Hospitals worldwide are under increasing pressure as patient intake rises while fewer staff are available to provide care. These challenges are particularly evident in Intensive Care Units (ICUs), where patients rely on one-to-one care. In the ICU, monitoring and frequent manual therapy adjustments lead to high workloads among clinical staff. (Semi-)closed-loop therapy systems, which automatically adjust or recommend therapy actions based on physiological data, offer the potential to reduce this workload while also improving the accuracy and consistency of care.
Despite growing evidence regarding the technical feasibility of these systems, limited research addresses how they should be designed for integration into the complex socio-technical context of the ICU. This thesis addresses this gap by developing a design framework aimed at overcoming adoption barriers and supporting their implementation within clinical workflows. Using blood pressure management as a case study, the research combined ICU observations, semi-structured interviews with ICU nurses, and a co-creation session involving clinical stakeholders and designers.
The findings resulted in nine guiding design principles: (1) Trust and confidence, (2) Regulation and legislation, (3) Reliable and safe functioning, (4) Context-aware automation, (5) System transparency, (6) Authority and control boundaries, (7) Critical clinical supervision, (8) Cognitive load management, and (9) Proven effectiveness. Each principle consists of underlying concerns, design challenges, and guiding questions to support practical application. Additionally, a structured design approach was developed to translate these principles into practice. By adapting the Systems Engineering Initiative for Patient Safety (SEIPS) model and integrating it with the Double Diamond process, this approach provides a step-by-step method for designers to move from analyzing today’s clinical reality to building a vision for future integration.
Together, the design principles and structured methodology demonstrate that successful implementation of (semi-)closed-loop therapy systems depends on carefully balancing technological capabilities with human and organizational needs. In doing so, this thesis offers both practical insights and a structured method for integrating (semi-)closed-loop therapy systems into intensive care practice.