MONARCA will develop and validate solutions for multi-parametric, long-term monitoring of behavioural and physiological information relevant to bipolar disorder. It will combine those solutions with an appropriate platform and a set of services into an innovative system for management, treatment and self-treatment of the disorder.

The MONARCA system will be designed to comply with all relevant security, privacy and medical regulations, will pay close attention to interoperability with existing medical information systems, will be integrated into relevant medical workflows, and will be evaluated in a statistically significant manner in clinical trials.

Policy Context

The project is funded by the European Commission’s Seventh Framework Programme (FP7), Subprogramme Area: Personal Health Systems.

Description of target users and groups

MONARCA conducted trials in Copenhagen and Tirol. For the first case, the trial included patients suffering from bipolar disorder that had experience of using smart phones, and it lasted for 6 months. For the Tirol trial, it was an uncontrolled, not randomised, mono-centric, prolective, observational study. The total sample size is estimated at 10-15 participants.

The population consists of patients with diagnosed bipolar disorder, categorised by the ICD-10 classification (class F31). Possible participants were patients of all psychiatric facilities of the TILAK.

The period of measurement was 12 weeks per patient.

Inclusion Criteria:

The relevant population comprises people who meet the following criteria:

• Age  >=18 and  <=65
• Ability and willingness to deal with the latest smartphones
• Contractually capable
• Mentally ill by bipolar disorder categorised by ICD-10, F31
• Frequently changing episodes

The participation of the study was voluntarily and quitting it would not affect the therapy in any way.

Description of the way to implement the initiative

The MONARCA system will consists of 5 components:

• a sensor-enabled mobile phone;
• a wrist-worn activity monitor;
• a novel “sock integrated” physiological (GSR, pulse) sensor;
• a stationary EEG system for periodic measurements;
• a home gateway.

It will combine GPS location traces, physical motion information, and recognition of complex activities (nutrition habits, household activity, amount and quality of sleep) into a continuously updated behavioural profile.

Figure 1

Physiological information from the “GSR sock”, the periodic EEG measurements, voice analysis from mobile phone conversations, and motion analysis will provide an assessment of emotional state and mood. Combining this information with patients’ medical records and established psychiatric knowledge quantitative assessment of the patient’s condition (expressed in Psychiatric Rating Scales like BRAM or HAMD) and prediction of depressive and manic episodes will be implemented.

Closing the loop between the system and the patient an interface for self-assessment (on the basis of the above information), provision of warnings and risk profiles and a coaching concept for self-treatment will be implemented. For the medical staff, interfaces for interpreting the data, therapy assessment and therapy planning tools (scheduling visits, planning medication) will be developed.

The aim of MONARCA aim is to develop and validate an approach which is both:

1. multi-parametric,
2. closed-loop.

Multi-parametric - Quantitative analysis of every-day life activities and state

• Location/travel patterns:
• Work, home, shopping, exercising
• General activity level analysis:
• Walking, resting, sitting, lying, etc.
• Complex behaviour pattern analysis:
• Sociability, “intensity” in life activities,
• Emotional state recognition (e.g. From acoustic analysis)
• Physiological state (e.g. Improved GSR analysis and brain function)

Closed-loop approach – feedback for main actors involved

• Doctors - objective information
• Therapy assessment
• Planning medication
• Scheduling appointments
• Caregivers - filtered information
• Warnings and risk profiles
• Patients - persuasive information
• Self-monitoring and assessment
• Motivation, coaching and self-treatment

Technology solution

The patient monitoring application was designed to measure two aspects, namely patients’ internal affective states, through the use of questionnaires; and their objective behaviour, through sampling of phone sensors. The application has been developed in close cooperation with the psychiatrics in order to capture relevant aspects of the disorder.

In order to increase patients’ motivation to provide daily experience sampling, the application provides alarms and reminders to fill out the questionnaire at a predefined time in the evening. Through the questionnaires the patients were able to provide their current state as well as activities they performed during the day, estimate their sleeping hours as well as quality, time spent outdoors and their social activities.

Psychiatric assessment and the psychological tests were performed every 3 weeks over a period of 3 months at TILAK (Department of Psychiatrics, State Hospital, Hall in Tyrol, Innsbruck). The schedule has been set by the psychiatrists in such manner as to reduce memory effect, which would have biased the evaluation outcome.

The clinicians used the following standard scales during the assessment of the patients:

• Hamilton Depression Scale (HAMD): the HAMD scale has been applied to rate the severity of depression in patients through the assessment of a range of symptoms. The higher the magnitude of symptoms, the higher is the scale of severity of depression (cut-off value: >=8)
• Young Mania Rating Scale (YMRS):  YMRS is most frequently utilized rating scale to assess manic symptoms. The baseline scores can differ in general, depending on the patients’ clinical features such as depression (YMRS=3) and for mania (YMRS=12).

Based on previous initial experiments and discussions with the medical personnel the following mobility based characteristics were assumed to be relevant:

1. Physical motion. Patients with depression tend to move less, move less forcefully and be slower overall. The opposite is true for manic patients.
2. Travel patterns. Most people have their travel routines dominated by a set of places, which they often visit in a certain temporal pattern. These patterns tend to change in both depressive and manic states (become less frequent or more erratic respectively). In addition depressive people tend to travel less and be outside less.
3. Social interaction. The way people interact with others can vary quite a lot. What people with bipolar disorder have in common is that in a depressive phase the desire and ability for social interaction is reduced, while during manic phase it is heightened.

Main results, benefits and impacts

The MONARCA system helps patients suffering from bipolar disorder to do daily self-assessment and to get timely feedback on how they are doing using persuasive visualisation and feedback mechanisms. Moreover, MONARCA also helps clinical professionals to obtain an objective picture of the patient state that is useful for therapy planning.

In the trials, we studied different aspects, including how the system was used and adopted, the usability of the system, and the usefulness of the system both for patients and clinicians in managing the illness. The results were positive; with the MONARCA approach it was found that there is correlation between the mood state and some variables monitored by the system such as amount of physical activity and location. This is a relevant result that is currently being further study in a complementary trial (being done beyond MONARCA).

Moreover, it turned out that the patients used the system almost every day; the system was considered very easy to use; and the usefulness of the system was high.

Furthermore the output of the data analysis as well as the preliminary results of a randomised clinical trial outcome seems promising. In other words, it was possible to introduce a monitoring system combining subjective and objective data, thereby getting all the benefits of having an electronic system, and most importantly that the system – according to the patients – did not introduce an overly high cognitive load and was able to improve the management of their illness.

Return on investment

Lessons learnt

When designing mobile health systems the focal point of the research is frequently concentrated on the design of innovative developments for improving the practice of healthcare and increase of wellbeing with a strong focus on functional requirements. In this regard, the aspects related to the definition of non-functional requirements of mobile health provisioning are often underestimated or left as a secondary item to take into consideration by researchers. However only through a thorough consideration of the potential implications for the design of non-functional requirements, can the mobile health innovations find an opportunity to be transformed into sustainable solutions that can be applied in real life contexts. These kinds of requirements comprise all the practical aspects of healthcare provisioning that are necessary to implement mobile health services ranging from human factors to important medical and technological issues.

As part of the experiences learnt in the MONARCA project we can identify not only aspects related to the innovative mobile health solution proposed by MONARCA but also on the technological and clinical aspects that were necessary for conducting multidisciplinary research in the context of such project and on other non-functional requirements that are key in the development of technological solutions for the design, development and evaluation of mobile health systems. Such requirements include aspects related to technology, human factors, medical practice, regulatory aspects and other practical issues that are identified as key challenges and lessons learnt useful for the development of future mobile personal health systems and services (see Figure 2).

Figure 2: Relevant Aspects in Multidisciplinary IT-based Clinical Research