In November 2025, HUD organized an expert co-design event at the University of Copenhagen. The workshop expanded on HUD’s thematic focus on mental health in detention, as well as the development of digital mental health tools. The workshop drew together a transdisciplinary and transvocational mix of perspectives. Humanitarian practitioners included mental health and psychosocial support specialists based at the International Committee of the Red Cross, Terre des hommes, and the International Federation of the Red Cross and Red Crescent. Researchers from across disciplines including political science, global health, medicine, engineering, computer science, and beyond also attended.

The workshop built on HUD’s basic research into humanitarian conditions in prisons – following its core co-design methodology. In that methodology, a bottoms-up needs assessment was conducted with local civil society, researchers, and other actors – in Colombia and the DRC – before soliciting the expertise of international humanitarian actors themselves. Going forward, this constant dialogue between those affected by crisis, and those with the potential to help reduce its effects, will continue. Below, a report summarizes the main findings of the workshop, tracing ‘where’ digital mental health intiatives are today, towards imaginaries of their future design, before ending on a discussion of the next steps that HUD aims to take in this area.

To capitalize on the opportunities that emerging technologies might contain for improving mental health and psychosocial support, while avoiding their attendant risks, the workshop reached a general consensus that an ideal-world digital e-CDSS for MHPSS would pair validated technical models with a validated user ‘platform’ that is:

Based on the discussions at the workshop, below is a schematic description of where such a digital e-CDSS system would fit in the ‘flow’ of a primary health care system – alternating between health worker training, patient screening, and data analytics. As discussed further below, and at length during the workshop, this is an ideal-type schematic and, in reality, choices would need to be made vis-à-vis which healthcare workers are targeted, whether specific populations (of patients) are targeted, and so forth.

Given resource constraints, the risk of siloed-thinking and product development, and other concerns, designing an e-CDSS system that gains sustained buy-in from as many possible actors was a crucial discussion point across the workshop. Figure 2, below, and Table 1, below, demonstrate the general idea that emerged from the workshop of achieving this through developing a transversal tool of value to three key stakeholders (healthcare workers, patients, and authorities).

Actor	Value Propositions
Authorities	An e-CDSS of this kind could provide society-wide healthcare system capacity building and birds-eye-view data analytics for population-level screening of MHPSS conditions. It would operate as a simple, cost-effective ‘air traffic control’ system or ‘satellite’ image of mental health burdens and where best to target resources.
Primary healthcare providers	An e-CDSS of this kind would save time, increase efficiency, be easy to use for non-specialist health workers, and preserve the value of humans in the care system. Analogously, it would be a basic ‘x-ray’ tool for MHPSS allowing brief non-invasive screening, saving vast human resources on the ground of low-resource settings.
Patients (and families of patients)	An e-CDSS of this kind would have the potential to offer non-invasive, low-intensity, and low-cost treatment available at point of primary-health care contact for a much wider segment of society in low-income settings than at present.

The above is an ideal-world scenario, but one that the workshop nonetheless made clear can be pragmatically achieved albeit in a processual manner, following further research and development. At this stage, the workshop also concluded that such research and development should focus on a platform for primary healthcare providers in low-income/humanitarian settings, which has the potential for the highest impact given the shortage of qualified mental health and psychosocial professionals in these settings, as well as other contextual factors (stigma, lack of resources, etc.). Beyond this, choices also need to be made at both technical and social levels. Specifically:

• A vertical ‘algorithm’ or computerized adaptive testing model for screening;
• A horizontal LLM-driven model for screening.

Each has its advantages and disadvantages as well as ease/complexity of validation, and different concerns relating to maintenance and sustainability, confidentiality and data, protection, and so forth. For the ‘training’ component, different options are equally available, but the workshop generally agreed on continuing the work of the MhGAP solution, but finding ways to make it more interactive, expanding beyond a simple digital copy of the paper tool, and finding ways to integrate forms of certification of training that may open up access to more complex digital tools.

• Developing a frontend designed with an integrated user-experience in mind that can triangulate between the three elements identified above (screening, training, analyzing), and which balances ease-of-use with extended functionality;
• Judging the degree to which such an e-CDSS integrates ‘local knowledge’ – including e.g. traditional medicine – into its functionality (whether through a modular platform structure or otherwise);
• Integrating such an e-CDSS into existing local, national, and international legal frameworks;
• Testing this user interface and technical backend in multiple different contexts, to further user-centred design work in this area for further ideas here;
• Researching, through a mix of interviews, observation, and controlled tests, the effects of introducing an e-CDSS into clinical settings on e.g. patient openness, the patient-clinician relationship, etc.

HUD’s Copenhagen and EPFL teams are now launching an intensive eighteen-month long period of resesarch and techncial development, supported by a technical working group of humanitairan professionals and researchers. This will involve a team made-up of a data scientist and research engineer, a specialist in development engineering, medical professionals, and social scientists working across Denmark, Colombia, the Democratic Republic of the Congo, and Switzerland to co-design prototype versions of the e-CDSS sketched above. At this stage, our work remains at the research and development stage, albeit with a long-term view towards viable product development and implementation in order to help improve global mental health and psychosocial support in – yet also beyond – humanitarian crisis.