Secondary healthcare systems around the world are facing ever increasing pressures and governments frequently revise the provision of important services, often without available data or studies necessary to understand the demand placed on clinical teams. The situation is particularly severe during Out of Hours (OoH) settings, which account for over 75% of the working week. Here, doctors work in a stressful environment, performing complex tasks and making difficult prioritisation decisions. In this work, we discuss methodology and applications demonstrating the potential of data science and machine learning, in order to design intelligent systems capable of informing, supporting and driving improvements in health care policy and management. For that matter, we introduce the use of combined location and tasking informatics, sourcing diverse data allowing for the study of clinical behaviour and workload patterns in OoH secondary care settings; and we review example applications and ongoing work on Bayesian statistics, graphical models and queueing networks. Specifically, we exploit comprehensive data for the purposes of location identification and evaluation of task demandandworkload. Andweshowitispossibletogainvaluable information for the purposes of rota scheduling, task prioritisation or resource allocation, among other things.