NEW ORLEANS – Residents in obstetrics and gynecology appear to have sizeable deficiencies in statistical literacy, results of a national survey suggests.
A survey completed by 4,713 ob.gyn. residents in 2011 included two questions aimed at assessing their statistical knowledge. The residents’ performance in this area, overall, "was not good," according to Britta L. Anderson, Ph.D. who works in the research department of the American College of Obstetricians and Gynecologists, Washington.
Indeed, on a multiple-choice question asking residents to identify the positive predictive value of a positive screening mammogram, a mere 26% picked the correct answer.
The other question was true/false. It asked whether the P value is the probability that the null hypothesis is correct. Only 42% of ob.gyn. residents correctly responded ‘no.’
These were not trick questions. A patient’s ability to make informed medical decisions depends upon her physician’s ability to effectively communicate numerical information. And based upon the high wrong-answer rates to the two survey questions on statistics, most ob.gyns.-in-training are on shaky ground in this area, according to Dr. Anderson. She noted that statistical literacy is going to be increasingly important to clinicians as evidence-based medicine moves to the fore.
The survey also asked ob.gyn. residents if they thought their statistical literacy training was adequate, and how they obtained it.
In all, 81% of survey respondents were women, 25% of whom deemed their statistical literacy training inadequate. In contrast, only 18% of male ob.gyn. residents characterized their training in this area as inadequate.
As to where they received their training in statistical literacy, 42% of residents indicated it came solely through participation in journal club, 12% said their training came entirely through course work, 22% indicated their training was solely informal, and 15% said they had received no training at all in statistical literacy. The remaining residents said their familiarity with statistics came from various combinations of journal club, course work, and informal sources.
Dr. Anderson reported having no financial conflicts.