A 65-year-old man with bipolar disorder was scheduled for maintenance electroconvulsive therapy (ECT), a procedure he had received dozens of times before. These procedures are usually administered in the morning by an anesthesiologist. Later in the morning, that anesthesiologist staffs the preoperative anesthesia clinic. A second anesthesiologist in the group spends the day in the operating room (OR) supervising three nurse anesthetists.

On the day of the scheduled ECT, the clinic anesthesiologist called in sick. The service, which had no policy to cover unexpected absences and rarely cancelled a case, scrambled to try to cover the ECT procedure, finally asking the OR anesthesiologist to attend on the case. He protested, noting that he was already responsible for overseeing the anesthesia for three surgeries in the OR. But finally, to avoid delays in the ECT administration, he reluctantly agreed to come to the day surgery unit to perform the quick (usually less than 10 minutes) procedure.

After the ECT was performed, the patient did not wake up promptly as expected. The anesthesiologist checked his medication cart and realized that he had inadvertently administered the intermediate-acting muscle relaxant rocuronium, instead of the short-acting agent succinylcholine. He later attributed this error to being rushed and stressed, leading him to pull the wrong vial from the refrigerator. He had never made this mistake before.

The patient was given more sedation to prevent awareness and bag-and-mask ventilation to support his respirations. After about 30 minutes, his neuromuscular blockade was pharmacologically reversed with neostigmine. The patient was informed of the error, and there were no long-term adverse consequences.

by Pascale Carayon, PhD

This case—in which an OR anesthesiologist is pressured to step in at the last minute to replace a colleague and administer ECT—vividly illustrates the issue of production pressure. Production pressures are the "overt or covert pressures and incentives on personnel to place production, not safety, as their primary priority."(1) After the clinic anesthesiologist called in sick, clinic personnel desperately sought a short-term solution to accomplish the scheduled ECT. Adding to the pressure, it is likely that there was no contingency planning for this situation. Instead, we have a patient who has already arrived at the clinic and is waiting to receive his scheduled ECT, creating pressure on the clinic staff to get the procedure done that day, and subsequently on the OR anesthesiologist to attend on the case. The OR anesthesiologist already had a high workload—supervising the work of three nurse anesthetists for three surgeries in the OR. Faced with the pressure to avoid delaying the ECT administration—some of which was likely self-generated, sympathizing with the needs of this elderly patient with bipolar disorder who has received the ECT procedure many times before—he finally agreed to perform the procedure. Although proving causality is difficult, it seems logical to attribute the medication error of the OR anesthesiologist, one he had never before committed, at least in part to the stressful circumstances and the ball juggling he undoubtedly had to perform.

Production Pressures in Health Care

Research has shown that various categories of health care workers experience different types of pressure, such as time pressure (i.e., having to perform tasks in a short period of time) and production pressure (i.e., having to "produce" at the expense of safety).(1) Gaba and colleagues (1) reported that 49% of the surveyed anesthesiologists have "observed an anesthetist pressured to conduct anesthesia in a fashion [one] considered unsafe given the level of urgency of the situation." These different types of work pressure can be sources of stress, and even burnout (2,3), and may threaten patient safety by increasing the likelihood of errors and workarounds.(4) Given the shortage in many medical specialties and nursing, it is important to understand how to design the work system to minimize pressures and stress on the health care providers, as well as to improve the quality and safety of care.(5) Those in upper management who are responsible for designing the work systems need to understand the negative impact of production pressures on both the health care providers and the patients.

Various organizational and systemic factors can contribute to production pressures, such as poorly designed workload policies, inadequate staffing, poor incentive systems, hierarchical systems, competition for cases, long work hours, and low safety culture. A system analysis of production pressures examines all of the work system factors that can be responsible for creating both objective and subjective pressures. In approaching this type of analysis, the work system can be conceptualized as having five elements—person, tasks, tools/technologies, physical environment, and the organization (6,7)—that should each be systematically considered. For instance, high workload or having a lot to do, a characteristic of the "tasks," can be a source of production pressure. The case shows this clearly: the anesthesiologist is faced with the tasks related to his regular workload of supervising three nurse anesthetists, but also those of a sick colleague.

Workload Management

In addition to the amount of work, the case illustrates another critical factor in production pressure—the management of workload. This factor is comprised of two elements: (i) predicting and planning workload and (ii) establishing policies and procedures for handling changes in workload. Safe organizations put effort into analyzing the work system and attempting to predict and plan for workload. Such information can be used to quantitatively model workload, therefore leading to more reliable, effective management of resources, including human resources (e.g., staffing) and other resources (e.g., OR allocation), and possibly reducing production pressures.(8) For instance, Mullinax and Lawley (9) present a methodology for assigning patients to nurses in neonatal intensive care (NICU) based on (i) quantification of nursing workload for each NICU patient and (ii) a decision analysis model (integer linear programming) that assigns patients to nurses while balancing nurse workloads. A test of the methodology with 10 case studies in an academic medical center shows that this novel approach to managing workload led to a better balance of nurse workload. McManus and colleagues (10,11) provide another example of the application of industrial and systems engineering tools to workload management in the context of surgical caseload and availability of beds in an intensive care unit. This research demonstrates that the variability of scheduled admissions was greater than the variability of emergencies.

Very often, changes in workload occur in health care in an unpredictable manner. Therefore, it is also important to establish policies and procedures for handling those sudden, unpredictable changes in workload. Such policies and procedures need to rely on the front-line staff as much as possible and provide autonomy to the local actors in their response to the sudden change in workload. The local actors need to be provided with the resources, responsibility, and autonomy to be able to respond to the workload changes, and therefore minimize the impact of production pressures. The Toyota Production System is famous for providing front-line workers with this type of local control, but the case of the NUMMI (New United Motor Manufacturing, Inc.) plant that implemented the Toyota Production System illustrates how local control can sometimes be an "illusion of control" if not accompanied by "real" resources and empowerment. As this one influential case study described, workers on the assembly line were told that they could stop the line when they were experiencing some problem. This control over workload can be a very positive job characteristic. However, the investigators found that when a worker did stop the line, there were so many negative consequences (e.g., being identified as the "weak link," feeling that workers down the line are let down) that actually exerting that control is not an option.(12) Without a robust management commitment and a culture that clearly prizes safety over production, it is very easy for the promise of local control or autonomy to deteriorate into an illusion of control. In addition, there has to be some high-level management and monitoring of the process in order for the entire system and process to function safely, effectively, and efficiently while ensuring appropriate and sufficient local autonomy.

What Can Be Done?

A possible solution to reduce production pressures is through staffing policies and procedures. If there is sufficient staff to perform all of the tasks, it is likely that production pressures will be minimized. It is important to understand that staffing policies and procedures need to go beyond the number of people and examine the entire work system to understand the factors that contribute to workload.(13) Adequate staffing levels depend not only on the number of people but also their skills, knowledge, and experience, as well as many characteristics of the work system or clinical microsystem, such as effective scheduling and facility planning that will allow better use of staff resources. Therefore, health care organizations need to evaluate their work systems and the factors that contribute and/or add to workload and production pressures. This case illustrates an everyday phenomenon in workplaces throughout the world: Without a culture in which safety trumps production, adequate staffing, clear policies and procedures, adequate working conditions, and front-line empowerment and resources, proclamations about the importance of safety are belied by actual events. It is hard to blame the staff for wanting to get the ECT done that day, or the anesthesiologist for reluctantly agreeing to staff the case. But the result is a medication error that could have caused major harm.

A better system would be one in which:

1. A prospective analysis assesses the daily workload of the anesthesiologists in both the OR and the outpatient center, ensuring that staffing is adequate for both usual and unusually busy days. Such an analysis would also consider how often anesthesiologists (or other staff) have unplanned absences and the impact of these absences on workflow and workload. This analysis would need to be performed on a continuous basis and take into account organizational changes.
2. Staff would be brought together to discuss how to handle these unusually busy days—whether caused by an unusual volume/complexity of patients or unplanned staff absences.
3. Management would help staff identify additional resources that could be made available on such days—for example, could an anesthesiologist be pulled from a critical care unit or an affiliated hospital to help out?
4. Everyone would agree that, when staff members felt that a certain level of production was unsafe, they would be uniformly supported if they chose to "stop the presses." This is precisely what the High Reliability Organization approach (14,15), along with other organizations with impressive records of safety and quality, now emphasizes, and it needs to become a standard part of the culture of health care.
5. At a broader policy level, accreditors should consider standards related to methods for managing workload. In aviation, for example, the Federal Aviation Administration (FAA) strictly regulates the frequency of take-offs and landings. This would be far trickier to do in health care. However, health care organizations should be strongly encouraged to design and implement systematic methods for managing workload and making decisions regarding the number of cases per day in an OR or staffing ratios in order to appropriately balance the inevitable tensions between production pressures and patient safety.

Take-Home Points

• Production pressures are caused by various factors embedded in the structure of health care work systems and organizations. They can affect health care providers in the form of stress and burnout, and the safety of care provided to patients.
• Consider identifying work system factors that contribute to production pressures, and implement interventions targeted at those system factors.
• Health care organizations should analyze their work systems and workflows, helping them to appreciate the frequency and circumstances—unplanned increases in patient volume or complexity, unplanned staff absences—that lead to increased production pressures.
• Develop methods, policies, and procedures for managing workload. Although some of these will be "top down," the front-line providers also need to be empowered to call up resources, solve problems, and if necessary "stop the presses" when production pressures threaten to compromise safety.

Pascale Carayon, PhD
Procter & Gamble Bascom Professor in Total Quality, Department of Industrial and Systems Engineering
Director of the Center for Quality and Productivity Improvement
University of Wisconsin-Madison

1. Gaba DM, Howard SK, Jump B. Production pressure in the work environment. California anesthesiologists' attitudes and experiences. Anesthesiology. 1994;81:488-500. [go to PubMed]

2. Carayon P, Zijlstra F. Relationship between job control, work pressure and strain: studies in the USA and in The Netherlands. Work Stress. 1999;13:32-48.

3. Maslach C, Schaufeli WB, Leiter MP. Job burnout. Annu Rev Psychol. 2001;52:397-422. [go to PubMed]

4. Reason J. Managing the Risks of Organizational Accidents. Burlington, VT: Ashgate; 1997.

5. Carayon P, Hundt AS, Karsh B-T, et al. Work system design for patient safety: the SEIPS model. Qual Saf Health Care. 2006;15(suppl 1):i50-i58. [go to PubMed]

6. Carayon P, Smith MJ. Work organization and ergonomics. Appl Ergon. 2000;31:649-662. [go to PubMed]

7. Smith MJ, Carayon-Sainfort P. A balance theory of job design for stress reduction. Int J Ind Ergon. 1989;4:67-79.

8. Reid PP, Compton WD, Grossman JH, Fanjiang G, eds. Building a Better Delivery System: A New Engineering/Health Care Partnership. Washington, DC: National Academies Press; 2005.

9. Mullinax C, Lawley M. Assigning patients to nurses in neonatal intensive care. J Oper Res Soc. 2002;53:25-35.

10. McManus ML, Long MC, Cooper A, et al. Variability in surgical caseload and access to intensive care services. Anesthesiology. 2003;98:1491-1496. [go to PubMed]

11. McManus ML, Long MC, Cooper A, Litvak E. Queuing theory accurately models the need for critical care resources. Anesthesiology. 2004;100:1271-1276. [go to PubMed]

12. Parker M, Slaughter J. Management by stress. Technol Rev. 1988;91(7):36-44.

13. Carayon P, Gurses A. A human factors engineering conceptual framework of nursing workload and patient safety in intensive care units. Intensive Critical Care Nurs. 2005;21:284-301. [go to PubMed]

14. Weick KE, Sutcliffe KM. Managing the Unexpected: Assuring High Performance in an Age of Complexity. San Francisco, CA: Jossey-Bass; 2001.

15. Roberts KH, Bea RG. When systems fail. Organ Dyn. Winter 2001;29:179-191.