In the Original Investigation titled “Preventability and Causes of Readmissions in a National Cohort of General Medicine Patients” published in the April 2016 issue of JAMA Internal Medicine,1 4 incorrect P values appeared in Table 1. In that table, the P values (last column) compared data for nonpreventable vs preventable readmissions. The value for where a patient was admitted from, which was given as .77, should be replaced with .75; the value for whether a patient was screened for low health literacy at the index admission, which was given as .009, should be replaced with .002; the value for whether a patient received a postdischarge follow-up telephone call within 72 hours after the index admission, which was given as .006, should be replaced with .98; and the value for the patient’s discharge location, which was given as .33, should be replaced with .48. This article was corrected online.
Incorrect P Values in a Table. JAMA Intern Med. 2016;176(10):1579. doi:10.1001/jamainternmed.2016.4948