A national survey of ob/gyn residents indicates that most do not grasp important statistical information. And practicing physicians don't perform much better.
A national survey of nearly 5,000 ob/gyn residents indicates that most do not fully grasp important statistical information.
The survey presented residents with one multiple choice question and one true/false question. The multiple choice question asked residents to select the positive predictive value of a screening mammogram. Just 26% chose the correct answer. The true/false question asked whether the P value is the probability that the null hypothesis is correct. Just 42% correctly answered ‘no.’
Although 74% of residents got at least one of the questions wrong, just 25% of female respondents and 17% of male respondents deemed their statistical training inadequate.
Think this is just part of residents’ learning curve? A study published in the Journal of Healthcare Quality (online first, January 2012) examined statistical literacy of practicing ob/gyns, and only 49% were able to calculate the positive predictive value of a mammogram. Sixty six percent were able to use sensitivity and specificity to choose a test option.
What impact does this have on patient care?