Online Predictive Tools for Intervention in Mental Illness

Mental disease is one of the greatest personal, societal and economic problems of the modern world. Millions suffer in silence; families and friendships suffer, and health services struggle to keep up. Mental disease is already responsible for over a third of total healthcare costs in the EU countries and this figure is growing. The costs in terms of lost productivity is at least as high as the direct cost of care.

The most common of all mental disorders is Depression. Depressive disorders cause immense suffering for the individuals concerned and for those close to them. They also have immense costs in terms of lost working days and in many cases lead to suicide. The prevention of depression and suicide is thus one of the five priority themes in the European Pact for Mental Health and Well Being. It is also the focus of OPTIMI.

Currently the main treatments for depression are drugs and evidence based Cognitive Behavioral Therapy (CBT). Clinicians have few options for prevention; the only available diagnostic tools for early diagnosis or for measuring the effectiveness of interventions, are standardized inventories. OPTIMI aims to improve this situation.

Depression is often associated with poor coping behavior in the face of stress. Some individuals are extremely resilient but others find it difficult to cope. Based on these premises, OPTIMI has set itself two goals: first, the development of new tools to monitor coping behavior in individuals exposed to high levels of stress; second, the development of online interventions to improve this behavior and reduce the incidence of depression.

To achieve its first goal, OPTIMI will develop technology-based tools to monitor the physiological state and the cognitive, motor and verbal behavior of high risk individuals over an extended period of time and to detect changes associated with stress, poor coping and depression.  A series of “calibration trials” will allow the project will test a broad range of technologies.  These will include wearable EEG and ECG sensors to detect subjects’ physiological and cognitive state, accelerometers to characterize their physical activity, and voice analysis to detect signs of depression. These automated measurements will be complemented with electronic diaries, in which subjects report their own behaviors and the stressful situations to which they are exposed. All participants will be regularly assessed by a psychologist, who will use standardized instruments to detect stress, poor coping and depression.  A few will also be asked to wear implanted devices levels of corticol in the blood, an objective physiological correlate of stress. The project will use machine learning to identify patterns in the behavioral and physiological data that predict the findings from the psychologist and the corticol measurements.  The final OPTIMI monitoring system will consist of the subset of tools that proves useful for this purpose and acceptable to users.