draw more meaningful conclusions from  secondary analyses in papers.

     

    Source:

    Sedgwick, P.  Multiple hypothesis testing and Bonferroni’s correction. BMJ 2014; 349 doi: https://doi.org/10.1136/bmj.g6284

    • Welcome! This is the website for the Mount Sinai Emergency Ultrasound Division. It serves as an information resource for residents, fellows, medical students and others seeking information about point-of-care ultrasound. There is a lot ofRead more

    • renal handling of water

      If you were on a tea & toast diet, how much water would you need to drink before you develop hyponatremia? I haven’t seen anyone work out the numbers before so here are my calculations. AndRead more

    • acute acidemia physiology

      As alluded to in the first post, don’t be fooled by a “normal” potassium in the setting of DKA because osmotic diuresis and H+/K+ exchange means that total body potassium is actually LOW. You all knowRead more

    • renal handling of potassium

      the first symptom of hyperkalemia is death Earlier post covered temporizing measures to counter hyperkalemia — namely, intracellular shift, increasing cardiac myocyte threshold potential. Give furosemide if the patient still urinates and consider dialysis, but then askRead more

    • bicarbonate revisited

      Previous post reviewed the safety of balanced crystalloids in hyper K. But what was up with serum bicarbonate decreasing with saline administration? This post introduces a new way of looking at the anion gap to possiblyRead more

    • hyperkalemia and balanced crystalloids

      Is it safe to give LR or plasmalyte to a hyperkalemic patient (these balanced crystalloids have 4-5 mEq/L K as opposed to 0 mEq/L K in normal saline)? Postponing the discussion of renal handling of potassium toRead more

    • hyperkalemia physiology

      You’ll likely encounter hyperkalemia on your next Resus / Cardiac shift, and you’ll instinctively treat it. But take a moment to review the fascinating physiology behind the “cocktail”! First, consider how K+ is buffered byRead more

    • Slow down your tachycardia (but not really)

      You’re sitting in resus bemoaning the departure of your most beloved attending when suddenly a patient wheels in without warning. The patient looks relatively stable but the triage RN tells you her heart rate wasRead more

    • Otitis externa: use the ear wick!

      Acute otitis externa (AOE) is a common complaint seen in pediatric as well as adult emergency departments. AOE is typically not accompanied by acute otitis media, although concurrent cases are possible. Also called “swimmer’s ear”Read more

    NextPrevious