To the Editor: The motivating article by Bent and colleagues1 is a most welcome addition to the literature of what works in the movement for performance improvement. We can hardly overemphasise the need to share knowledge on innovations (ie, what works and what does not work) in the quest for best quality and safety practices. However, to aid efficient lesson-drawing, we are inclined to look for more contextual information and levers in any quality-of-care interventions.

For a safety research discussion, the report by Bent and others has at least three important elements: the use of a technology (personal digital assistants [PDAs]), the clinical performance of healthcare professionals (here, anaesthetists), and the permissive culture (to want to learn and improve). Nevertheless, what such innovative pilot practices should also incorporate and report are the contextual factors responsible for successful acceptance,2 application and appraisal of quality interventions.

Bearing continuity and sustainability in mind, one should be interested in the "characteristics" of anaesthetists who would voluntarily engage in personal monitoring and feedback. Initial technology use is seen among the "technologically proficient few" before becoming widespread.3 The introduction of PDAs for incident monitoring calls for the evaluation of the sociotechnical meta-system4 in which it will ultimately exist. Therefore, it is important for us to add a qualitative assessment to such a pilot study to identify personal motivating factors, climate for action, and the personal performance effects. Failure to evaluate technology "deployment" in healthcare results in lack of commitment, slow technology adoption, and perhaps decreased patient safety.5

The application of PDAs in reporting adverse events will increase within and across clinical disciplines and borders, but so must the rigorous appraisal to aid transference of knowledge. The global stage for international comparative research is widening, necessitating the need for integrated study designs, contextual analysis and robust reporting. It is often desirable to look for cost-effective means of improving patient care, with a dual learning carriage between institutions and nations. Patient safety and quality care studies will therefore continue to enjoy inputs from epidemiology, health services research, health economics, health policy, cognitive engineering, and information and communication technology. However, the main challenge remains: where is the patient in "patient safety"?.