Healthcare systems that rely on ubiquitous computing often take it for granted that more data means better awareness, interpretation, and action. A new paper on arXiv argues that this assumption deserves closer scrutiny in end-of-life care, where the stakes of data-driven decision-making are especially high.

The study draws on semi-structured interviews to explore patients' experiences of breakdowns in the data-interpretation-action loop. Rather than celebrating the benefits of continuous health monitoring, the authors focus on moments when the flow of data fails to support patients, suggesting that visibility alone is not enough.

The paper's significance is its cautionary message for designers of ubiquitous health systems. If data can stop helping at the end of life, then system design must account for the limits of data visibility and the importance of human interpretation and action.