The critical care expert's working mental model is a temporally sequenced, multidimensional, physiologically based matrix, in which 20 to 60 simultaneous clinical variables can be assimilated nearly instantaneously. The mental data-seeking exercise follows an expert's cognitive pattern of simultaneously developing answers, while still seeking more data. Any computerized, paperless incarnation of clinical data management in NICUs should have instant data access at the bedside for full support of focused, expert-level, cognitive work and decision making. High-speed information reporting is essential in all ICU environments.

Critical care clinicians often are interrupted by a more urgent situation. Standard computer access/use/log-off is very difficult to incorporate seamlessly into a NICU workflow. Critical care physicians report feeling cognitively blinded by computer-related fragmentation of the temporally flowing clinical data streams. Nurses are distracted by time-consuming, adult-designed charting systems requiring typed data entry. Reports are slow and fragmented.

Neonatologists and criticalists find that existing computerized charting methods in NICUs waste time. Without bedside data tracking, quick understanding of the overall situation of a particular patient, at the bedside, at a particular moment is virtually impossible. New computerization introduces change that often disrupts generations-old clinical workflow functions and may have many unintended consequences. Planning, developing, or purchasing and implementing effective systems for totally computerizing an NICU is an interdisciplinary work in progress.

Enormous political pressures are urging health care to "get computerized." Most hospitals are in the process of installing (or reinstalling) "hospital information systems (HISs)." Such efforts have exposed significant problems and limitations of current HIS systems for critical care situations. Extending office-designed computer systems safely and effectively into intensive care units and other real-time hospital venues has presented substantial difficulties. In late 2008, most NICUs still integrate multisource clinical data at the bedside by charting each hour with pen in small boxes on folding paper flowsheets, a slow, error-prone, and imprecise method for tracking unstable situations. To improve minute-by-minute management of unstable patients, all intensivists need automatic, real-time integration of machine, laboratory, and bedside observational data, presented in medically comprehensive, user-definable formats. Clinicians need reflex response speed (14 msec) in every clinical computer utility used in their "workflow." Similar sets of unique human computer/interactions of teams working in safety critical situations such as nuclear power plants, battleships, and airplane cockpits have been studied extensively by scientists known as "human factors engineers."

Human factors situations rarely have been studied in hospitals, and not at all for NICUs. Overall, this neonatal informatics series defines the current state of the "paperless" effort. Part 1 provides a historic and situational overview of current problems in optimizing hospital computer systems for critical clinicians, including communication, NICU computerization history, and an introduction to systems analysis, which is a key exercise for planning future NICU computer systems.