Measurement of user performance and attitudes assists the initial design of a computer user display and orientation method.

OBJECTIVE: The objective of our study was to assess the acceptability of a proposed user interface to visually interfaced computer-assisted anesthesia record (VISI-CAARE), before the application was begun. The user interface was defined as the user display and its user orientation methods. METHODS: We designed methods to measure user performance and attitude toward two different anesthesia record procedures: (1) the traditional pen and paper anesthetic record procedure of our hospital, and (2) VISI-CAARE. Performance measurements included the reaction speed (identifying the type and time of an event) and completion speed (describing the event). Performance also included accuracy of the recorded time of the event and accuracy of the description. User attitude was measured by (1) the physician's rating on a scale of 0 to 9 of the potential usefulness of computers in anesthesia care; (2) willingness to use the future application in the clinical environment; and (3) user suggestions for change. These measurements were used in a randomized trial of 21 physicians, of which data from 20 were available. RESULTS: After exposure to VISI-CAARE, the experimental subjects' ranking of computer usefulness in anesthesia care improved significantly (4.2 +/- 1.1 to 7.6 +/- 1.5, p = 0.0001), as did controls' (5.2 +/- 2.6 to 8 +/- 1.5, p = 0.0019). All the volunteers were willing to try the proposed prototype clinically, when it was ready. VISI-CAARE exposure was associated with faster and more accurate reaction to events over the traditional pen and paper machine, and slower and more accurate description of events in an artificial mock setting. VISI-CAARE 1.1 demonstrated significant improvements in both reaction speed and completion speed over VISI-CAARE 1.0, after changes were made to the user display and orientation methods. CONCLUSION: With graphic user interface prototyping environments, one can obtain preliminary user attitude and performance data, even before application programming is begun. This may be helpful in revising initial display and orientation methods, while obtaining user interest and commitment before actual programming and clinical testing.

A collaborative approach to standards, practices. Setting the stage for continuous quality improvement.

In retrospect, the most important thing we did was work together. We analyzed, refined, and validated our philosophical approach to patient care. We provided an information data base that is readily available for on-the-job reference and serves as a starting point for CQI activities. The very act of joint documentation of practices encourages open discussions about improvements to patient care. One physician states, We know that flaws in the process through which we produce care are everywhere--waste, duplication of effort, unnecessary complexity, and unpredictability . . . I believe that modern total quality management offers enormous hope to a medical care field that is rather desperate. . . . Collaborative practice and CQI activities are one hope. The scope of what nurses and physicians traditionally consider when discussing standards and practices must widen. We should no longer look only at patient care. We must simultaneously focus on how the management of total systems influences quality care for all patients. The CQI process, a proactive method, requires an accurate data base of information that is easily retrieved when looking for systems and individual patient care improvements. Our Computerized Collaborative Standards and Practices Manual is the reservoir for documenting practice plans developed and approved by all the disciplines involved. The process described here began with two closely knit operating room disciplines; this framework, however, offers the potential for expansion into a hospital-wide system of information organization and use.

Comparison of high-frequency lung ventilation with conventional mechanical lung ventilation. Prospective trial in patients who have undergone cardiac operations.

High-frequency lung ventilation was compared with conventional mechanical lung ventilation following elective cardiac operation. The results indicate that this high-frequency ventilator works as well as conventional mechanical ventilators and that it accomplishes the desired gas exchange at lower peak airway pressures. We conclude that routine use of high-frequency ventilation in the postoperative period is possible and that it may be indicated if lower peak airway pressures are desired.

Computer assisted patient evaluation (CAPE): a multi-purpose computer system for an anesthesia service.

We have designed and built a database management system, the computer assisted patient evaluation (CAPE) system, for use in patient management, research, and administration in our anesthesia practice. An important part of the system is the use of specially designed forms on which anesthesiologists record patient histories and management information during the course of patient care. The forms provide means for convenient and complete record keeping, as well as for direct computerization. We demonstrate the flexibility and utility of the CAPE system by presenting a series of examples of its use. These include development and implementation of a preoperative screening program to identify patients at high risk of postoperative respiratory complications; a study of anesthesia technique and outcome; auditing for quality of care; and utilization review of respiratory therapy.

Bacterial contamination of arterial lines. A prospective study.

One hundred seventeen patients had indwelling arterial illness for hemodynamic monitoring and blood sampling. The duration of catheterization varied from 25 to 439 hours, during which time no components of the system were replaced. In contrast to other reports, our study showed no instance of contamination of transducer dome fluid when the continuous flush device was located just distal to the transducer. The sampling stopcock showed bacterial growth in 16.2% of patients. In the one case in which the arterial catheter tip, stopcock, and patient's blood showed the same organism, culture of the transducer fluid was negative. Our results suggest that elimination of a static inline fluid column and proper aseptic sampling technique limit risk to the patient of transmitted bacterial infection from the fluid in the system. Routine changes of components of the system are not indicated and a substantial cost saving can be achieved.

Anesthesia and the control of blood pressure in patients with spinal cord injury.

The medical records of 78 procedures performed on 50 spinal cord injured patients at risk for developing autonomic hyperreflexia were evaluated for blood pressure changes during anesthesia and surgery. Hypertension was arbitrarily defined as a systolic blood pressure of greater than 140 torr. The 78 procedures were divided into three groups: group A, 19 procedures in which the patient received topical anesthesia, sedation, or no anesthesia; group B, 13 procedures conducted under general anesthesia; and group C, 46 procedures carried out under spinal anesthesia. Hypertension occurred in 15 of 19 procedures (79%) in group A, in three of 13 procedures (23%) in group B, and in three of 46 procedures (7%) in group C. Group A differed significantly from group B (p = 0.011) and group C (p = 1.2 X 10(-8)). There was no significant difference between groups B and C (p = 0.114). Results indicate that patients at risk for autonomic hyperreflexia are protected from developing intraoperative hypertension by either general or spinal anesthesia.

The dynamic responses of liquid-filled catheter systems for direct measurements of blood pressure.

Direct measurement of blood pressure with a fluid-filled catheter and transducer is widely accepted in clinical practice. However, errors associated with the measurements are often not appreciated. The system frequently is unable to reproduce rapidly changing waveforms and overshoots to produce higher peak pressures. The most common causes of this phenomenon are trapped small air bubbles and long connecting tubing. To assess the inaccuracy in pressure measurements, we calculated the weighted sum of the percentage difference between reference and recorded amplitudes of sinusoidal waveforms for several catheters and connecting tubings. We found that when the connecting tubing was shorter than 3 feet long and no air bubbles were trapped readings were accurate. On the contrary, connecting tubings 7 feet long or longer, and/or air bubbles, were frequently associated with inaccurate results. For example, minimal air bubbles (0.25 ml) in a system exaggerated the systolic pressure measurement by 41 torr when simulated blood pressure was 150/50 torr.