New system to diagnose bipolar four years before symptoms

doctor supporting patient in clinic
The reduced delay in diagnosing patients could lead to increase medical support. Source: Edwintp (via Pxhere)
Results from a recent study, using machine learning, shows high accuracy predicting bipolar in young adults up to four years before the onset of symptoms.

By Holly Giles | Deputy Editor

Bipolar is a mental health condition that is characterised by severe mood swings including “manic highs and depressive lows” as described by BipolarUK. Despite many people not being aware about the condition, it affects over 3 million people in the UK and equates to 1 in 50 people. 

A large problem with bipolar disorder is getting a diagnosis; many people experience a delay of six years between the onset of symptoms and receiving a diagnosis. This is not only frustrating for the patient but can be clinically important as this delay in support leads to a progression of symptoms, less time between episodes and increased harm to the patient’s mental health. The ability to diagnose bipolar more rapidly would transform the experience of many patients.

This is the hope that has been offered by researchers in Brazil who have been able to predict bipolar disorder four years before symptoms start. This would allow four years of preventative treatment, which could mean a decrease of symptoms when they do appear.

The team looked at 3810 individuals born in Pelatos, Brazil in 1993. They took measurements and interviewed the participants aged 11, 15, 18 and 22, looking at the general physical health as well as their mental health and lifestyle. At the end of the study, when participants were 22, 255 people in the study had been diagnosed with bipolar. 

When looking back retrospectively at the interviews, they were able to highlight several factors which increased the risk an individual would go on to be diagnosed with bipolar. Some of these risk factors included showing suicidal tendencies, anxiety, parental abuse and financial problems, all of which correlate with an increased risk of bipolar. 

During the study the team used the same software used to predict weather to input the data and predict the risk of bipolar. “It was the job of the machine learning to weigh these factors and estimate the risk of developing bipolar disorder” explained Lead Researcher, Francisco Biego Rabelo-da-Ponte, when reflecting on the highlighted risk factors.

When thinking about the future implication of the research, Rabelo-da-Ponte continued:

It’s very difficult and expensive to replicate such a long-lasting study, but what we have found indicates that we need more of these longitudinal studies. We’ve already learned a lot from the study itself, for example if we were to set it up now we would include many more mental health parameters, which we hope would allow us to identify even more psychological benefits. We see too many false positives (indicating someone is at risk when they are not) to rely 100% on this system alone. Nevertheless, this system will allow doctors to see who might be at risk, and the gain of 4 years before diagnosis could make a huge difference to the life of a young person.”

In the future it is hoped that this machine learning will be used to highlight those at risk, to enable them to see a doctor before reaching breaking point. “This may be a new additional tool for the diagnosis of bipolar disorder; this will not replace a doctor’s diagnosis, but may allow them to take preventative measures to slow or avoid the onset of the condition, and so gain 4 years of preventative treatment” explained Rabelo-da-Ponte.

Despite its promising results, the study is of a small cohort and is only in one area so it is yet to be determined if the results can be replicated; an example of this is that there may be different risk factors in different areas. The need for more longitudinal studies is apparent but with their high cost and length, it may be some time before we see the fruits of this labour in practice.


Science and Technology Holly Giles

Add Comment

Click here to post a comment

Your email address will not be published. Required fields are marked *