Mining electronic medical records could help depression patients find the right treatment

December 7, 2017, American College of Neuropsychopharmacology
Credit: CC0 Public Domain

Personalized medicine has been one of the most promising medical developments in recent years. To personalize treatment for patients, some doctors have turned to predictive models to help determine which patients will benefit from different treatments.

A new study from Massachusetts General Hospital and Harvard University finds that using data from in those models could hold a lot of promise.

Researchers MGH's Roy Perlis and Harvard's Finale Doshi-Velez, joined by postdoctoral researcher Michael Hughes, looked over large amounts of electronic medical record data from more than 50,000 people with , a condition that is notoriously difficult to treat. Patients often try several different treatments before arriving at an effective one.

The researchers applied machine learning to sort the data into depression subtypes, some of which weren't necessarily standardized. The subtypes represented clusters of symptoms and features spotlighted by the data—such as age, sexual dysfunction, weight gain, or obesity.

Doshi-Velez's team also looked at 16 types of antidepressants that were ultimately successful in treating the patients' depression. From these data, they were able to develop an algorithm to predict an antidepressant's effectiveness based on depression subtypes.

When implemented in clinical settings, the new tool was more accurate than existing methods, the researchers found. And clinicians polled by the researchers said they found the tool more user-friendly than existing .

As there is a 50 percent failure rate among patients trying a first for depression, and a 50 percent failure rate among those who try a second type of treatment, cutting down on the guesswork involved could be life-changing. "Each try can take multiple weeks to assess, while you're playing around with dosages. And these are already people for whom morale is sapped," says Doshi-Velez. "Even if we could make a modest improvement—for instance, one less try—this could be a very significant result."

While the researchers studied the impact of mining records on depression, their method could apply more broadly and help with various health problems, mental and physical.

Explore further: Depressed with a chronic disease? Consider alternative therapies

Related Stories

Depressed with a chronic disease? Consider alternative therapies

November 6, 2017
Scientists are finding more evidence that commonly prescribed antidepressants aren't effective in people battling both depression and a chronic medical disease, raising a critical question of whether doctors should enact ...

Data on depression suggest possibility of personalized treatments

February 23, 2017
Depression is a complex disorder, characterized by clusters of symptoms that respond to treatment with varying degrees of success and are likely to reoccur after initial remission. However, a group of Yale researchers, combing ...

Antidepressant does not reduce hospitalization, death for HF patients with depression

June 28, 2016
In a study appearing in the June 28 issue of JAMA, Christiane E. Angermann, M.D., of University Hospital Wurzburg, Germany, and colleagues examined whether 24 months of treatment with the antidepressant escitalopram would ...

Personalized psychiatry matches therapy to specific patients with depression

May 1, 2017
Selecting the antidepressant that will be most effective for a specific patient suffering from depression can be a "try and try again" process. Examining new personalized and precision psychiatry approaches, a new study in ...

Neuroimaging categorizes four depression subtypes

December 6, 2016
Patients with depression can be categorized into four unique subtypes defined by distinct patterns of abnormal connectivity in the brain, according to new research from Weill Cornell Medicine.

Study finds no increased risk of autism, ADHD with prenatal antidepressant exposure

January 5, 2016
- An analysis of medical records data from three Massachusetts health care systems finds no evidence that prenatal exposure to antidepressants increases the risk for autism and related disorders or for attention-deficit hyperactivity ...

Recommended for you

Infants are more likely to learn when with a peer

October 16, 2018
Infants are more likely to learn from on-screen instruction when paired with another infant as opposed to viewing the lesson alone, according to a new study.

Researchers use brain cells in a dish to study genetic origins of schizophrenia

October 16, 2018
A study in Biological Psychiatry has established a new analytical method for investigating the complex genetic origins of mental illnesses using brain cells that are grown in a dish from human embryonic stem cells. Researchers ...

Linguistic red flags from Facebook posts can predict future depression diagnoses

October 15, 2018
In any given year, depression affects more than 6 percent of the adult population in the United States—some 16 million people—but fewer than half receive the treatment they need. What if an algorithm could scan social ...

Early changes to synapse gene regulation may cause Alzheimer's disease

October 15, 2018
Alzheimer's disease (AD) is the most common form of dementia, involving memory loss and a reduction in cognitive abilities. Patients with AD develop multiple abnormal protein structures in their brains that are thought to ...

Clues that suggest people are lying may be deceptive, study shows

October 12, 2018
The verbal and physical signs of lying are harder to detect than people believe, a study suggests.

Why don't we understand statistics? Fixed mindsets may be to blame

October 12, 2018
Unfavorable methods of teaching statistics in schools and universities may be to blame for people ignoring simple solutions to statistical problems, making them hard to solve. This can have serious consequences when applied ...

0 comments

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.