JAMA publishes inaccurate study on data breaches

While there have been numerous data breaches in the past few years, it seems a research letter on the topic published in the Journal of the American Medical Association in April overstated the problem.

JAMA’s correction notice acknowledged five mistakes in the text and a dozen in a data table, but the authors said none of them affected the study's conclusions. Retraction Watch, a blog devoted to such retractions, called the situation a "mega-correction" because one result changed from statistically significant to “borderline significant” after the revision.

Lead author Vincent Liu, MD, MS, research scientist at the Kaiser Permanente Northern California Division of Research, told Retraction Watch that JAMA inadvertently published the table with outdated information. “Once we became aware that the older version was published, we corrected the table with the editorial staff. The overall study findings remained consistent,” according to the blog post.

The research letter was a review of 949 breaches of protected health information, as defined by HIPAA, between 2010 and 2013. The breaches affected the records of 29 million patients, not the 29.1 million originally mentioned in the original study published in April.

Read the original study.