Can we use a blood test to predict suicidal behaviour?

Suicide is the second leading cause of death among young people and worldwide there is approximately one person every 40 seconds who takes their own life. For every person who commits suicide, there are many more people who attempt it each year. There is a strong link between a persons mental health and their suicide risk. Schizophrenia, bipolar disorder, and depression are often associated with suicide or suicidal thoughts. The difficult part in prevention is identifying when someone is thinking of committing suicide. Often people will not talk about it and the outwards signs may be hard to miss. As a solution to this problem, a research group in Indianapolis has published research in Nature showing that the combination of a blood test and clinical information may be able to predict the risk that someone with mental illness will commit suicide 92% of the time.

The researchers studied 37 people with documented mental health issues and grouped them into those with or without thoughts of suicide. They then analyzed the gene expression in the blood of these patients and looked for indicators (biomarkers) of suicidal thoughts. The top biomarkers from this analysis were then put to the test on a group of 26 individuals who unfortunately had committed suicide. After this validation step, the biomarkers were put through a series of other validations to further whittle down the list to the genes that were the most predictive of future suicidal thoughts. Out of this list, one gene was able to predict with 72% accuracy the likelihood that anyone with a psychiatric disorder would commit suicide. This gene was SLC4A4 and is important in maintain brain pH. Interestingly enough, brain pH is considered to be an important factor in the development of panic attacks and anxiety disorders. This one gene was even better at predicting suicidal thoughts in people with bipolar disorder (93% accurate).

In order to increase the accuracy of their test, the researchers combined the top biomarkers with an app that asks questions about the general mental health of people. These include questions about their feeling of the hope for the future, history of abuse, history of suicide in family, and any current substance abuse. When the score from the app was combined with the blood test, they were able to predict 92% of the time that someone with a mental health issue would have suicidal thoughts. It was able to predict suicidal thoughts 98% of the time in patients with bipolar disorder. Additionally, the test could predict with 94% certainty how many of these patients with mental health concerns would end up in hospital from a suicide attempt.

The work has a long way to go before this test is used in the general public. It was able to predict only 94% of the time that patients with prior diagnosed mental health issues would contemplate suicide. That means 6% of the time it was wrong.  This is likely to be even worse in the general population where only 16 in every 100,000 people have suicidal thoughts. 6% incorrect would mean 6,000 people out of 100,000 would be diagnosed incorrectly. Still, this tool has the ability to be useful in populations of people most at risk for committing suicide. Additionally, the knowledge gained from the analysis of genes involved in the risk of suicide may shed some light on the underlying cause of depression, bipolar disease, schizophrenia, and other mental disorders. Any test that could be used to spot those at risk for committing suicide would be useful in helping start treatment and preventative measures.

Photo Credit: Suicide by Edouard Manet

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s