Introduction
Medical errors continue to be a significant cause of injury and death. In the United States, it is estimated that medical errors result in 44,000–98,000 unnecessary deaths each year and over one million unintentional injuries [1]. Sentinel events are unanticipated outcomes resulting in death or serious physical or psychological harm, unrelated to the presenting disease. While sentinel events are the errors that make the headlines, such as wrong site or wrong patient surgery, the majority of medical errors are less dramatic, may even escape notice, and do not always translate directly into an adverse outcome [2, 3]. These small errors are frequently the result of system errors and deficiencies rather the mistakes of individuals. The Institute of Medicine’s reports on quality in healthcare delivery, To Err Is Human and Crossing the Quality Chasm, describe designing systems with an emphasis on human factors and safety [3, 4]. These principles have been the center of many recent research efforts. The focus of error prevention continues to shift away from individual mistakes to systemic failures and system inadequacies.
In a complex setting such as the operating room (OR), gathering meaningful data for quality improvement is a challenge. Recently, surgical flow disruptions (SFD) have been used as a proxy measurement to understand better how to improve current systems. The goals of this chapter are to summarize how human factors apply in the context of surgery, to define and describe previous research into SFD, and to offer strategies to start to manage disruptions in the OR.
Human Factors in Surgery
Human factors is a discipline that explores the interface between humans and the systems in which they work. The field includes the study, design, implementation, and testing of environments and processes to ensure safe, effective, and efficient use to accomplish an intended goal. Concerning healthcare, human factors research focuses on maximizing human performance and efficiency, to promote health, safety, comfort, and quality of life for not only patients but also practitioners [5]. There are three fundamental principles to keep in mind when evaluating systems based approaches to medical errors: (1) human error is unavoidable, (2) defective systems allow human error to cause harm to the patient, and (3) systems can be designed to prevent or detect failure before actual harm to the patient occurs [6]. With these principles in mind, there have been many different frameworks and strategies established to build and improve on existing systems. One widely applied and cited model that helps to conceptualize a complex system is Reason’s Swiss Cheese Model.
British psychologist James Reason pioneered the modern field of systems analysis with his famous and widely studied “Swiss Cheese” model. Reason presented this model an explanation of preventable error. The model explains that every step in a process has the potential to contribute to systemic failure. The slices represent multiple layers of defense while the holes that make up Swiss cheese are systemic shortcomings. An error, when the holes in different layers align, perfectly penetrates all defensive layers. In summary, most accidents or adverse outcomes result from the combined effect of many smaller errors due to underlying system flaws that when aligned lead to disastrous consequences [7, 8]. A major aspect to understand in this model is that the holes (errors) will always exist; therefore, it is important to provide multiple layers of defense to decrease the chance of alignment, allowing individual errors to amass and result in an adverse outcome. This model also relates to the idea of failure mode and effects analysis.
The Swiss Cheese Model has been extended to include the concepts of active and latent failures. Active failures are unsafe acts committed by humans, typically by accident, with an immediate negative impact [9]. An example of an active failure in surgery would be an unintentional ligation of an artery. A latent failure is the result of circumstances, deficiencies, or holes in the system. An example of a latent factor could be improper labeling of a medication vial that eventually leads to incorrect dose administration. What makes latent failures particularly dangerous is that the impact may not be immediately apparent. A latent failure may only become apparent once an adverse outcome has occurred and is recognized. The OR provides an environment and opportunity to study latent failures due to the inherent complexity and acuity of activity that routinely takes place.
Despite representing only 1.9–3.6 percent of all hospitalizations, surgery represents 46–65 percent of all adverse events [10, 11]. ORs are complex, high-stress environments that challenge systems design for many reasons. First, each OR has a different personnel dynamic with multiple care teams such as nursing, anesthesia, and surgery. Further complicating this is the fact that each individual member of the team has different training, varying immediate goals, different measures of productivity and success, and yet all members are expected to work seamlessly in a tight setting [12]. In addition to diverse team members in the OR during a single procedure, the hospital operating area likely houses multiple specialties. High patient volume and turnover with many different specialties, each with their unique configurations, and confounded by personnel frequently coming in and out of the OR, create a complex daily environment.
In addition to personnel complexities, the actual working environmental plays a role in surgery performance and outcomes. For example, Spaghetti syndrome, a term used to describe the congestion created by wires, cables, and lines is a typical scenario first described in the intensive care unit [13]. An environment with complex technical equipment that is often not be well organized illustrates an accident waiting to happen. In addition, clutter, noise, temperature, and lighting have all been demonstrated to play a role in surgical outcomes.
It is self-evident that the OR has the potential to be an incredibly high-stress environment. There is clear evidence that excessive levels of stress have an effect on outcomes in surgical performance. Quantitative and qualitative studies have recognized stress as an essential component of surgical performance [14]. In addition to environment, communication, and team dynamics, interruptions and distractions are frequently a factor in surgical systems failure.
Both active and latent errors occur in the OR. Active failures usually are identified quickly as they become evident and are typically straightforward to measure. In contrast, investigators and quality improvement teams are challenged with how to measure, characterize, and quantify systemic issues and potential latent errors. In order to properly measure latent errors, teams often utilize proxy measurements that operationalize different hard to measure components of systems, such as SFD.
Flow Disruptions
Mihaly Csikszentmihalyi first described the concept of flow. Csikszentmihalyi wrote that flow is that state of optimal experience when individuals or groups are fully immersed in the task in front of them, leading to optimal production in that particular situation [15]. Individuals are so absorbed that nothing outside seems to matter [15, 16]. This state is colloquially known as being “in the zone,” and the theory describes that performance is at its peak when the state of flow is reached. When in the state of flow, individuals operate at their personal capacity [16]. To reach a state of flow, there must be a balance between the task, the skill of the individual or team, and the environment. Flow has been tied to performance and is explained by improved concentration and intrinsic motivation. Although not explicitly stated, the idea of flow can also be connected to mindfulness or being fully present and attentive in the task or moment. Practices that emphasize being fully present and reaching one’s “flow” have been linked to better clinical decision making, with due attention to present moment, improved clinician mental health, and reduced burnout.
As a complex activity, surgery possesses natural flow when procedures progress with ease and fluidity. Csikszentmihalyi even mentions the field of surgery specifically in his writing about how performers involved in demanding tasks with critical implications can reach flow [17]. Wiegmann et al. defines SFD as deviations from this natural fluidity of a procedure that can thereby potentially compromise the safety of the operation [18]. Flow disruptions provide a proxy measurement for quality and safety by operationalizing latent errors. Also, application of the previous concepts show that flow disruptions represent barriers to optimal performance and can lead to increased intraoperative stress and increased mental workload. Individual flow disruptions may not lead to an adverse outcome, but the accumulation of disruptions may highlight the process and system holes that could lead to error.
The significance of flow disruptions and their relationship to quality has been explored. Some studies have looked at the nature and frequency of SFD. In 2006 Healey et al. looked to quantify flow disruptions in urology procedures. A trained observer recorded a mean of .45 event per minute and found most of the distractions to be related to team communication, equipment or environmental issues, and procedural challenges. This study also found disruptions relating to outside or case irrelevant conversations, work environment issues, telephone calls, and equipment failures. Healey found discussions to be the most frequent and severe interruptions in their observed procedures [19]. Healey and colleagues later examined distractions and interruptions during fifty general surgical procedures. They defined disruptions as a break in attention, evidenced by observed behavior. The interruptions were characterized by a rating system from least severe or potentially distracting (1) to a completion interruption (9). The team recorded phone calls, beepers, communications with external staff and other noise of varying volumes that may have influenced teamwork in OR and outcomes [20].
In 2007, Wiegmann and team also explored disruptions in surgical flow. The team also recorded errors and disruptions in flow during 31 cardiac procedures. These recordings were classified and analyzed by a team of human factors experts. Similar to Healey’s findings, Wiegmann found the flow disruptions to be mostly related to communication or communication breakdown. In fact, 52 percent of the disruptions were related to teamwork or communication, and the number of disruptions was the strongest predictor of error. An example of communication breakdown provided in this study was “surgeon was under the impression that the patient had been given x when x had not been given.” Wiegmann found that there was a correlation between rate of errors and SFD. Specifically, the relationship was significant for disruptions classified under teamwork or communication errors. This was one of the earlier studies to present the empirical link between flow disruptions and OR errors [18].
Building on the previous work, in 2010 Parker and her team developed and validated a tool to categorize latent errors that may lead to adverse outcomes. The tool provides a tested framework to categorize “precursor events” and help teams to design their own interventions to reduce error through system diagnosis and improvement. The tool was validated and demonstrated to be simple and clear for observers of diverse academic backgrounds to use [21].
Catchpole expanded the scope of the topic to look outside the OR and to include trauma patients and handoffs [22]. They found that for trauma patients, the most common flow disruptions were coordination issues and communication failures, which is consistent with previous studies. They analyzed flow disruptions during care transitions that patients experience during trauma care. By studying 181 trauma patients, they found that 42 percent experienced at least one disruption during the transition of care and 53 percent of the disruptions were related to poor coordination of care. Disruptions included: patient taken to the incorrect room, critical team members missing, and imaging equipment occupied, all resulting in a delay of care. One important finding presented in this study was that the sicker, more acute patients had less standardized handoffs, and more flow disruptions, despite not having an increased number of handoffs [22].
One gap in the current literature is the question of to what extent individual OR team members recognize and perceive OR flow disruptions. Silver et al. performed a survey of over 100 staff members from three academic institutions using the validated tools described. Sixty-five percent of respondents acknowledged that flow disruptions happen either several times a day or every procedure and 40 percent of respondents ranked poor communication between teams to be the most frequent cause of SFD. Finally, respondents felt that staff burnout, patient safety, and economic consequences were the most significant consequences of repeated flow disruptions [23]. Consistently throughout the literature, authors note that flow disruptions in surgery are common and in and of themselves do not indicate a failure. Furthermore, flow disruptions may or may not be significant in causing adverse outcomes due to an overwhelming number of variables within an OR. The challenge for researchers and those charged with improving outcomes effectiveness is accurately measuring and categorizing flow disruptions to further understand where individual processes may have holes.
Opportunities for Leadership and Improvement
Understanding the concept of SFD provides an opportunity for improvement when applied with the goal to improving OR systems. The first step is to quantify and categorize SFD with the validated tools previously mentioned. The information gathered with these tools can be used to help improve operations and also provide a foundation for understanding factors leading latent failures and adverse events [21].
In addition to measurement, one strategy that has been explored to decrease SFD is an approach borrowed from aviation called “sterile cockpit.” In commercial aviation, the sterile cockpit is that time from 0 to 10,000 feet and vice versa. A sterile cockpit protocol is activated when critical periods of a procedure are occurring. The sterile cockpit concept ensures that during critical times, all communication or movement is restricted to essential information. Once the critical period is over, everything else can resume. This has been explored by teams in cardiothoracic operating. In contrast to aviation, there is generally no discrete period, such as takeoff or landing, during a surgical operation; therefore, to successfully implement a sterile cockpit protocol, the OR team would have to identify high risk or high mental workload tasks, or critical time intervals [24]. To achieve this, communication and training are essential.
Team huddles, briefings, and debriefs are opportunities to look at one or multiple cases for a particular team and have a discussion as a unit. The huddle can be used in the OR to improve situational awareness, to bring up potential issues or high-risk factors, to identified factors which could contribute to error or adverse outcomes, and to initiate actions for future procedures. A daily huddle has been shown to improve team satisfaction and decrease interruptions and delays [25].
Conclusion
Strengthening systems to support people working within the OR environment is essential if meaningful and sustainable improvements in outcomes and effectiveness are to be achieved. Transient, “feel good” fixes do little to address underlying causes of error and adverse outcomes, but are typically the focus of management because they are easy. Understanding SFD and the relationship to latent and active errors assists teams responsible for systems change to characterize, measure, and assess the existing work environment and proposed and implemented systems changes. The next decade is sure to witness dramatic advancements in systems engineering in complex healthcare delivery environments such as the OR. The application of artificial intelligence to this field will make error and outcome analysis, including analysis of SFD, more powerful than ever, resulting in systems that better support the people working to ensure patient outcomes are the best possible.
References
Definitions
Vocabulary is an important part of understanding research in operating room (OR) management [1]. Paradoxically, it is not possible to search the OR management literature successfully using PubMed without first knowing the corresponding vocabulary [1]. When people refer to “efficiency,” but what they really mean is “make more money,” then their search will not find the papers about what they truly care about [1]. Precise vocabulary matters, just like for drug names and anatomy [1].
Staff Scheduling and Assignment
Staff scheduling is the process of deciding which nurses, anesthesia providers, and other personnel work each shift on each day. Staff scheduling for a future date usually is performed before the surgical cases to be performed on that date have been scheduled.
For example, the OR nurses create their work schedule two months in advance. Ms. Jamison and Mr. Green are scheduled to work 8 a.m. to 4 p.m. on March 8.
Staffed hours are the hours into which cases are scheduled (e.g., 7 a.m. to 3 p.m.).
Staff assignment is the process of deciding who will take care of patients in a given location on a specific day. Most assignments are typically made on the workday prior to the date of surgery.
For example, tomorrow, Dr. Waters will be supervising two Certified Registered Nurse Anesthetists (CRNAs) in ORs 9 and 10.
Elective, Urgent, and Emergent Cases
We are not aware of any one best answer as to what constitutes an elective, urgent, or emergent case [2]. Still, differentiating among such cases is necessary to plan staffing. The following is a set of reasonable definitions that provide for different operational decisions.
An elective case can be defined as one for which the patients can wait at least 3 days for surgery without sustaining additional morbidity [3]. The choice of 3 days corresponds to patients waiting from Friday to Monday. At a facility with patients scheduled for elective surgery on Saturdays, two days would be used as the threshold.
For example, shoulder arthroplasty and breast biopsy cases are elective.
An emergent case can be defined as one for which the patient is likely to sustain additional morbidity and/or mortality unless the surgical care is started in less time than needed for a team to be called in from home. In this context, “likely” to have a worse outcome means based on scientific studies, such as published observational studies. Through the use of this evidence-based definition, the appropriateness of the relevant staffing decision can be made by reviewing data on prior emergent cases [3].
For example, cesarean sections because of a prolapsed umbilical cord or placental abruption are emergent.
An urgent case is defined as one for which the safe waiting time lies between that of an emergency and an elective case [3]. Almost all nonelective cases are urgent cases.
For example, appendectomy is an urgent case. The procedure is not an emergency, because the patient can wait long enough for an OR team to come to the hospital from home. A cadaveric renal transplant would also be urgent, not emergent for the same reason.
From a practical viewpoint, sufficient staff need to be present in-house to deal with emergent cases, whereas providing staff who take calls from home to cover urgent cases that cannot be covered by the in-house call team is a viable approach. The cost differential between having sufficient staff in-house to handle urgent cases vs. having them taking calls from home can be considerable [3, 4]. Planning sufficient staffing for the urgent cases is important. Surgical procedures for patients who are inpatients preoperatively generally represent urgent cases. Delaying surgery and thereby increasing hospital length of stay is counterproductive [5, 6].
Definitions Related to Service-Specific Staffing
Surgical service refers to a group of surgeons who share allocated OR time (i.e., service-specific staffing). An individual surgeon, a group, a specialty, or a department can function as a surgical service, depending on how OR time is allocated. Service simply refers to the unit of OR allocation.
For example, two ORs are allocated to the urological surgeons practicing at a hospital, any of whom can schedule cases in these rooms daily. Then, “Urology” is a service.
For example, two general surgeons are partners in General Surgical Associates (GSA), one of two independent general surgery groups that practice at a hospital. If these two general surgeons are together allocated OR time, then they represent a service, “GSA.”
For example, a busy orthopedic surgeon, Dr. Bones, is personally allocated 8 h of OR time every Tuesday. Then, from the perspective of allocating OR time and scheduling cases on Tuesdays, that “Dr. Bones” represents a surgical service, even though there may also be an “Ortho” service. The service-specific staffing is that one OR for 8 h every Tuesday.
Even when a surgical suite does not have a formal organizational plan for allocating ORs (i.e., “block schedule”), there can be service-specific staffing (i.e., OR allocations). In this regard, “services” need not be specific clinical subspecialties in the medical staff organizational structure. Rather, they reflect the activities of individuals or groups of surgeons who use the OR facilities and thus require organized staffing to support those activities. In some circumstances, several disparate subspecialties (e.g., “oral surgery” and “plastic surgery”) may share allocated OR time and thus function as a service.
For example, an 8-OR surgical suite has the official policy that all of its cases are scheduled on a first-scheduled, first-served basis. However, in reality, cases of the same specialty usually are scheduled into the same ORs, since this simplifies the distribution of resources (e.g., surgical equipment, video towers) and assignment of nursing and anesthesia teams who specialize in the area of surgical practice. Some nurses preferentially care for patients undergoing neurosurgery and otolaryngology cases, some mostly gynecology or general surgery, and so forth. In this case, the services correspond to the specialty teams.
Allocated OR time is the interval of OR time with a specified start and end time on a specified day of the week that is assigned by the facility to a surgical service for scheduling cases (Figure 8.1). Allocated OR time is effectively the same thing as staffed hours, but often the phrase “staffed hours” is used to refer to the sum of multiple services’ allocated OR time. Some facilities have OR time that is staffed and available for cases, but not allocated to a specific service. Such OR time has been allocated to a “pseudo-service,” variably named the “Open,” “Unblocked,” “First-Scheduled, First-Served,” or “Other” service. Knowing the allocated time for a service on a day of the week quantitatively describes the minimum anesthesia and nursing capacity for the surgeons. This is in contrast to “block time” that effectively predicts whether a surgeon will have many hours of cases on some future data.
Figure 8.1 An example of an OR schedule.
For example, Urology is allocated OR time in two rooms from 7 a.m. to 4 p.m. on Monday to Friday. This does not mean that the department’s surgeons are limited to scheduling cases only if they can be completed by 4 p.m. Instead, it means that staffing has been planned for the department’s surgeons between 7 a.m. and 4 p.m. The definition applies whether or not at that hospital it happens that the department’s surgeons actually finish by 4 p.m. If the urology service’s cases run past 4 p.m., and if nursing and anesthesia teams were to plan its staff scheduling to match the allocated OR time, then they would need to work beyond the end of regularly scheduled hours.
OR time of a case is defined as the interval from when a patient enters an OR until that patient leaves the OR (i.e., “wheels-in” to “wheels-out”). This definition is used often because these events are unequivocal and thus have good interrater reliability. The use of anesthesia information management systems to provide such data automatically can make OR management easier [7].
Turnover time is the time from when one patient exits an OR until the next patient on that day’s OR schedule enters the same OR [8, 9]. Separating turnover time from the OR time of a case permits the two to be studied statistically as separate processes (see “Impact of Reducing Surgical and Turnover Times”). Cleanup times and setup time characteristically are recorded separately from OR times, and then combined. In part, this is because it is hard to define when cleanup has ceased and setup has begun for the next case, and these activities may overlap. Turnover times include cleanup times and setup times, but should exclude planned or unplanned delays between cases (e.g., when a to-follow surgeon is given a scheduled 12:30 p.m. start time and the prior case in the OR ended at 11:00 a.m., or if the second of third cases in an OR cancels and the third patient is not available; see Figure 8.1). Hospital surgical suites may consider times between cases that are longer than a defined interval (e.g., 90 min) to represent delays, not turnovers, when computing turnover times to focus statistics on the latter cleanup and setup times [8]. This is because it is difficult to determine retrospectively the cause of such outliers, and these are usually unrelated to the process of room setup and cleanup. It is common for hospital analysts to ignore this distinction, resulting in invalid estimates of the actual turnover time and its variance.
For example, staffing is planned from 7 a.m. to 3 p.m. A patient arrived at the holding area at 7:45 a.m., her intravenous catheter was placed at 7:50 a.m., she entered the OR at 7:59 a.m., the trachea was intubated at 8:12 a.m., the operative site was prepared at 8:15 a.m., and the incision was made at 8:23 a.m. The patient left the OR at 10:59 a.m. From the perspective of OR scheduling, the case started at 7:59 a.m. The OR time of the case was 3 h.
For example, a surgeon is scheduled to perform a hepatic resection. However, soon after incision, the patient is found, unexpectedly, to have widespread peritoneal metastases and the incision is closed without performing the planned procedure. The patient exits from his OR 2.5 h earlier than planned. Including a planned 0.5-h turnover, the second case of the day could start 3 h earlier than planned. However, the second case of the day in that OR will be performed by a different surgeon. He is unavailable, caring for patients in his outpatient office. The result is a delay of 3 h. That delay should not contribute to the calculation of turnover times.
OR workload for a service is its total hours of cases including turnover times (Figure 8.1). This excludes the urgent cases for that service if separate OR time is allocated for urgent cases performed by all services. Turnover times are applied, by convention, to the service performing the prior case in the OR. Thus, there is no turnover time for the last case of the day in the OR.
Underutilized OR time = [allocated OR time] – [OR workload], or zero if this value is negative (Figure 8.1) [10]. This means that underutilized OR time equals the allocated OR time minus the OR workload, provided the allocated OR time is larger than the OR workload. Otherwise, the underutilized OR time is 0 h. Thus, underutilized OR time represents the time for which staffing was planned, but no work was performed. This equation applies only when the allocated OR time is calculated based on minimizing the inefficiency of use of OR time, defined several paragraphs below.
For example, staffing is planned from 8 a.m. to 4 p.m. An OR’s last case of the day ends at 2 p.m. The OR workload is 6 h. There are 2 h of underutilized OR time. The adjusted utilization is 75 percent, where 75% = 100% × (1–2 h / 8 h) [9]. The maximum value of “adjusted utilization” is 100 percent.
Overutilized OR time = [OR workload] – [allocated OR time], or zero if this value is negative (Figure 8.1) [10]. Thus, overutilized OR time represents the time during which work was performed, but staffing was not planned in advance. The adjusted utilization ignores overutilized OR time, which is one of the reasons why utilization is not a useful metric for facilities that typically do not finish the OR schedule prior to the end of allocated OR time. This equation applies only when the allocated OR time is calculated based on minimizing the inefficiency of use of OR time.
For example, an OR is staffed from 8:30 a.m. to 6 p.m. The last case of the day in the OR ends at 8 p.m. Then, there are 2 hours of overutilized OR time.
Inefficiency of use of OR time = [(cost per hour of underutilized OR time) × (hours of underutilized OR time)] + [(cost per hour of overutilized OR time) × (hours of overutilized OR time)] [10–12]. The cost of an hour of overutilized OR time is always more expensive than the cost of an hour of underutilized OR time. This equation, the one for underutilized OR time, and the one for overutilized OR time collectively are the three simultaneous equations defining OR efficiency.
OR efficiency is the value that is maximized when the inefficiency of use of OR time has been minimized [10]. “Efficiency” is characteristically thought of as the ratio of an output to the necessary input. For example, in a factory producing widgets, efficiency can be the number of widgets produced divided by the labor cost of producing those widgets. Thus, efficiency could be increased by producing more widgets with the same number of workers, or the same number of widgets with fewer workers. In the OR setting, when surgeons and patients are provided open access to OR time on any future workday, the output (e.g., number of cases performed) is a constant (see “Tactical versus Operational OR Management Decisions”). One cannot “manufacture” more cases on the day of surgery. Maximizing “efficiency” then is achieved by minimizing the input. That occurs when service-specific staffing and case scheduling are so good that there are both 0 h of underutilized OR time and 0 h of overutilized OR time. In practice, this is an unachievable goal, due to variance in surgical times for identically scheduled cases. Thus, optimizing OR efficiency requires a managerial objective to minimize the total cost of the underutilized and overutilized OR time.
For example, Vascular Surgery is allocated two ORs every Friday. Why was this decision made? If the department’s surgeons were allocated three ORs, then much of the OR time would be underutilized. That would reduce OR efficiency. If the department had been allocated one OR, then the surgeons would have been working late to finish their cases, resulting in much overutilized OR time. That would reduce OR efficiency. The choice of two ORs provided the best balance.
In our experience, this example provides what most facilities consider the objective of OR allocation, providing the right amount of OR time to get the cases done (i.e., not too much or too little). That is the essence of operational OR management decision making. This objective must be differentiated from the longer-term tactical stage of OR allocation, wherein an increase or reduction in allocated OR time is expected to result in a change in OR workload [13].
The example also shows why good operational decisions cannot be made based on OR utilization. True, OR allocation would not be three ORs, based on either OR efficiency or OR utilization, because there would be many hours of underutilized OR time. However, the choice of one or two ORs would not be clear based on OR utilization because the resulting hours of overutilized OR time are not included in the calculation of utilization (which, by definition, cannot be higher than 100%). In contrast, decision making based on OR efficiency considers both the expected underutilized and overutilized OR time. At facilities where there is substantial overutilized OR time, basing allocations on utilization statistics will have a large negative impact on OR efficiency.
For example, Dr. Sato is an orthopedic surgeon in a solo practice. She is allocated OR 1 on Mondays and Wednesdays for 10 h, from 7 a.m. to 5 p.m. Dr. Ho is another orthopedic surgeon in a solo practice. He is allocated OR 1 on Tuesdays and Thursdays for 10 h, also from 7 a.m. to 5 p.m. Both Drs. Sato and Ho consistently perform slightly less than 10 h of cases in their allocated OR time, virtually never more. Both perform spinal surgery cases. However, Dr. Sato tends to perform one more case of the same type as Dr. Ho within the allocated OR time because Dr. Sato operates much more quickly than Dr. Ho. OR efficiency is identical and unaffected by how quickly Dr. Sato operates (see continuation of the example at the end of this section).
Managerial Cost Accounting
Labor cost equals the sum of two products: staff scheduled hours multiplied by the cost per hour of staff scheduled hours and hours worked late multiplied by the cost per hour of hours worked late [14, 15]. More complicated managerial accounting models generally are not needed for purposes of OR allocation and case scheduling. Labor cost can generally be estimated as the sum of the allocated OR time multiplied by the cost per hour of staffed hours and the hours of overutilized OR time multiplied by the cost per hour of overutilized OR time.
OR productivity equals the OR workload divided by the labor costs [14].
For example, the only anesthesia service that a group provides at an outpatient surgery facility is OR anesthesia. Staffing is planned for five ORs from 7 a.m. to 5 p.m. There is virtually never any overutilized OR time. Then, each increase in OR workload (i.e., the cases performed) results in an increase in OR and anesthesia group productivity.
For example, a hospital has substantial underutilized OR time and substantial overutilized OR time. Recent increases in elective OR workload have resulted in cases finishing in the early evenings, resulting in increased overutilized OR time. The increase in OR workload could be reducing OR productivity. That would be happening if the cost per hour of overutilized OR time is much higher than the cost per hour of regularly staffed hours.
Although it may seem good to make operational OR management decisions based on increasing OR productivity, we recommend against the approach. Instead, make operational OR management decisions to maximize OR efficiency. Usually the decisions will be the same, but not always [14].
We recommend decision making based on OR efficiency, for two reasons [2, 16]. First, whereas decisions based on OR efficiency are invariant to the perspective of the cost assessment, decisions based on labor cost are not. There is no one best answer as to whose labor costs should be used to make decisions. For example, although from the perspective of the anesthesia group, the ideal would be to make the decisions based on its labor costs, other reasonable options include the labor cost of the hospital or society. Second, labor costs vary depending on staff scheduling and staff assignment, whereas OR efficiency does not. If labor costs were used, distributed decision making would no longer be consistent depending on the perspective of who makes the decision. For example, if one CRNA works overtime to cover for another CRNA who has called in sick, that would affect decisions based on labor costs but would not affect decisions based on OR efficiency.
Revenue is the money received from third parties in return for having provided care for a specific patient.
Variable costs are costs that increase proportionate to the volume of patients receiving care [17].
For example, the amount of anesthetic medications used will vary with the number of patients who receive anesthesia care. Hence, pharmacy costs are variable costs.
Fixed costs are those costs that are not related to the volume of patients receiving care.
For example, the surgical tables cost the same regardless of how often they are used. Surgical tables are a fixed cost. For example, a new eight-OR ambulatory surgery center has virtually no overutilized OR time but considerable underutilized OR time. On a short-term basis, labor costs can be viewed as fixed. Even if OR workload increased moderately, all the cases would still be completed within allocated OR time. The number of OR nurses needed to staff the ORs would be unchanged. However, on a longer-term basis, labor costs could be reduced by closing an OR if the OR workload does not increase sufficiently.
Contribution margin equals revenue minus the variable costs for providing care to those patients. These include revenue and variable costs associated both with the current case and those related to subsequent care due to complications.
For example, consider the calculation of contribution margin for a colon resection in which the wound becomes infected. Revenue and variable costs need to be included due to the original surgery as well as the full hospitalization including three trips back to the OR to wash out the wound.
Profit equals revenue minus the sum of fixed and variable costs. This is the same as contribution margin minus fixed costs.
For example, let us return to the orthopedic surgeons, Drs. Sato and Ho, who were described previously. Dr. Sato performs one extra spinal surgery case in the same number of hours of OR time than does Dr. Ho. For the anesthesia group, Dr. Sato is more profitable than is Dr. Ho, because the anesthesia group gets more revenue for the same fixed costs of staffing the OR. However, the implants that Dr. Sato chooses cost 80 percent of the revenue while those that Dr. Ho chooses cost 50 percent of the revenue. Thus, for the hospital, Dr. Sato is less profitable than Dr. Ho [18]. Both still have a positive contribution margin, but only slightly so for Dr. Sato.
OR Efficiency on the Day of Surgery
OR efficiency is maximized by choosing staffing and scheduling cases to minimize the inefficiency of use of OR time, the latter being the [(cost per hour of underutilized OR time) × (hours of underutilized OR time)] + [(cost per hour of overutilized OR time) × (hours of overutilized OR time)]. If one considers the cost of one hour of overutilized time to one hour of underutilized time to be the fixed ratio, R (typically 1.5–2.0), the value to be minimized can be expressed in terms of hours: (hours of underutilized OR time) + R × (hours of overutilized OR time). This relationship is further simplified on the day of surgery.
At most surgical facilities, OR nurses are full-time hourly or salaried employees. Thus, on the day of surgery, the increment in nursing labor cost from 1 h of underutilized OR time is negligible relative to the cost from 1 h of overutilized OR time. Finishing cases early, but still before the end of staffed hours, reduces labor costs negligibly versus the labor cost that would result from a reduction in overutilized OR time. The same applies to CRNAs and/or anesthesiologists who are employees of the surgical facility or corresponding anesthesia group.
Few anesthesiologists and CRNAs in private practice can earn enough money to cover the cost of their salary plus benefits unless they are scheduled to care for whatever patients may need urgent surgery (i.e., who are inpatient preoperatively [5, 6]), along with patients having elective, scheduled surgery. Thus, the incremental revenue lost on the day of surgery by having 1 h of underutilized OR time is negligible relative to the indirect/intangible costs from working late unexpectedly (i.e., the opportunity cost of being idle is effectively zero) [19, 20].
Consequently, on the day of surgery, the cost per hour of underutilized OR time is negligible relative to the cost per hour of overutilized OR time [2, 21]. Thus, on the day of surgery, minimizing the inefficiency of use of OR time (see “Definitions”) requires only that management minimize the hours of overutilized OR time, since the cost per hour of this time is a constant [2, 21]. As explained below, “minimizing” on the day of surgery includes case and staff assignment decisions, as all cases are performed unless patient safety would be affected.
Case scheduling to maximize OR efficiency minimizes hours of overutilized OR time, as previously reported for surgical suites [22]. The following two scenarios illustrate the implications of the results.
For example, an anesthesiologist is assigned to an OR staffed from 7 a.m. to 3 p.m., but with one expected hour of overutilized OR time. The anesthesiologist works quickly. She places every intravenous catheter and arterial cannula on the first attempt and performs a fiber optic intubation in 10 min. Because of her rapid work, the cases finish at 3 p.m., preventing 1 h of overutilized OR time. Thus, the anesthesiologist increased OR efficiency [21].
A different anesthesiologist is assigned to another OR staffed from 7 a.m. to 3 p.m., but with 7 h of scheduled cases. The anesthesiologist works equally quickly, resulting in cases finishing at 2 p.m. instead of at 3 p.m. Because overutilized OR time was not reduced, the anesthesiologist did not increase OR efficiency [21].
These scenarios show that “working fast” is not synonymous with increasing OR efficiency. The last scenario of the preceding section showed that working fast is not synonymous with maximizing profit, either. Analogously, “working slowly” is not synonymous with decreasing OR efficiency. Sometimes, “working fast” may increase OR efficiency, and “working slowly” may decrease OR efficiency [2, 23]. But this will be entirely dependent on the circumstances.
For example, a different anesthesiologist is supervising resident physicians in two ORs. Staffing is planned from 8 a.m. to 4 p.m. The anesthesiologist needs to decide which of the two ORs to start first. One OR is scheduled with two cases from 8 a.m. to 6 p.m., the other with five cases from 8 a.m. to 3 p.m. To maximize OR efficiency, the anesthesiologist should first start the OR expected to have 2 h of overutilized OR time [21].
By following this simple principle, individual and collective decision making can be closely linked to enhancing OR efficiency. Without understanding the principles of OR efficiency, the anesthesiologist is likely to have made the opposite decision because there are more cases in the other OR.
The same principles and use of scenarios can be applied to housekeepers, OR nurses, managers, postanesthesia care unit nurses, etc. [2, 7, 12]. In essence, all decision making on the day of surgery that has “improving efficiency” as the goal revolves around this concept of reducing overutilized time. Again, working faster per se does not increase OR efficiency; rather, OR efficiency is increased only when working faster reduces overutilized time.
For example, staffing is planned from 7 a.m. to 3 p.m. Recently the hospital hired a new OR nurse. On Monday, she assisted in OR 12, resulting in cases finishing at 2 p.m. instead of 3 p.m. On Tuesday, she assisted in OR 14, resulting in cases finishing at 4 p.m. instead of 5 p.m. She increased OR efficiency more on Tuesday than Monday, because reducing 1 h of overutilized OR time increases OR efficiency more than does reducing 1 h of underutilized OR time.
Tactical versus Operational OR Management Decisions
Consider a common OR management problem: staffing is planned from 7 a.m. to 3 p.m. A surgeon has been allocated 8 h of OR time every Wednesday for years, and the hospital has an “official” policy that elective cases may only be scheduled into allocated time. The surgeon has always underestimated the OR times of his cases in order to bypass this constraint. He has never finished before 6 p.m. and usually ends between 7 and 8 p.m.
The anesthesiologists and OR nurses may complain about working late every Monday because the surgeon is being allowed to “overbook” his schedule. They may lobby to have a committee meet to rectify the situation. Simultaneously, the administrators may discuss the surgeons’ lack of respect for rules and hospital resources. Nevertheless, physicians who refer their patients to the surgeon reward him by continuing to send him work because their patients are pleased with his expeditious service.
The fundamental issue is the surgeon’s frequent misrepresentation of the estimated OR times of his cases, in order to get them onto the OR schedule [24]. The merits of the tactical issue (i.e., whether this is overall good or bad practice) have little relevance to OR productivity. The relevant operational decision is clear: managers should change staffing to match the reality of the existing workload [15]. Doing so neither increases nor reduces OR capacity or convenience for the surgeon and his or her patients. What it does is to reduce labor costs by reducing the hours worked late, since staff scheduling is adjusted based on staffed hours [14]. From the surgeon’s perspective, the only thing that will change is that he can provide more realistic estimated OR times, since there will no longer be a need to “adjust” the times in order to get the cases running past the end of the regular workday on the schedule. From the perspective of the anesthesiologists and the OR nurses, complaints about working “late” will disappear, as the regular hours in the surgeon’s OR now extend to 12 h, and staff working in that OR can expect to work for this period of time.
For example, when an anesthesiologist was hired, the job description said that work hours were 7 a.m. to 5 p.m., and he accepted a salary based on this assumption. Yet, every Wednesday for the past 5 years, the anesthesiologist has finished working between 7 p.m. and 8 p.m. Staffing is subsequently changed to be to 8 p.m. because that is the reality of the existing OR workload. Planning the staffing to 8 p.m. does not change the workload. Rather, it results in the work being planned long in advance.
In the two preceding scenarios, the surgeon and patient are choosing the day of surgery. Cases are not being turned away, provided they can be done safely, even if they will likely be performed in overutilized OR time [25]. Subject to that priority, OR time can be allocated based on maximizing OR efficiency. To describe operational reality, mathematics needs to be based on the surgeon and patient having open access to OR time on the workday of their choosing.
For example, all ORs are allocated at a hospital for 8 h. The adjusted utilizations range from 75 to85 percent among the surgical services. Thus, there is essentially only underutilized OR time. At this hospital, allocating OR time based on OR efficiency would give precisely the same result as allocating OR time based on adjusted utilization. This is because virtually no OR ever finishes late. A zero has been substituted for the hours of overutilized OR time in the equation for the inefficiency of use of OR time (see “Inefficiency of Use of OR Time”). The surgeons can be considered to have open access to OR time on the workday of their choosing, and they have chosen to perform cases only when they can be completed within allocated hours.
The two preceding scenarios demonstrate that service-specific staffing can be considered for any facility when decisions are made based on OR efficiency and on surgeon and patient open access to OR time on any future workday. The next scenario shows that the assumption of fixed hours applies only to a minority of surgical suites [11].
For example, an ambulatory surgical center has a policy that OR time is allocated based on OR utilization. Staffing is planned from 7 a.m. to 3 p.m. This policy is enforced strictly. A surgeon asks to book a case to start at 1 p.m., with an expected (realistic) OR time for the case of 2.5 h. He is told “No,” that would be unacceptable, because the case will likely end at 3:30 p.m.
The preceding scenario will seem unreal to most clinicians in the United States. That is the point. Only scheduling cases if they can reasonably be expected to finish by the end of allocated OR time is not the reality of short-term operational decision making at many facilities. Although considering a facility to have fixed hours of OR time is an accurate and practical model from a tactical perspective, it is not realistic for day-to-day decision making for all surgeons at a surgical suite [2, 25–27].
We return to the first scenario of this section, which describes persistent overutilized OR time. Should the surgeon be encouraged to continue to schedule cases beyond the hours that have been allocated? That is a reasonable tactical question, which includes consideration of the financial impact of the surgeon’s cases versus the long-term effects on hiring and retention of OR nurses and anesthesia providers [13]. The tactical decision can, and probably should, be considered from multiple perspectives, including societal. However, the operational decision making focuses on the reality of the existing workload. Operational decisions, specifically service-specific staffing, are what most managers can control.
Truly not having fixed hours of OR time, despite an official policy against overbooking elective cases, is particularly common at hospitals at which surgeons mischaracterize cases as “urgent” to get them onto the OR schedule.
For example, an academic department is allocated three ORs from 8 a.m. to 4 p.m. on all workdays. No elective case is scheduled unless it will fit into the 8 h based on mean historical OR time data from the OR information system. The service schedules 20 percent of its OR hours as urgent cases. The patients are inpatient preoperatively [5, 6]. Many of these patients likely could have waited safely for several days for surgery. Thus, these were elective cases. However, having the cases wait for surgery would be counterproductive economically. The surgeons reasonably called the cases “urgent” to achieve open access to OR time and thus bypass the policy against overbooking. OR efficiency would have been greater had more OR time been allocated originally. This would have allowed the cases to be performed in allocated, rather than in overutilized, OR time.
Suppose that on a long-term (tactical) basis, the behavior of the academic surgeons was considered so bad that penalties were applied. Then, there would be very little overutilized OR time. The methods described in this chapter would be valid and appropriate, but not necessarily a useful improvement. Consequently, there is reason to consider whether the behavior of the above surgeons is inherently bad.
From the societal, hospital, and surgeons’ perspective, likely the behavior is good, or at least not bad enough to penalize the surgeons. They are serving as their patients’ advocates, assuring timely surgery. Among patients who are inpatient preoperatively, the surgeons are reducing hospital lengths of stay. Among patients who are outpatient preoperatively, most patients only have two preoperative visits with the surgeon, making surgeon flexibility to schedule initial consultations very important to growth in surgical practices [28]. Further, in some healthcare systems, including that in the United States, the more cases that the surgeons perform, the higher are hospital and physician contribution margins.
Hospitals receiving fee-for-service reimbursement achieve an overall positive contribution margin for the elective cases of almost all surgeons, [13, 26, 29] because a large percentage of OR costs are fixed (e.g., surgical robots, video equipment for minimally invasive surgical suites, and anesthesia machines). If professional revenues for the anesthesia providers and surgeons were also considered in the calculation of contribution margin, then every surgeon would provide an overall positive contribution margin for his or her elective cases. The implication, then, is that if a case can be performed safely, it is economically irrational not to perform the case [13, 26, 30].
The rationale for providing surgeons with open access to OR time, provided a case can be performed safely, makes particular sense for hospitals with ICUs that often are full. For patients needing such care, the ICU is a frequent bottleneck that results in delays or cancellations of surgical cases. There are two ways to approach this problem, other than simply providing and staffing more ICU beds.
One strategy to reduce the risk of delays or cancellations is to adjust the days that services are scheduled to perform surgery [31, 32]. Although such techniques can be implemented practically [31, 32], the incremental benefit to hospitals may be small. If most surgeons schedule patients for ICU admission on the same days of the week, usually the cause of case cancellations is visible to the surgeons. The surgeons generally suffer more, financially, from case cancellations and delays than do hospitals and anesthesiologists. In this situation, the hands-on facilitation of a local OR manager or an expert in managing organizational conflict can help, with tabular and graphical summaries of the impact of decisions on cancellations [32]. Such interventions are valuable and important [33]. However, they are not commonly decisions made by anesthesia group managers or OR nursing directors, although they can facilitate such processes.
The second of the two strategies is to provide surgeons with flexibility on the days when they have OR time. Cases should get onto the OR schedule to assure that the expensive bottleneck (the ICU) is always full. For example, although 90 percent of patients may have ICU lengths of stay <2 days following coronary artery bypass graft, there can be marked variability in length of stay [32, 34, 35]. Consequently, predictions can be inaccurate for the number of open ICU beds available daily as a result of patient transfers from the unit. When the bottleneck to doing surgery is downstream from ORs and the service time for that downstream process is highly variable, then flexibility in scheduling the OR cases is needed to maximize throughput. This does present some inconvenience to surgeons and patients in that they do not know with certainty the date when the procedure will be performed until very close to the day of surgery, but is preferable to having the case cancelled on the day of surgery due to inadequate ICU resources.
The same logic applies to expensive capital equipment (e.g., intraoperative magnetic resonance imaging), that, like the ICUs, is a fixed cost that is best kept as fully utilized as possible. In the future, more ORs will include more technologically advanced equipment, resulting in even higher capital costs. The percentage of hospital costs for surgery that are attributed to labor likely will decrease as capital costs increase to support these and other expensive technologies. To maximize use of that equipment, surgeons should have open access to OR time to do a case on whatever future workday they are available, provided the case can be performed safely using existing equipment. For example, if two surgical services have allocated time on the same day of the week and are vying for the one operative robot, providing the services the ability to book elective cases on days other than on the date of their surgical block will increase the utilization of this expensive resource.
The caveat of allowing open access to OR time “provided the case can be performed safely” is of strikingly large importance. Safety includes access not only to specialized surgical equipment, but also to limited ICU beds, hospital ward beds, postanesthesia care unit beds, nonfatigued staff, etc. What can be done safely limits how much work can be done in a surgical suite on any given day [2, 12, 13, 32, 36]. Characteristically, tactical decision making limits what can be done safely. Then, operational decision making functions within these boundaries.
Based on these arguments, realistic operational decision making needs to function within a structure that allows the surgeon and patient to choose the day of surgery. The reason why this is so important is that surgeons are not the individuals primarily responsible for OR efficiency through their filling of the OR time allocated to them. Rather, the parties primarily responsible for OR efficiency are the nursing and anesthesia group managers who choose the OR allocations to match staffing to the surgeons’ workloads. The latter refers almost entirely to the durations of the hours in each OR into which cases are scheduled, numbers of ORs usually only being numbers of flexible rooms to facilitate turnovers and urgent cases, etc.
For example, for 1 week each year, most of the otolaryngologists are away at a conference. There is substantial underutilized OR time, resulting in poor OR efficiency. This is an example of poor OR management. The managers should have increased OR efficiency by adjusting staffing to match the surgeons’ and patients’ hours (e.g., by encouraging months in advance for some nurses and anesthesia providers specializing in this area of care to use some of their accrued vacation).
Planning Service-Specific Staffing and Scheduling Cases Based on Increasing OR Efficiency
Allocating OR time (i.e., planning service-specific staffing) and scheduling cases based on OR efficiency can increase OR productivity by reducing labor costs.
Performing Calculations Using Complete Enumeration
In practice, OR allocations that are calculated based on OR efficiency are done by service and day of the week. That is because day of the week is the best predictor of a service’s workload [11, 37]. Calculating an OR allocation means determining how many ORs should be staffed daily for each service and, for each of these ORs, how many hours of staffing should be planned (e.g., 8, 10, or 13 h) [11, 37]. Calculations of optimal allocations can be done by complete enumeration [23]. Specifically, all possible staffing solutions are considered, starting with 0 h and progressively increasing staffed hours until additional increases in the staffed hours cause the efficiency of use of OR time to decrease for that service [37]. If shifts of 8, 10, and 13 h are considered, then the successive choices are 0, 8, 10, 13, 16, 18 h, etc. Increasing the staffed hours causes the efficiency of use of OR time to increase progressively to a maximum, after which it decreases [11]. The complete enumeration can be constructed such that every series of cases performed by the same surgeon on the same day would be performed in its original sequence and take the same amount of OR time [23]. The only change is in the start times.
For example, a surgeon is currently allocated 8 h of OR time individually on Tuesdays. The surgeon historically has done 9 h of cases every Tuesday. The hospital calculates that the expense of one hour of overutilized time is twice that of one hour of underutilized time; inefficiency is expressed in terms of the number of equivalent underutilized hours. Candidate allocations are 0, 8, 10, and 13 h. The inefficiency of use of OR time for each potential allocation is determined from the cost of the underutilized and overutilized hours that would have resulted. A 0-h allocation (A) would have resulted in 9 h of overutilized time, with an inefficiency of use of OR time proportional to 18 h. An 8-h allocation (B) would have resulted in 1 h of overutilized time, with an inefficiency proportional to 2 h. A 10-h allocation (C) would have resulted in one underutilized hour with an inefficiency proportional to 1 h. Finally, a 13-h allocation (D) would have resulted in 4 h of underutilized time with an inefficiency proportional to 4 h. Since the most efficient solution (i.e., smallest value of the inefficiency of use of OR time) was Allocation C, the surgeon should have been allocated 10 h of OR time in order to maximize the efficiency of use of OR time.
There is a unique solution to the choice of the OR allocation that will maximize OR efficiency if OR allocations can be of any duration (e.g., 9.27 h [11]), but not necessarily when fixed choices (e.g., 8, 10, 13 h) are considered. When two choices provide nearly the same inefficiency of use of OR time, the OR workload can be reviewed to consider which most closely matches how the surgeons in the service have historically been using their OR time.
For example, the cardiac surgeons perform an average of 14 h of cases each Tuesday, with a range of 12–15 h. Forecasted OR efficiency would be nearly identical whether 13 h of OR time were allocated in one OR or 8 h in each of two ORs. The cardiac surgeons have had two ORs (i.e., reliable first case of the day start times) for the past 6 years. They have consistently scheduled cases into those ORs such that there is only underutilized OR time, not overutilized OR time. Two ORs would be the most reasonable choice. In this example, planning OR allocation based on OR efficiency versus adjusted utilization results in the same decisions.
Maximizing OR efficiency is the same as minimizing the sum of underutilized hours and overutilized hours multiplied by the relative cost of overutilized to underutilized OR hours (see “Definitions” [11]). Thus, only the relative cost of overutilized to underutilized OR hours needs to be known, not the costs per se [11]. A commonly used [37] value for this ratio of costs is 1.75. This includes the direct costs of overtime at “time and a half” (1.50) and an increment (0.25) for indirect (intangible) costs of employee dissatisfaction, resignation, and recruitment and training [37]. Because of the marked effect of limiting consideration to common staff schedules (e.g., 8 or 13 h), the resulting inefficiency in use of OR time is characteristically highly insensitive to local experts’ uncertainty in the choice of the value of this parameter [38].
For example, on three Wednesdays, a service performed 12, 7, and 15 h of cases, including turnover times. There are 8-h shifts, with overtime scheduled by rotation using a late list. The relative cost of overutilized to underutilized hours is considered 1.75. If the service were allocated 8 h of OR time each Wednesday, then the cost of the inefficiency of use of OR time would be proportional to 20.25 h, where 20.25 h = (0 underutilized + 1 underutilized + 0 underutilized + 1.75 × [4 overutilized + 0 overutilized + 7 overutilized]). If the allocation were two 8-h ORs each Wednesday, the cost would be proportional to 14 h, where 14 h = (4 underutilized + 9 underutilized + 1 underutilized). If the allocation were three 8-h ORs each Wednesday, the cost would be proportional to 38 h, where 38 h = (12 underutilized + 17 underutilized + 9 underutilized). Therefore, the service should be allocated two 8-h ORs to maximize OR efficiency.
There is only one answer to the question, “How close are current OR allocations to those that would maximize OR efficiency?” In contrast, there is no one answer to the question, “How close are current OR allocations to those that are optimal based on OR utilization?” The reason is that there is then the subsequent question of how to determine the optimal OR utilization. The best OR utilization varies among services because it is sensitive to many parameters, such as staffed hours, turnover times, day-to-day variability in OR workload, statistical distribution of OR times of cases, and so forth [25, 39]. Years of data can be required to estimate these parameter values sufficiently accurately to use them to decide on the OR utilization to use as the service’s goal [40]. Allocating OR time based on OR efficiency simultaneously takes into account all of these issues. When a manager says “We allocate OR time based on OR efficiency,” that is close to a sufficient statement to describe precisely what happens in practice because the choice of the relative cost of overutilized to underutilized OR time is invariably close to 1.75 and insensitive to any differences. In contrast, when a manager says “We allocate OR time based on OR utilization,” that alone says virtually nothing about what happens in practice at the surgical suite.
Calculated Staffing (OR Allocations) Differ from Those in Current Practice
OR managers’ efforts to reduce labor costs must focus predominantly on OR allocation and case scheduling, because almost all of anesthesia providers’ costs are labor costs. The viability of a surgical facility depends on the economics of the anesthesia providers. For 11 of 12 facilities studied, allocating OR time based on OR efficiency achieved significantly lower labor costs than the plans that were being used by the local managers [37, 41–43]. For 9 of the 11 facilities, the statistical method approach resulted in plans that reduced labor costs by at least 10 percent [37, 41–43]. The percentage increases in OR efficiency were, by definition, even more.
A common anecdote reveals how poorly many facilities plan service-specific staffing. Often OR nurses and anesthesia providers report that every OR finishes at least an hour or two late every day. To consider the irrationality [44] of the situation, suppose that the relative cost of overutilized to underutilized OR time were 2.0. Then, it would be twice as expensive to finish late versus early. With appropriate OR allocations, the odds for each service and OR to finish early should be approximately two chances in three. That is, if staffing decisions were made rationally, a given OR would finish early on 2 of every 3 days.
In practice, percentage reductions in labor costs are not proportional to the number of ORs [37, 41]. Even at facilities for which each allocation is for one room, but either for 8 or 10 h, savings are found [45]. Surgical suites at which many hours of OR time are allocated to services do not have the largest percentage improvements from applying the operations research to OR management. The explanation for this observation is that the principal challenge faced by managers is not the number of ORs to be allocated to services, but how to manage variability in OR workload from week to week. The fact that the OR allocation decision is stochastic is the conceptual problem in the practicing managers’ decisions. The poor decisions are caused by cognitive biases that are observed for such decisions in other industries [44]. Implementation of improved decisions is not achieved by education alone, but rather by automating reliance on decision support software [44, 46].
For example, consider a service with OR workload averaging 6.5 h every Friday [47, 48]. Because there are no overutilized hours, allocation based on OR efficiency is identical to allocation based on OR utilization [45]. Once this principle is understood by managers, analysis is unneeded in the future. One analysis is sufficient. In contrast, suppose that the same facility has three of its eight ORs as unblocked, open, first-come, first-served “other” time. The surgical suite staffs in 8-, 10-, and 13-h shifts. Then, those three ORs could be allocated as 8/8/8, 8/8/10, 8/10/10, 10/10/10, 8/8/13, 8/13/13/, 13/13/13, 8/10/13, 10/10/13, and 10/13/13. Intuition will not help with this complex decision. The value of education is by increasing trust in relying on the statistical results [49, 50].
Hospitals generally should not artificially treat all ORs as having the same number of hours of cases [15]. The standard deviations among ORs in the daily total hours of cases including turnover times between 7:00 a.m. and 11:00 p.m. were estimated for 34 hospitals [51]. The hospitals did not have all ORs fully packed with the same end of the workday [12]. Many hospitals had standard deviations greater than 3 h [51]. Some ORs had underutilized OR time and other ORs had overutilized OR time; some ORs had 8 h allocations and other ORs had longer hours of allocated time; and/or some ORs finished before the end of an 8 h workday and other ORs were used for urgent cases, not finishing until late in the evening [2, 12].
Urgent Cases
Some hospitals have one or more ORs allocated for urgent cases during the regular workday. Typically, then, the appropriate number of ORs is chosen for such urgent cases by considering them to be performed by a pseudo-service, the “Urgent” service. The methods above are applied. At facilities not planning an OR for urgent cases, when calculating OR allocations for elective cases, each urgent case should be attributed to its surgical service.
The relative cost of overutilized to underutilized OR time may be appropriately higher for the urgent service than the other elective services, because the choice affects not just how often staff work late, but also patient waiting time for urgent surgery. However, urgent cases often cannot start immediately (e.g., because the surgeon is not available), such that overutilized OR time would occur regardless of calculations. In practice, the use of the same relative cost for overutilized to underutilized OR time as above (e.g., a factor of 1.75) can provide answers that clinicians consider reasonable.
Amount of Data Required for Calculations
To assess how much data are required to produce acceptable results, a long series of data from a surgical suite was divided into training and testing datasets, with different training periods [16]. The complete enumeration was applied to the training data, and the expected labor costs that would have occurred during the subsequent testing period were calculated. Each increase in the number of months of data up to 9 months resulted in a statistically significant reduction in expected labor costs. There were large incremental benefits in using at least 7 months of data. For the studied hospital, there was no advantage to using more than 1 year of data.
The minimum amount of data needed for calculating OR allocations based on OR efficiency can be particularly important to managers at facilities purchasing a new OR information system, anesthesia information system, or anesthesia billing system. The minimum period of data indicates the time from installation of the system to when management changes based on resulting data can be implemented. Application of the statistical methods using as little as 30 workdays of system data provided better OR allocations to reduce labor costs than OR allocations established by the practicing managers with years of data [16].
Sources of Data
Data for analysis can come from an OR information system, an electronic anesthesia information management system, or anesthesia billing data [52]. OR information system data have the advantage of virtually always having necessary data fields completed. Anesthesia billing data have the advantage of accuracy, because billing errors can be costly or even lead to challenges of fraudulent behavior. When using anesthesia billing data, if the OR in which the case was performed is not available from the data, then the anesthesia provider (i.e., the person in the OR delivering anesthesia care, not the supervising anesthesiologists) can be substituted for the OR field to calculate turnovers. Depending on the workflow among ORs, this substitution can be preferable.
Facilities with OR information systems that do not have data review at the time of data entry often have datasets that contain errors or omissions, including lack of knowledge of the actual ORs in which some cases were performed. This can occur if cases are moved during the day without correcting the corresponding information systems, or if the times of OR entry or exit are incorrectly entered. This manifests as the false appearance of two cases overlapping in the same OR at the same time. The typical fix is to change the recorded OR of each case that overlaps to a unique unknown OR. For example, suppose that one case is listed as being performed in OR 1 from 10 a.m. to 11 a.m. and another in OR 1 from 10:30 a.m. to 12 noon. Among all cases in the dataset, the latter case is the 139th for which the true OR is unknown. The second case can be considered to have been completed in the fictitious room “Unknown139.” Making such a change affects calculated turnover times, since some turnovers between cases will be altered, and thus may affect OR allocations. Nonetheless, studies demonstrated that the impact of this adjustment on the labor costs that result from poor OR allocations is of negligible importance, for three reasons [53, 54]. First, OR allocations are based on each service’s total hours of cases, a large number, plus total hours of turnover times, a much smaller number. Second, for cases in an OR that have a preceding case and a following case, two turnover times are lost. Yet, the turnover time between the remaining cases is increased between the two cases surrounding the reassigned case to the default maximum turnover time [53]. Third, the effect of allocating OR time only in fixed increments (e.g., 8 or 13 h) is of larger importance.
Assessing Trends, Seasonal Variation, and Data Errors
Use of complete enumeration assumes that there are no systematic differences among weeks in the expected OR workload (i.e., there are no trends or seasonal variation) [37]. National survey data show that these assumptions will hold for most facilities [55]. Raw data were reanalyzed from the 1994 to 1996 National Survey of Ambulatory Surgery. As a positive control, to assure that seasonal variation could be detected if present, the average number of myringotomy tubes inserted each day in ambulatory surgery centers of the United States was examined. As expected, myringotomy tube insertions peaked each winter, corresponding to the peak incidence of middle ear infections. Specifically, the average number of tubes inserted each day varied systematically among months for all 26 of the overlapping 11-month periods in the 36 months of the survey. In contrast, the average number of ambulatory surgery cases performed with an anesthesia provider each day in the United States per 10,000 persons was found not to vary systematically month to month on an 11-month basis.
Good routine practice is to test for statistically significant trends [56] or seasonality, to confirm that analysis is reasonable for each surgical suite. For example, the so-called runs test can be applied to the total labor cost over each consecutive 4-week period [37, 57]. Calculate the total labor cost for each 4-week period. Subtract the median from each value. Delete zero differences. Assign a “+” to positive differences and a “−” to negative differences. A “run” is defined as a series of one or more consecutive values that are the same. Finally, compare the number of runs of +’s and −’s to a critical value from appropriate statistical tables. For example, if over 10 weeks the values were + − 0 − − − + − + + there would be 5 runs (2 +, 1 + +, 1 − − −, and 2 −). At P <0.05, the expected number of runs is between 2 and 10, so the null hypothesis that there is a trend would be rejected. This test, the Wald–Wolfowitz one-sample runs test, is available in most statistics packages.
In our experience, it is almost never necessary to incorporate methods appropriate for data with trends and seasonality into the analysis [12].When the runs test detects trends or seasonality, characteristically this reflects a problem with the data or special conditions [56] that need to be modeled separately. For example, if a hospital opens a new three-room endovascular (interventional) suite in the middle of the data collection period, this may result in a positive trend in OR workload. Opening of a new surgery center may result in an abrupt decline in workload at the main facility [56].
In addition to using the runs test, plot each service’s OR workload for the days of the week when the service is allocated OR time. The graphs are helpful to detect unrecognized errors in the data. For example, plotting OR workload for a service against time can show if a service had no cases listed for a day of the week for some part of the data period being used. This usually occurs when the data sent for analysis include one or more surgeons who recently left the facility and operated on the empty days.
Finally, look for the presence of many zero values in the histogram of OR workload for each combination of the day of the week and the service allocated OR time. This usually happens when the service’s scheduling is characteristic of an individual surgeon rather than a group of surgeons. These “holes” often represent times when an individual surgeon is away (e.g., on vacation). These can be hard to identify in a graph of OR workload versus time. Such services may need to have their allocations of OR time combined with another service to achieve reliable staffing predictions.
Services with Low OR Workloads
Provided cases are scheduled sequentially into ORs, then services with average OR workloads that are consistently <8 h have no overutilized hours. Allocating OR time based on adjusted utilization does not differ from doing so based on OR efficiency. Many facilities appropriately apply a minimum adjusted utilization for OR allocations [58, 59]. For example, based on the relative cost ratio of 1.75 described above, if services’ workloads were always the same each workday, then the optimal (minimum) value would be 68 percent [58, 59].
For example, a service’s OR workload averages 6 h every Tuesday. The facility bases its decisions on the efficiency of use of OR time. The service’s adjusted utilization is 75 percent. Thus, the service is allocated a single OR for 8 h. Because there are no overutilized hours, allocation based on OR efficiency is identical to allocation based on OR utilization. There are 0 h of underutilized time caused by OR allocation and case scheduling.
Each service not receiving an OR allocation on a given day (due to low historical workload) can be combined into an “OTHER” service (i.e., open, unblocked, first-scheduled, first-served time). At facilities without substantial cross-training of staff, there may be different “OTHER” services for different nursing teams. The calculations of the preceding sections are repeated for the “OTHER” service(s) on each workday.
Importantly, do not simply measure the average OR workload of a service, observe that it is too low for an allocation of an 8-h OR for the day, and then automatically pool it into “OTHER” service time. Apply the graphical methods of the preceding section to assure that the reason for a low OR workload reflects an actual low workload, not a service that operates every other week on the studied day of the week [12]. Likewise, assure that incomplete data or a trend in OR workload is not being observed.
Using Qualitative Information to Improve Forecasts
Qualitative information not available from information system data should be used when finalizing OR allocations.
For example, a surgeon operates at an outpatient surgery center on Fridays in her 8 h of allocated OR time. For years, she has consistently performed 7.0–7.5 h of cases at the surgery center in her OR time. The OR allocations are being updated for the next quarter. Based on historical data, she would, of course, be allocated 8 h of OR time on Fridays. However, she is 8 months pregnant and has requested 3 months of maternity leave. She should not be allocated OR time during the next quarter because it would be underutilized, thereby reducing OR efficiency. Even without personally allocated OR time, she would continue to have open access to OR time on any future workday, if she were to change her mind and work for a few days during her period of maternity leave. Note that if she were not provided open access to OR time, then there would be an adversarial relationship between the facility not wanting to plan a “block” for her versus her desire to keep some block time to provide herself and her patients some flexibility [44]. This highlights that providing open access to OR time on any future workday generally increases OR productivity.
While applying qualitative knowledge, though, focus on the cognitive bias that results in most of the inefficiency of use of anesthesia time, the bias being lack of use of the mathematics [44, 46]. The qualitative information should be used to update the forecasts of workload, not used to create an ad hoc process of converting from workload to OR allocations. We humans are good at forecasting changes in workload, not in making the mathematical conversions from mean workloads into appropriate OR allocations (staffing) [44].
Forecast Remaining Underutilized OR Time
A concern at some facilities is that underutilized OR time is needed for nonclinical, but nonetheless important activities. For example, equipment for the next day’s cases may be set up by nurses whose ORs finish earlier than the end of their shift. The nursing supervisors at such facilities may express concern that changing OR allocations to increase OR efficiency will impair processes that function well by taking advantage of existing underutilized OR time.
Expected underutilized time can be estimated empirically after future OR allocations have been determined. Applying the allocations, each historical day’s resulting total underutilized hours are calculated. The statistical distribution of each day’s total hours of underutilized OR time can be described using histograms or percentiles.
Case Scheduling
Allocating OR time to increase OR efficiency is of little value unless cases are also scheduled into the OR time appropriately.
A series of thought experiments and computer simulations was performed to evaluate case scheduling based on maximizing OR efficiency [21]. The performances of different case scheduling heuristics were compared. The analyses showed that managers can achieve efficient OR scheduling while leaving case scheduling decisions to the convenience of surgeons and patients, provided three simple scheduling rules are followed. In other words, there are small differences in the resulting OR efficiency among different scheduling heuristics, with three exceptions.
The first of three scheduling rules is that a service should not schedule a case into another service’s OR time if the case can be completed within its own allocated OR time [21].
For example, two thoracic surgeons are partners in a group that has been allocated 10 h of OR time on Tuesdays. One of the surgeons has scheduled 6 h of cases into the OR time, leaving 4 h of allocated but unscheduled OR time. A cardiac surgeon has scheduled 2 h of cases into his personally allocated 8 h of OR time. Nine days before the day of surgery, the second thoracic surgeon wants to schedule a new 2-h case. The available start time would be after her partner who has already scheduled cases. The case would not be scheduled into the cardiac surgeon’s OR time, even if the second thoracic surgeon wants to start earlier. The reason is that the thoracic surgeons have available OR time for the case.
The reason for this result is that OR allocations are calculated based on expected OR workload on the day of surgery. Services fill their allocated OR time at different rates [60]. Some services have many patients who are inpatient preoperatively, and those cases typically would be booked in the OR scheduling system on the day before surgery [6]. Almost all facilities with allocated OR time follow the preceding scheduling rule. Thus, the importance of this finding was not that it showed a new way to schedule cases but that it showed that most facilities make decisions based on OR efficiency [21]. By definition, the decision would not represent a change in facility practice, but an unusual request of the second thoracic surgeon, because otherwise the thoracic surgeons would not have been allocated 10 h of OR time on Tuesdays.
The second of the three scheduling rules is that a case should not be scheduled into overutilized OR time if it can start earlier in another of the service’s ORs [14, 22]. This applies to services allocated two or more ORs. Suppose that OR workload is 23 h. The expected hours of overutilized OR time would be slightly less if two OR were allocated for 13 h (total 26 h) versus three OR for 8 h (total 24 h). This result would be less reliable if case scheduling did not result in similar packing of the cases into the allocated OR time [14, 21, 61] Simulations show usually it does.
For example, a service has been allocated OR 3 and OR 5 from 7 a.m. to 3 p.m. One surgeon in the service has scheduled cases in OR 3 to finish around 2 p.m. OR 5 is empty. A second surgeon in the service wants an afternoon start. He asks to start an elective 3-h case at 2:30 p.m. in OR 3. Even though OR workload would be the same, scheduling the case into OR 3 would be expected to result in overutilized OR time and thereby reduce OR efficiency. His request should be denied. The surgeon should take the first case of the day, start in OR 5, or schedule the case on a different workday.
The preceding scenario matches what is done at most surgical suites. Cases are generally not scheduled into overutilized OR time when a service has another allocated OR that is empty. Consequently, as the first rule above, this rule shows that scheduling cases based on maximizing OR efficiency differs little from what is commonly done in practice [21]. Changes resulting from decision making based on OR efficiency generally do not affect case scheduling. Rather, they affect OR allocations (as above) and in the third rule regarding how OR time is released.
The third of the three scheduling rules is that if a service has already filled its allocated OR time, then, to maximize OR efficiency, its new case should be scheduled into another services’ OR time instead of into overutilized OR time [21, 61].
For example, a service has filled its allocated OR time but has another elective case that it desires to schedule. If the OR time of another service were not released, the case would be performed in overutilized OR time. OR efficiency is greater by performing that case in the OR time allocated to another service that otherwise would be underutilized on the day of surgery.
For example, a surgeon appears to be subverting the case scheduling system for the “OTHER” service, which provides first-scheduled, first-served OR time. The surgeon seems to be creating fictitious patients to “hold” OR time for his cases (e.g., at the desirable 7 a.m. start time). At the OR Block Committee meeting, a manager suggests that there be the policy that when a case is cancelled, first access to cancelled OR time goes to other surgeons with waiting cases, not the surgeon canceling the case. That recommendation is not sound. When a service has filled its allocated OR time and has another case to schedule, OR efficiency is enhanced by releasing the OR time of the service expected to have the most underutilized OR time [60]. No cases should be waiting to be scheduled.
To evaluate which service should have its OR time released, simulations were performed scheduling new hypothetical cases into actual OR schedules. Services fill their allocated OR time at different rates. Thus, theoretically, the service that should have its OR time released for a new case should be the service that is predicted, at the time the new case is booked, to be the service that will have the most underutilized OR time on the scheduled day of surgery. In practice, performance is made only slightly worse (versus having perfect retrospective knowledge) by scheduling the case into the OR time of the service with the largest difference between allocated and scheduled OR time at the time when the new case is scheduled [60]. The latter is practically straightforward to implement.
In contrast, releasing the OR time of the service with the second most, instead of the service with the most, allocated but unscheduled OR time has a large negative effect on OR efficiency [60]. The reason is that usually a particular case can only be scheduled into one or two services’ OR time without resulting in overutilized OR time. The differences among those few services in their amount of expected open OR time often are large. This occurs because day-to-day variability in the OR workload of services on a day of the week generally exceeds variability due to the timing of how quickly different services filled their allocated OR time.
The timing of when allocated OR time should be released has been studied [62]. Potentially, the scheduling office could wait to release the allocated OR time until closer to the day of surgery, when data may be available on subsequently scheduled cases, in order to improve the quality of the decision. Slightly more than half of ORs have a change in one or more cases within 1 workday of surgery [6, 63]. However, simulation results were equivocal as to the benefit of such a decision [62]. Under two conditions, postponing the decision of which service had its OR time released for the new case until early the day before surgery had a negligible effect on resulting OR efficiency versus releasing the allocated OR time when the new case was scheduled [62]. First, this finding applied to an ambulatory surgery center with brief cases. At such facilities, typically there is only one good choice for the service to have its OR time released [60]. Thus, there is no good reason to wait in making the decision. Second, this finding often also applies to large surgical suites in which cases are scheduled as if there were many smaller suites. For example, at a 30-OR surgical suite, one nursing and anesthesia team may staff the six ORs used for general and vascular surgery. From the perspective of releasing OR time for a new general or vascular surgery case, only six ORs are available, not all 30 ORs.
For example, a hospital contains a team cross-trained in neurosurgery and otolaryngology. One week hence, on next Thursday, neurosurgery has been allocated one 10-h OR. Otolaryngology has been allocated one 10-h OR also. The otolaryngologists have scheduled 11 h of cases into their OR. A third otolaryngologist wants to schedule another 2-h case. The neurosurgeons have scheduled a case for 3 h from 7 a.m. to 10 a.m. The otolaryngologist with the new case can book the case because the surgeons have open access to OR time on whatever workday they choose. Provided the otolaryngologist is available at 10:30 a.m., then the neurosurgeons’ OR time would be released. There is no advantage to waiting to schedule the case. Yet, if the neurosurgeon with the 7 a.m. to 10 a.m. case was to schedule another case, the scheduling office should contact the otolaryngologist and perhaps she would not mind starting her case later in the day.
Despite this consideration of how best to release allocated OR time, it is important to appreciate that results are highly sensitive to the OR time being allocated appropriately based on OR efficiency [14]. Issues of when to release allocated OR time vastly pale in practical importance to OR allocation and staffing [14]. The difference typically is less than one case between whether a 2 OR service has overall underutilized or overutilized time [23, 63]. This remarkable observation explains why the OR allocation decision is so important to reduce overutilized OR time, whereas the case scheduling decision is usually obvious (and, by corollary, absolute differences between scheduled and actual OR time generally can be neglected [2]). Although OR management problems are observed on the day of surgery, often the root cause and only practical way to fix the problem is to plan OR allocations and staffing properly several weeks or months before the day of surgery [14, 64]. The balance between the role of case scheduling versus OR allocations in causing inefficient use of OR time can be assessed for each facility by reviewing multiple examples as in this chapter and comparing each to the facility’s current practices [65].
For example, OR information systems data are used to calculate OR allocations, which then are reviewed by the “block” committee. An ophthalmologist complains that his allocated OR time on Tuesdays has been “released” for 2 of the past 3 weeks. Each time, the otolaryngology service has filled its allocated 8 h of OR time and so has booked cases into his OR time. The ophthalmologist is upset that the schedulers are treating him unfairly by repeatedly releasing his allocated OR time. Although he schedules many cases a couple of days before the day of surgery, his OR workload is consistently at least 7 h each Tuesday. The ophthalmologist’s concerns are well founded; this should not be happening. However, the problem is not that the schedulers are releasing his OR time. Rather, they are making the proper decision to maximize OR efficiency. The problem is that the otolaryngology service should be allocated more than 8 h of OR time. This is either a failure of statistical forecasting of the otolaryngology service’s workload, which is uncommon, or a failure in appropriating allocating OR time based on the forecasted workload, which is more common [44]. At facilities with frequent concerns about releasing of allocated OR time, be sure to focus on who is responsible for statistical calculations of the OR allocations and their use.
Although this chapter has focused on decision making before the day of surgery, the same principles apply to decisions made on the day of surgery [2, 23, 65, 66]. The principles described can also be used to decide how cases are moved on the day of surgery [65, 67], how staff members are assigned on the day of surgery [68], and how cases are sequenced in each OR [51, 69, 70].
Impact of Reducing Times on Productivity
Impact of Reducing Surgical and Turnover Times
The impact of interventions on labor costs can be forecast using each facility’s own data, along with corresponding confidence intervals [12, 19]. For example, turnover times can be reduced between each case [12, 19]. Surgical times can be reduced to national average values for each procedure [20]. First case of the day starts can all be on-time [12, 59]. For all interventions, first the labor cost is calculated assuming that OR time is allocated and cases are scheduled based on OR efficiency. Second, the intervention is performed, thereby reducing OR workload by service. Third, using the revised workload values, OR time is reallocated based on OR efficiency and the new estimates for labor costs projected. Fourth, the differences are calculated. By analyzing the differences in 4-week periods, to prevent effects of variation by day of the week, confidence intervals can be calculated for the differences [19, 20, 37].
For example, consider a hospital that allocates 8 h of OR time to each of many small services, and each has an adjusted utilization less than 85 percent [20]. Cases are being scheduled based on OR efficiency (i.e., sequentially into ORs [21]). Reducing OR times cannot result in reduced overutilized hours because there are none. Labor costs will not be reduced (i.e., they are fixed to achievable reductions in OR times).
For example, a different hospital has few surgical services, most with more than one OR, and many ORs with workloads exceeding 8 h [20]. Then, reducing OR times can result in reductions in workload sufficient to reduce allocated OR time (e.g., an OR allocated for 10 h would now be allocated for 8 h). At this hospital, unlike the one in the preceding example, there would be financially important reductions in labor costs from reducing OR times.
Equivalent analyses can be performed at teaching facilities to calculate [20] the impact of longer OR times (due to factors such as teaching time and development of skills in trainees) [71, 72] on labor costs.
These examples show that, generally, cost reduction from reducing OR or turnover times can only be achieved provided OR allocations are reduced [12]. The initial impact of reductions in OR or turnover times may be increased underutilized OR time and/or reduced overutilized OR time. This initial step is evident to clinicians. The secondary step is revisions of OR allocations based on the new values of decreased OR workload. The latter step provides for the large reductions in labor cost.
Usually, reductions in labor costs from reducing turnover times tend to be small. At four academic tertiary hospitals studied, reductions in average turnover times of 3–9 min would result in 0.8–1.8 percent reductions in labor cost [19]. Reductions in average turnover times of 10–19 min would result in 2.5–4.0 percent reductions in labor costs [19]. These analyses can be fruitful in educating stakeholders that achievable reductions in the times to complete tasks often have less effect on OR efficiency than does good management decision making.
Impact of Not Changing Service-Specific Staffing
Some facilities do not make decisions systematically based on increasing OR efficiency and are unlikely to change their practices [44]. Then, the methodology above can be used to calculate the higher labor costs that the facility sustains from OR time not being allocated and cases not being scheduled based on OR efficiency [16, 37, 41].
For example, anesthesia group expenses exceed revenue at a facility. The calculation is performed using labor costs of anesthesia providers. The estimate of the resulting additional labor costs can be used by the anesthesia group and hospital when negotiating an appropriate administrative support agreement from the hospital [58, 73].
Development of administrative support agreements can also apply to negotiations with medical schools, ambulatory surgical facilities, or a multispecialty group. At two academic medical centers, estimated annual excess labor costs were $1.6 million and $1.0 million, respectively [41].
Impact of Not Reducing the Number of Allocated ORs
Some organizations aim to adjust their OR allocations to be as close as possible to those that are expected to maximize OR efficiency while not reducing the number of allocated ORs. This approach does not result in maximal OR efficiency. Instead, this approach reflects organizational support for opening as many ORs as are available for first case of the day starts (e.g., to achieve on-time starts for surgeons) [74, 75]. The mathematics can be weighted to allocate more ORs by repeating the analyses using a higher relative cost of overutilized to underutilized hours (e.g., 3:1). An increase in the relative cost gives an increase in how many ORs are allocated [12, 25]. The smallest value is chosen for which the allocated number of staffed ORs matches the desired, usually current, number of ORs. This analysis is run separately for each day of the week [25].
Increasing the number of allocated ORs results in a slightly smaller percentage increase in OR labor cost than in staffing [25]. The reason is that opening more ORs than are needed to maximize OR efficiency does not change OR workload. Thus, the increase in allocated OR hours increases underutilized OR time and reduces overutilized OR time. The cost per hour of overutilized OR time exceeds that of underutilized OR time. Consequently, the percentage reduction in OR efficiency is less than the percentage increase in allocated OR hours. The same argument applies to labor costs.
Forecasting the Time Remaining in Ongoing Cases
The preceding sections have focused almost entirely on decision making before the day of surgery, because good decision making cannot be done on the day of surgery unless the OR allocations chosen months ahead are appropriate. As considered in the section “OR Efficiency on the Day of Surgery,” when there is consistent overutilized time on the day of surgery, first and foremost this is a failure months before in statistical forecasting of workload and managerial decision making [14]. However, to use those OR allocations in practice on the day of surgery, another set of data is needed: the forecasted time remaining in cases that are ongoing. In most hospital ORs, the cases running at the end of the day are those that took longer than scheduled [2]. Therefore, good decision making cannot be done in the late afternoon without estimating the time remaining in late running cases. The solution to this problem is not intuitive.
Forecasting the time remaining in cases is one of the most important determinants of decision making on the day of surgery, as it affects decisions such as calling for the next patient, moving cases, and staff relief. Even where there is no bias (i.e., systematic difference) between estimated OR times provided by surgeons and the actual durations from the OR information system (e.g., the average difference equals 0 min), there is substantial variance among historical OR times for the same surgeon and scheduled procedure or procedures that comprise a case [76].
Consider a laparoscopic small bowel resection scheduled for 2 h that has been in the OR for 0.5 h. The median expected time remaining is around 1.5 h. In contrast, suppose that the patient has been in the OR for 1.8 h. The median expected time remaining is not 0.2 h, but longer. The reason is that many of these resections took less than 1.8 h, so the median duration of the cases that took longer than 1.8 h is more than 2.0 h. The shorter duration cases have been excluded. For some of these longer cases, the laparoscopic approach may have been abandoned in favor of an open resection due to the presence of adhesions, or a complication ensued requiring additional time.
For any given combination of surgeon and procedure, there is considerable variation between the time when skin closure begins and when the patient leaves the OR (e.g., from 15 min to 90 min) [77, 78]. This “extra time” comprises the time for irrigation, inspection, and closure, for the patient to recover sufficiently from anesthesia to allow removal of the endotracheal tube, for monitors to be removed and intravenous lines secured, and for the patient to be transferred to the stretcher and transported out of the OR.
The time remaining in a case can be forecasted [77–80] using Bayesian methods by combining the scheduled OR time, historical case duration data, and elapsed times in the OR determined from real-time anesthesia information management system data. Practically, this requires computerization for two reasons. First, many cases running late at the end of the day include rare combinations of procedures with little or no historical data [76, 80–83]. These cases have a markedly disproportionate impact on the overall variability in decisions involving case durations on the day of surgery [83]. Second, accurate predictions require data about how long cases have been underway and in which OR they are being performed. That can be inferred automatically based on the identifier of the anesthesia information system workstation transmitting pulse oximetry, electrocardiogram heart rate, and end tidal CO2 partial pressures [84]. The method of automatic forecasting of the remaining time works well even in the absence of any historical data, and automatically incorporates predictive variability in case durations due to changes from the scheduled procedure [77].
Conclusions
OR allocation is a two-stage process [13]. During the initial tactical stage of allocating OR time, considering OR hours to be fixed is reasonable. For operational decision making on a shorter-term basis, such a conceptual model produces results markedly inconsistent with how surgical suites are and should be run. Instead, consider the workload to be fixed on a short-term basis. Provide staff flexibly to match the existing workload, not vice versa. Do so by making operational decisions based on maximizing OR efficiency, as this is an important step to maximizing OR productivity.
References
In its essence, operations management is the process by which one seeks to match supply and demand. This standard economic principle exists in all industries, whether it is an energy company, the big box retailer, or the emergency room of a community hospital. Matching supply and demand can be challenging in the operating room (OR); however, as the supply is relatively constant, demand can fluctuate greatly. For example, the ambulatory surgery center may be fully staffed for ten operating suites every Tuesday. On any given Tuesday, the number of cancellations may spike far above the average, leaving empty operating suites attended by wage-earning employees. Conversely, the surgeon in one room may encounter difficulty during his or her first case of the day, causing the case to run significantly over its allotted time. The schedule of operation for the operating suite is then thrown off.
ORs are, in general, profitable. In fact, ORs generally create seventy percent of a hospital’s revenue, while constituting only forty percent of the total expenses of the hospital. In the model of the free-standing ambulatory surgery center, the OR is a profit center. So although the revenue stream is vital, running the OR efficiently with the lowest costs possible becomes increasingly important. The OR manager must, therefore, have a profound understanding of cost structures.
In addition, OR managers should monitor key performance criteria, in order to facilitate management decisions and to improve OR efficiency, safety, and satisfaction. Examples of key performance criteria include contribution margins, case cancellation rates, start-time delays, turnover times, etc. In more recent years, Medicare reimbursement has become partially linked to patient satisfaction and patient outcomes; as a result, many ORs include such criteria when monitoring performance and making management decisions.
Cost Structures
“Profit margin” is defined as the difference between revenues and costs. The profit margin can be increased by improving revenues, or, alternatively, decreasing costs. “Costs” can first be categorized as fixed, variable, or semivariable. “Fixed costs” are costs the OR will incur regardless of volume of surgeries performed and generally account for more than fifty percent of an ORs total expenses. Examples of such costs might include the salary of administrative staff, mortgage, capital equipment, billing costs, and information systems. These costs will be in place whether the OR sees an increase in volume, no change in volume, or even a decrease in volume and are generally long-term costs. “Variable costs” fluctuate based on the volume of cases performed and are generally short-term costs. For example, as more surgeries are performed, more supplies are used, and supply costs increase. If there are no surgeries performed on a given day, no additional supplies will be used. Some costs are “semivariable” or have elements of fixed and variable structures. An example of semivariable cost is a full-time employee (FTE) who is paid on an hourly basis. The first 40 hours of their wages are more or less fixed. Any overtime pay, however, is a variable cost, which is dependent on OR volume. Labor and materials are the two significant fixed and variable costs in the OR.
As one allocates costs, it is important to differentiate direct and indirect costs. Direct costs are directly associated with the running of the OR. Such costs can be attributed back to a source such as a thyroidectomy or other procedure. All hospitals and ORs must factor in “indirect costs,” or “overheads,” as well. Overhead costs are allocated costs that are spread among all departments or parts of a business. They often originate from support departments such as laundry, dietary services, and housekeeping. Overhead costs may not be directly attributed to the OR but must be factored into the OR budget.
The OR manager may not be able to influence overhead costs, but does have a hand in establishing direct costs in the categories of supply costs, practice management costs, and personnel costs. “Controllable costs” are costs that can be influenced by a manager’s decisions. Staffing and supplies are often costs controllable by OR managers. Several strategies for reducing personnel costs have been utilized in the OR.
Personnel costs include the nurses, surgical technicians, and aides in the OR and postanesthesia care unit (PACU). Anesthesia providers may also be included if they are employed by the hospital or surgery center. Personnel costs can make up to 60 percent of an OR budget. Thus, any variance in these costs can have a great impact on the profitability of the OR. Efficiently allocating full-time, part-time, and salaried OR employees in order to avoid excess overtime or underutilized paid time are key to optimizing labor costs.
Supply costs involve the materials needed by the OR to perform surgeries. This is a major component of the OR budget. Pressure exists to hold costs down as much as possible. Thus, this goal must be kept in mind all the way from purchasing the supplies to the analysis of the supplies’ use. The OR must have reliable, vital supplies on hand at all times, while avoiding excess or unused inventory, which raises costs without adding value. Thus, optimizing supply costs can be improved by accountability and engagement of all OR personnel and physicians to reduce waste.
Contribution Margin
In managerial accounting, a key concept to understand is the idea of the “contribution margin.” This is a method of looking at the relationship between revenues and costs so that the manager can use them to make decisions regarding planning. The key attribute of this approach is the strict delineation between fixed and variable costs, as fixed costs are more constant and will remain in place regardless of how many surgeries are performed. The contribution margin is defined as the difference between revenues and variable costs.
The resulting amount, or contribution margin, is then applied to cover the fixed costs. If the contribution margin is greater than the fixed costs, the firm makes a profit. Conversely, if the contribution margin is less than the fixed costs, the firm suffers a loss. Because contribution margin depends upon revenue, payer mix and reimbursement levels will affect the contribution margin. Payer mix and reimbursement levels can vary greatly between institutions. Thus, when filling OR time, one should focus on procedures with higher contribution margins to improve profits, rather than those with high profit but lower contribution margins. Otherwise, when improving utilization of ORs with low contribution margin procedures, one can decrease margins. This is more easily understood if viewed graphically.
In Figure 9.1, the fixed cost is held fixed at $10,000. Total cost starts at $10,000 then has a positive slope up, based on variable costs. Variable cost is the difference between total cost and fixed cost. The point at which total cost and revenue are equal is the “breakeven point,” or the point at which the contribution margin covers the fixed cost exactly.
Figure 9.1 Contribution margin.
Any additional contribution margin will be applied to profit. The difference between the revenue and total cost line equals the profit (to the right) or loss (to the left of the breakeven point). For this OR, the breakeven point (again, the point at which costs are covered by revenue) is fifty procedures. If this were the data for one week’s time, the manager would know that as long as the OR were able to book and perform fifty procedures per week, the unit would not lose money. In addition, any procedures after the fiftieth procedure would be profitable. At any point below fifty procedures, however, the contribution margin would be unable to cover the OR’s fixed expenses.
Relationship between Financial and Operational Performance
Although OR managers are not typically responsible for overall financial and operational performance, they must be aware of at least a qualitative relationship between the financial and operational performance of the OR. In order to understand the relationship between operational performance and financial performance, OR managers must have a basic understanding of both financial and managerial accounting. Financial accounting is known as external accounting because the principles of financial accounting aim to create standardized reports with the purpose of comparing organizations within an industry. Managerial cost accounting is known as internal accounting, because it is primarily used as a tool for managers to measure trends in financial performance internally.
Financial accounting statements are composed of four basic documents: (1) the balance sheet, (2) the income statement, (3) the cash flow statement, and (4) the notes to the financial statement.
The balance sheet contains a categorized list of assets and liabilities at the end of an accounting period. It is a picture of the organization’s financial position at one point in time, and is often considered the best single indicator of the financial condition of the organization (see Figure 9.2).
Figure 9.2 Community health care system balance sheet.
The income statement contains a list of all revenues and expenses (Figure 9.3). Revenue for a hospital is generated from both patient care operations and non–patient care operations (parking, cafeteria, research grants, rental space, or equipment sales). Expenses include salaries, benefits, supplies, depreciation, and professional fees.
Figure 9.3 An example of an income statement.
The cash flow statement contains a list of cash-producing and cash-consuming transactions for the accounting period (Figure 9.4). The cash flow statement shows the financial status in terms of cash flow, rather than according to revenues and expenses in the income statement or accounting entries in the balance sheet.
Figure 9.4 Community health care system cash flow statement.
The notes may include valuable information about the operational status of the organization. Much of the benefit derived from financial statements results from tracking trends in values from one period to the next.
Because healthcare organizations function in different geographic regions and environments while delivering different mixes of services to patient populations, operational indicators make diverse organization more comparable. Many operational indicators are adjusted for case mix and prevailing local wages. Case mix adjustment is a mathematical correction made to account for differences in case type and severity of illness between patient populations. Wage index adjustment is a mathematical correction made to account for differences in employee wages between geographic areas.
Measures of Financial Performance
In addition to the general observation of whether or not an OR is producing enough revenue to meet expenses, financial metrics or indicators can further describe the financial health of the entity. Such measurements describe an institution’s ability to meet its responsibilities. For instance, a business can meet its short-term demands by taking on additional debt. This will allow the business to pay its bills at the end of a given month. This, however, is unsustainable in the long term. The business must be able to develop enough equity to meet these obligations. An inability to build such equity will hinder the business from making purchases, developing new technology, or continuing to provide its current standard of service.
Specific financial ratios are also accepted indicators of financial performance (Figure 9.5). “Liquidity ratios” measure the ability to meet short-term obligations. “Capital structure ratios” measure ability to meet long-term obligations. “Profitability ratios” measure ability to generate retained earnings and thus to increase assets. “Activity ratios” measure the efficiency with which assets are used to generate revenues.
Figure 9.5 Community health care system financial performance ratios.
Indicators of financial performance expressed as ratios make the information easier to interpret and facilitate benchmarking between similar institutions.
Operation room managers understand that factors such as reduced OR costs, increased OR operational efficiency, and increased patient volume will improve financial performance.
Liquidity is the measure of an institution’s ability to pay its debts or liabilities with its current assets. One can consider this to be “cash on hand.” The amount by which assets exceed liabilities is known as “working capital.”
As an organization’s liquidity increases, it has a greater ability to meet its liabilities. If an organization is managed poorly or is struck by unexpected financial hardship, a crisis can ensue when its bills cannot be paid. Such a crisis may be short lived or may be an indication of deeper financial problems. Liquidity is often studied in terms of ratios. The current ratio is the measure of current assets to current liabilities.
Rarely is this ratio equal to 1. Rather, a very high acid test ratio is approximately 0.3.
Another interesting liquidity ratio useful for making decisions and deciphering financial health of an organization involves the number of “days of cash on hand.”
Simply put, this measures the number of days the organization could meet its expenses using solely the cash it has on hand. This indicates an organization’s ability to survive a sudden downturn in cash flows. For the manager, or even an investor unfamiliar with operations management, this is a concept that can be missed. This is particularly true in the setting of the freestanding ambulatory surgery center. The capital to make a purchase may be available upfront (the building is financed, the equipment is purchased); however, the cash on hand is insufficient to maintain the operation should volume decrease due to market forces. A rudimentary example may be the ability to purchase the car, but being unable to buy the gasoline to fuel the vehicle.
Capital Structure
Even in healthcare, where the mission is to serve and care for the patient, an OR must generate income. Without doing so, the organization cannot buy equipment or hire additional personnel. As stated earlier, profit margin is the difference between total revenues and total costs. Profitability ratios indicate ability to generate income.
The total margin ratio simply measures the amount of revenue that contributes to income. This can come from patient care or non–patient care sources.
Operating margin excludes non–patient care-related activities and focuses solely on the operational revenues from patient care. This is subsequently usually less than the total margin ratio. Within the OR the main mode of increasing operating margin is by decreasing expenses.
Finally, return on equity is often considered the primary test of profitability. This measure relates income to equity or net assets of the organization. Equity is determined as the difference between assets and liabilities. The resulting number indicates the rate at which an OR or other organization produces profit relative to its net assets. A higher percentage indicates greater return on equity, and thus greater profitability.
Measures of Operational Performance
Although measures of financial performance are vital to determine the health of an organization’s finances, they do not characterize the efficiency by which the organization provides services with the resources at hand. In an ideal model, an organization would utilize resources perfectly. In other words, the OR would consume exactly what is required for its services, no more, no less. Every additional suture required or labor hour needed above this point decreases efficiency. A manager can work to increase operating income by increasing revenues or decreasing expenses. By and large, the overwhelming method for increasing income is by decreasing expenses. These fundamentals are analyzed much like the financial statements through the use of performance ratios (Box 9.1). These ratios are complicated by the fact that case mix is not universal across institutions. A case mix index is therefore applied that weighs a facility’s cases and patient illness against an average. A similar index is utilized to adjust for wage differences between regions.
Box 9.1. Operational Performance Indicators and Suggested Example Values
| Occupancy | 50% |
| Length of stay, case mix adjusted | 4.5 days |
| Revenue per discharge, case mix, and wage index adjusted | $5000 |
| Revenue per visit, wage index adjusted | $225 |
| Cost per discharge, case mix, and wage index adjusted | $5,000 |
| Cost per visit, wage index adjusted | $225 |
| Inpatient staff hours per discharge, case mix, and wage index adjusted | 135 |
| Outpatient staff hours per visit | 6 |
| Salary per full-time equivalent employee, wage index adjusted | $33,000 |
| Capital costs per discharge, case mix, and wage index adjusted | $5400 |
| Outpatient revenue | 33% |
One indicator of interest is the revenue a facility generates per discharge. This can be used for inpatient hospitals as well as ambulatory centers. The net inpatient or outpatient revenues depending on the facility are divided by the number of discharges, adjusted for case mix and wage discrepancy.
A similar indicator can be used to measure the costs incurred per discharge by substituting net costs for revenue. This is again adjusted for case mix and wage.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
