The study compared the feasibility and usability of the Trilogis-Monsenso and PULSO systems among patients with bipolar disorder in Denmark, Italy and Spain. The article highlights differences in patients evaluations based on country as the Trilogis-Monsenso system was preferred over the PULSO system in Denmark, but there were no significant differences in evaluations in Italy and Spain. This paper provided an interesting insight into innovations in technology and how this can be applied within the mental health sphere. This is important given that integrating technology into health has the potential to improve patient care and outcomes.

The title and aims of the paper are clear. The introduction provides an overview of the prevalence and level of disability in people with bipolar disorder, and trends towards technological innovations in the field of physical and mental health, which sets the scene for the paper. The introduction also highlights the incidence of smart phone ownership, and the potential opportunities available with technology compared with classical EMA assessments. Acronyms are written out in full first and then are abbreviated. References appear correct, are relatively recent and appropriate key studies are included which relate to the aims of the paper. The research question is clear, and appears justified given the wider picture around digital innovation, the program of funding, and getting 'value for money' by investing in products that have an evidence base and are effective. The research appears likely to add value to current understanding, and is justified as it is important to assess these two systems feasibility and usability, and in what way these systems could potentially improve current practice. The potential usefulness of such technology is highlighted; illness management, early detection of deterioration, and the data aiding collaboration between patient and clinicians where the clinician can give feedback to patients based on the 'real time' data collected.

The methods section is written clearly and provides a useful summary of how the study was conducted and a description of both the Trilogis-Monsenso and PULSO systems. The study was conducted at all three NYMPHA partners’ locations (Italy, Spain, Denmark) by local doctors, clinicians and researchers during a discrete period from January 2018 to March 2018. This approach of is of benefit as the results are not exclusive to one country, and was not conducted at different time periods. The inclusion/exclusion criteria for the study, how clients aged 18+ were recruited, and the questionnaires used to confirm diagnosis and assess patient’s symptoms are outlined. The study recruited 20 patients per country who were randomised to one of the two technologies, which is strength as it avoids researcher effects from influencing the outcome of the study. Researchers who were blinded to the patients’ smartphone data and not involved in patients treatment conducted the evaluations. This is a strength as clients are less likely to feel inclined to report a particular view to please the researcher, and are more likely to be honest in their feedback.

The results section outlines further details about the participants included in the study, including the subtype of bipolar disorder, gender and the age range which was 18-69 years, except for Spain where it was 32-67 years. Although not explicitly stated here, there appears to have been a high level of adherence and engagement as only one patient in the Spain sample was noted to have not used the technology at all and so could not evaluate it. The two systems were evaluated as a combined system in Italy and Spain, but separately in Denmark. The results are clearly summarised, and Figures and Tables are included to summarise data and highlight comparisons.

The discussion succinctly highlights the outcome of the results, the limitations and future recommendations. The authors highlight how different evaluation methods between the three countries may have potentially played a role in the results obtained as users would have been either evaluating technologies separately or as a homogenous unit, highlighting a useful consideration for future research protocols. In addition, small sample size was also highlighted as a potential issue related to statistical power. The authors highlight the high level of participant retention in the study indicating adherence, but report that they were unable to obtain information about adherence to the app and how long/often participants wore the wrist band, which would have provided meaningful information. This provides useful considerations for the design of future research to ensure that an agreement is made between the research investigators and IT companies prior to beginning a study.

The paper would benefit from the following minor revisions including:

• Title/Aims: The title would benefit from the addition of the word “the” to make it more readable - i.e. “evaluation of THE usability...”. There is also a typo in the aims section (i.e. “testes” instead of “tested”). • Introduction: The introduction could have been improved in the following ways to provide a more balanced overview: 1) Including a brief overview of current research into how technologies have been helpful to date for people with bipolar disorder and in what way; 2) Elaboration of the problems with illness management currently within this population using classical methods, and whether technology could potentially resolve all or some of these issues; 3) A note about research around the problems with technological methods for some patients (e.g. skills deficit, cognitive impairments impacting engagement, etc). • Methods: The two systems were evaluated as a combined system in Italy and Spain, but separately in Denmark. It would have been helpful for the reader to understand the reasons for this difference in more detail. • Results: It would have been useful for the authors to comment on reasons behind non-engagement from one patient as this could have yielded useful learning. The samples of each country also included unequal male and female ratios. It would have been useful to include information about the ethnicity of the clients, as well as the mean age of each sample. It would have been helpful to have a key in Figures 3, 4, 5 and 6 rather than the labels in text along the bottom of the figures as this would help the reader understand the information more readily. Furthermore, Table 2 provides a useful overview of clinician’s feedback on both systems, but the methods section had not stated that clinicians’ feedback would be sought. This would need to be added to the methods section along with more specific details of how many clinicians were interviewed, their role, and how the feedback was obtained. Further analyses may have been warranted to understand whether evaluations varied according to patients’ age, gender or bipolar subtype, although I am unsure whether the sample may have been too small for these statistical tests. • Discussion: It would be useful for the authors to consider any additional explanations which could have explained the differences between the three countries, and/or suggest what other methods in future would help ascertain reasons behind the differences (e.g. qualitative methods to obtain feedback from patients about their experience of the systems). The discussion could have been improved by placing the current paper in the wider landscape of global evidence around digital technology which had been discussed in the introduction. The discussion could have explored the following as examples: Do the results fit with current understanding about this patient group's engagement with digital technology? Does it fit with other feasibility and usability studies for this group? (etc).