Category Archives for Statistics
A measure of risk of peripheral nerve injury after total hip arthroplasty
Regional anesthesia is popular for lower extremity procedures. Despite its popularity, however, any procedure has an attendant risk. Fortunately, when we explain a procedure to a patient, we can also tell them about the risk involved. Earlier this year, the … Continue reading
Beginning statistics: Student’s t-test
William Sealy Gosset worked as a brewer at Arthur Guinness & Son in Dublin, Ireland. His second article, “The Probable Error of a Mean,” is the basis for the t-test that we know today. He published the article with his friend, … Continue reading
The Normal Distribution: Part 2
Last month, Page2 considered the normal distribution. As was stated in that particular post, there are tests that can be used to determine whether a data set is normally distributed, and those tests would be considered in a later entry. … Continue reading
Anemia, transfusion, and the risk of acute kidney injury
Anemic patients have a higher risk of postoperative complications than non-anemic patients. Karkouti and colleagues, using a nuanced analytical approach, further explore this seemingly obvious statement in their article, “Influence of Erythrocyte Transfusion on the Risk of Acute Kidney Injury … Continue reading
Paul Meier: a remembrance
Paul Meier earned his doctorate in statistics from Princeton University in 1951, taught at Johns Hopkins from 1952 to 1957, and from there joined the University of Chicago statistics faculty. In 1992, he moved to Columbia, where he was the … Continue reading
When the going gets tough, try bootstrap resampling
In the August issue of Anesthesiology, Graham et al., in their article, “A Cost-Benefit Analysis of the ENIGMA Trial,” used a technique called “bootstrap resampling” (1) in their cost-benefit analysis of the ENIGMA trial. Since nitrous oxide was shown in … Continue reading
The normal distribution
In our basic statistics series, Page2Anesthesiology has considered using mean and median to describe data and blinding. Our next intent is to describe the t-test, though first we need to understand the concept of normal distribution. The t-test is not … Continue reading
Describing data: mean and median
When studies are performed, it`s useful to summarize data. The average value or mean is the sum of the values divided by the number of values. The standard deviation represents the spread around the mean. The smaller the standard deviation, the closer … Continue reading
Basic statistics: Blinding
Most clinical studies approved for publication in Anesthesiology are blinded. Why that is the case may not be clear to all readers. This entry will hopefully clarify the reasons for this practice. When a study is single-blinded, either the observer … Continue reading
Statistics Corner: Propensity Score Analysis Versus Multivariable Modeling: Should They Agree and Can Both Methods Be Biased?
Hansen et al.`s report on a fascinating cohort study investigating the effects of infant inguinal hernia repair and the associated anesthesia (versus a non-surgery age- and sex-matched control group) on 9th grade test scores in Danish children was published in … Continue reading