[Skip to Navigation]
February 27, 2019

Big Data From Health Records in Mental Health Care: Hardly Clairvoyant but Already Useful

Author Affiliations
  • 1Kaiser Permanente Washington Health Research Institute, Seattle, Washington
JAMA Psychiatry. 2019;76(4):349-350. doi:10.1001/jamapsychiatry.2018.4510

The data science wave that has transformed industry and commerce is now reaching mental health care. Algorithms developed from millions of patient health records have been reported to predict suicidal behavior more accurately than traditional clinical assessment.1-3 Computerized rating of clinicians’ notes can uncover the sentiment of clinical encounters and predict subsequent suicide.4 Machine learning can identify symptom patterns predicting response to specific depression treatments,5 a goal unattained through decades of small-data research. Inclusion of genomic data in clinical records will multiply the possibilities for rapid learning from large records databases.