Newsweek Article on Colin Walsh's Research in Predicting Suicide Risk

February 28, 2017

From Newsweek:


Machine-Learning Algorithms Can Predict Suicide Risk More Readily than Clinicians

by Matthew Hutson

Each year in the United States, more than 40,000 people die by suicide, and from 1999 to 2014, the suicide rate increased 24 percent. You might think that after generations of theories and data, we would be close to understanding how to prevent self-harm, or at least predict it. But a new study concludes that the science of suicide prediction is dismal, and the established warning signs about as accurate as tea leaves.

There is, however, some hope. New research shows that machine-learning algorithms can dramatically improve our predictive abilities on suicides. In a new survey in the February issue of Psychological Bulletin, researchers looked at 365 studies from the past 50 years that included 3,428 different measurements of risk factors, such as genes, mental illness and abuse.


Colin Walsh, an internist and data scientist at Vanderbilt University Medical Center, along with FSU’s Franklin and Ribeiro, looked at millions of anonymized health records and compared 3,250 clear cases of nonfatal suicide attempts with a random group of patients. To make their prediction method widely scalable, they restricted themselves to factors that would be documented in routine clinical encounters, such as demographics, medications, prior diagnoses and body mass index. Then they let a computer churn through the data and find patterns that would predict suicide attempts within various time frames, from a week to two years.

The accuracy score for each algorithm could range from 0.5 to 1, with 0.5 being no better than chance and 1 being perfect prediction. For comparison, the single factors from the meta-analysis achieved scores of about 0.58, little better than flipping a coin. The computer, however, achieved scores ranging from 0.86, when predicting whether someone would attempt suicide within two years, to 0.92, when looking ahead one week.


(Read the complete article here)

Kevin Johnson on Making Precision Medicine Patient Friendly

December 21, 2016

(From MedCity News)


How to Make Precision Medicine More Patient-Friendly

by Neil Versel


Dr. Kevin Johnson, chief informatics officer at Vanderbilt University Medical Center, has a theory about how to explain precision medicine to laypeople.

“Homeless patients should not get medications that require refrigeration. That is precision medicine,” he said during the opening session of the Connected Health Conference in National Harbor, Maryland.

Another panelist, Gregg Simon, executive director of the White House Cancer Moonshot Task Force, agreed with that philosophy. “We need to avoid being grandiose when we talk about precision medicine,” he said.

(Read the full story at MedCity News)

Jeremy Warner on Precision Cancer Medicine Apps

December 5, 2016

(from GenomeWeb)

Vanderbilt's Precision Cancer Medicine App Brings Genomic Data to Point of Care

by Uduak Grace Thomas

NEW YORK (GenomeWeb) – Enabling precision medicine at the point of care requires ready access to genomic information within the clinical workflow as well as tools to help clinicians make sense of the information presented to them.

As Electronic Health Records' vendors work to develop functionality that will enable the use of genomic data at the point of care, researchers from Vanderbilt University and elsewhere have developed a prototype of a clinico-genomic mobile application that provides some features that clinicians might use in interactions with patients. It also "demonstrates how to achieve end-to-end integration with a data warehouse operating in near-real time with the accompanying EHR system," the researchers wrote in a paper
published earlier this year in the Journal of the American Medical Informatics Association that describes the app.


Jeremy Warner, an assistant professor of medicine and biomedical informatics at the Vanderbilt-Ingram Cancer Center and one of the PCM app's developers, discussed the tool during a presentation at the HL7 Genomics Policy Conference held last month in Washington DC. 


Read the complete story at GenomeWeb.

WebMD Article on Josh Peterson and the Challenges of E-Prescribing

December 3, 2016

(from WebMD)

Hospital E-Prescribing: Trouble for Older Adults?

by Kathleen Doheny

TUESDAY, Nov. 29, 2016 (HealthDay News) -- Preprogrammed doses of medications that can raise the risk of falls are often set too high for older hospital patients, new research shows.

In the study, doctors looked at the records of 287 patients over the age of 65 who fell while staying in a large urban hospital. Some patients fell more than once, adding to a total of 328 falls in the study. Of those falls, 62 percent occurred in patients who had been given at least one high-risk medication in the 24 hours before their fall.

Of that 62 percent, 16 percent had been given two high-risk medicines, while another 16 percent had been given three or more.


However, the research does suggest that dosages matter and that reducing the default dose for vulnerable patients might be an easy way to lower the risk of falls.

Dr. Josh Peterson, an associate professor of medicine and biomedical informatics at Vanderbilt University Medical Center in Nashville, said the new findings build on previous research that suggests the default dosing system sometimes needs to be tweaked.


Read the complete story at WebMD.

Jeremy Warner's Research on Precision Medicine Apps Featured

December 3, 2016

(from Healthcare Informatics)

Health systems focus on data integration, decision support, clinical work flow

by David Raths

When the American Society of Clinical Oncology surveyed its members to ask what they got when they order a cancer gene panel from a lab, approximately 50 percent said the lab faxes over a PDF or sends a PDF file as an attachment. Only 22 percent said the lab can send the results as discrete data and that their EHR can store it.  

The average genomic report is 30 pages long and dense, noted Jeremy Warner, M.D., M.S., assistant professor of medicine and biomedical informatics at Vanderbilt University School of Medicine. Integrating genomics into clinical workflow is a step into unknown territory, he added. Health systems have to make sure the report is not interruptive of the patient-doctor relationship. “I can’t imagine reading a 30-page PDF in front of a patient in the office,” said Warner, who was speaking at an Oct. 26 HL7 meeting on the future of cancer genomics, interoperability and precision patient care.


Read the complete story at Healthcare Informatics.