B3518 - Using a machine learning approach to develop and validate a prediction model for the onset of hypomania - 12/05/2020

B number: 
B3518
Principal applicant name: 
Steven Marwaha | University of Birmingham
Co-applicants: 
Dr Pavan Mallikarjun, Dr Sam Leighton, Miss Danielle Hett, Professor Daniel Smith, Mr Joey Ward
Title of project: 
Using a machine learning approach to develop and validate a prediction model for the onset of hypomania
Proposal summary: 

Bipolar disorder (BD) is a debilitating mental health condition, characterised by severe shifts in mood, that can range from disabling highs (i.e., mania/hypomania) to extreme lows (i.e., depression). Approximately 1% of the population are affected by bipolar (Pini et al., 2005), with most people experiencing the onset of mood symptoms prior to their 20s (Geoffroy et al., 2013). Despite this, little is known about the predictors to bipolar disorder and hypomania symptoms, particularly among young people. Intervening early in the development of bipolar is a top clinical priority, and one that may have the potential to limit its functional and symptomatic impact on those affected. Thus, predicting the onset of bipolar/hypomania prior to its onset, may help clinicians/researchers to develop novel, tailored preventative strategies and interventions for young people.

Impact of research: 
These results will have direct clinical implications for young people who wish to know their risk of developing hypomania, a strong signal of bipolar disorder. Currently, due to the overlap in symptoms, people presenting with bipolar symptoms are often being managed within early intervention services developed for the treatment of psychosis. This could be preventing young people presenting with bipolar symptoms from receiving adequate diagnosis and appropriate treatments early on in their care. Indeed, research suggests that, compared to out-patient treatment, early intervention may be a more clinically and cost-effective tool in managing bipolar. Thus, developing a prediction model for the onset of bipolar, novel preventative/ early intervention programmes— that are tailored to young people specifically—can be developed. This research remains a key clinical priority and will ultimately help towards reducing the psychological suffering of those with bipolar disorder. A prediction model could also enable clinicians to avoid potentially harmful treatments for people at heightened risk of bipolar, such as antidepressants which are known to trigger hypomania in people who are vulnerable to the condition.
Date proposal received: 
Monday, 27 April, 2020
Date proposal approved: 
Friday, 1 May, 2020
Keywords: 
Mental health - Psychology, Psychiatry, Cognition, Mental health, Statistical methods, Statistical methods