Editor’s Note: As stated on above, “In My Opinion . . ” is not an APSF Editorial, but rather an opportunity for invited respondents to address interesting and controversial issues.
Vigilance is the motto of the ASA and the AANA. Is it important? Our emblem and most individuals say “Yes”. However, our textbooks imply “no.” A review of fourteen textbooks on display at the 1990 ASA Exhibits showed twelve without the word “vigilance” in the index. These textbooks covered the areas of pediatric, obstetric, intensive care, cardiovascular, and introductory reference anesthesia. Miller’s text, considered by many to be the best text available, makes only a brief statement about vigilance in context with anesthesia for the elderly. One British text has somewhat more than passing remarks, but that statement is less than a page in length.
A review of the anesthesiology literature also can lead to the conclusion that vigilance is not considered important. In six hours at a university library, only three editorials (1,2,3) and three original scientific paper (4,5,6) on vigilance in anesthesiology from all of the 1980s could be found. Two papers dealt with the effects of fatigue on vigilance. (4,5) The other one dealt with the effect of automation on vigilance. 6 Of the 1,273 abstracts presented at the 1990 ASA one purports to address the subject of vigilanM7 To an untrained em more than one of the above authors appear not to understand what vigilance is.
What is vigilance? No one in an average-sized community hospital department could come up with a good definition. No one in & audience at three different prestigious anesthesia programs could give either a dictionary or descriptive definition of vigilance. During the same library search, literally thousands of articles on vigilance published since World War II were found. Dozens of articles from the last five years seem to be relevant to the aspects of vigilance which might apply to anesthesiology. Several textbooks about vigilance are available. (8,9) Vigilance has been researched in workers with multiple various tasks including: food and industry inspectors, truck drivers, subway operators, nuclear power plant operators, air traffic controllers, pilots, submarine sonar operators, and multiple other task scenarios, but not anesthesia providers.
Vigilance is sustained attention. Attention toward what? What aspect of the anesthesia cockpit are we to keep our attention sustained toward the operative field, the monitors, the patient’s pupils, the I.V. drip rate, or the blood loss? How can we sustain our attention toward all these things at the same time? In fact, vigilance literature specifically states that we should be vigilant toward any signal which indicates a potential injury to the system which me human beings are monitoring. Vigilance is sustained attention, watching for a signal to which one should respond. The objective measure of vigilance and its effectiveness is the response time to that signal. Research on vigilance deals with the environmental, human, and signal characteristics which affect the response rate to that signal.
Characteristics intrinsic to the vigilance paradigm are:
1 . The task must be prolonged and continuous.
2 . The signals to be detected are highly discriminable to the alert observer.
3 . The signals are infrequent and aperiodic.
4. The human’s response has no effect on the signal rate
Initially, this may also sound confusing until one realizes that a signal is different from an event. An example of an event is an unchanged blood pressure recording of 1 30/80. An example of a signal is a new, sudden blood pressure recording of 70/55. A signal is an event to which @Km should respond. Since we are not supposed to respond to a normal and unchanging blood pressure, vigilance toward that event (absence of change) is at best inconsequential, and at worst harmful. Likewise, are we to be vigilant toward the intentional clamping of an esophageal stethoscope, or toward an event which can injure the patient, such as the disconnection of an endotracheal tube? The former will not injure the patient by itself, and may not even change our detection of the latter signal, depending on our other monitoring techniques.
Fatigue certainly plays a role in vigilance, and yet it is only one of more than thirty factors which have been clearly shown to affect vigilance tasks. What are the effects of things such as: background event rate, caffeine, propranolol, smoking, false alarm rate, length of case, and number of cases per day on vigilance in anesthesiology? As the event rate (i.e. number of vital signs per hour) increases, and the signal rate decreases (as implied through apparent increased safety in anesthesiology), we appear to be heading toward a scenario in which human vigilance can be severely negatively impacted. At what level of monitoring does the anesthetist hit his or her “biological barrier” where the mind becomes confused by processing too much data and the speed and accuracy with which one responds to a signal decreases?
Anesthesiologists should be students of the science of vigilance. We should apply that knowledge to research in concerning anesthesia tasks. It is possible we have not done this because we have not had the tools. It is believed that the automated anesthesia record may serve as one of those tools. Mackie states that “vigilance is extraordinarily time dependent, as the four aspects of the vigilance paradigm demonstrates” (11) A computer can be used as a time monitor of the chronology of events in the operating room. Therefore, response times can be studied. Once this tool is developed, possibly we can determine the effect of circadian rhythms and other often-cited factors on Vigilance in clinical anesthesia practice.
Other types of tasks that have been studied appear to show that over-automation can lead to boredom resulting in decreased vigilance with the “human-computer”, team. (12,8) However, correct application of the human-computer team in the task involved can dramatically improve Vigilance. For instance, humans are notoriously poor with negative signals, whereas computers are relatively good at detecting negative signals. Humans are good at prioritizing signals, computers are relatively poor. As automated records and servo-control mechanisms linked to end-tidal carbon dioxide, agent concentration, and blood pressure gain acceptance, the issue of human-computer interactive performance becomes more important.
It can be argued that complex multioriented tasks do not have a vigilance decrement. (13) If that is the case in anesthesiology, then addressing the above questions will demonstrate that vigilance is not a real issue in our profession. Should we not then change our oft-quoted motto?
If “vigilance” in anesthesiology is only a euphemism, lets change our motto. If it is a valid concept, let’s apply it to our daily education and practice
Dr. Edsall is Chairman, Department of Anesthesiology, Burbank Hospital, Fitchburg, MA.
1. A matter of vigilance(Editorial). Anesthesia 41:129-130, 1986.
2. Minimal monitoring and vigilance (Editorial). Anesthesia 42:683-684,1987.
3. Why investigate vigilance? (Editorial). J Clin Monit 2(3): 145-147, 1986.
4. Denisco RA, Drummond JN, Gravenstein JS: The effect of fatigue on the performance of a simulated anesthetic monitoring task. I Clin Monit 3(l): 22-24, 1987.
5. Paget NS, Lambert TF, Sridhark K. Factors affecting an anesthetist’s work: Some findings on vigilance and performance Anaesth intens care 9:359, 1981.
6. Kay 1, Ned M: Effect of automatic blood pressure devices on vigilance of anesthesia residents. J Clin Monit 2(3): 148-150,1986.
7. Yablok, DO: Comparison of vigilance using automated versus handwritten records. Anesthesiology 73 (3A): A416, 1990.
8. Warm, IS (ed): Sustained attention in human performance The Wiley Series on Studies in Human Performance. Volume 4.
9. Wiener, EL(ed): Human Factors in Aviation. Harcourt Brace Jovanovich, Publishers.
10. Percival LC, Noonan TK: Computer network operation: Applicability of the vigilance paradigm to key tasks. Human Factors 29(6): 685-694, 1987.
11. Mackie RR: Vigilance research are we ready for counter-measures? Human Factors 29 (6): 707-723, 1987.
12. Parasuraman R: Human-computer monitoring. Human Factors 2 9(6): 695-706, 198 7.
13. Wiener EL: Application of vigilance research: Rare medium or well done? Human Factors 29 (6): 725-736, 1987.