There are several strategies to reduce the labor cost of providing anesthesia care. One approach is to achieve more efficient matching of staffing to the anticipated workload using analytics (operations research),1–7 thereby reducing unnecessary staffed hours without changing the number of operating rooms (ORs). Another approach is for fewer anesthesia providers to provide the clinical care in a manner that does not increase a department’s incidence of adverse outcomes.8–10
Communication systems can increase productivity by coordination of care among ORs and perioperative locations.11 Such coordination includes the timing of administration of oral midazolam or vancomycin to patients in the preoperative holding area. St. Jacques and Rothman12 refer to such coordination as “computer enhanced workflow.” Such technology increases productivity when applied to activity outside of ORs.11 Computer recommendations improve the quality of anesthesiologists’ decisions for case coordination.11 Such technology is used when provided,13 perhaps because anesthesiologists perceive that their professional work includes coordination with other health care providers to complete each assigned list of cases as quickly as possible.14
However, the maximal potential of communication systems to increase the productivity of anesthesia care by enhancing anesthesiologists’ coordination of care (activities) among ORs is unknown. If few messages to anesthesiologists originate from outside of ORs, the preceding statistically significant11,13 benefits would be of small magnitude, managerially.
Permissions to perform the studies using data from hospitals A and B were obtained from the respective Institutional Review Boards of their affiliated universities (Thomas Jefferson University and State University of New York Upstate Medical University), without requirement for written consent. Both hospitals were appropriate for study because at neither did anesthesiologists have a remote method of knowledge of what was happening in the ORs they were supervising other than from the devices studied (e.g., no access to displays of hemodynamics, no remotely located slave monitors, and no video feeds from ORs [e.g., accessible via smartphones or desktop computers]).12,13
Our first hypothesis was that when anesthesiologists supervise, as opposed to provide care personally, they receive as many messages (≥50%) originating from outside of ORs as originating from the ORs they are supervising. We use the word “supervise” in the generic sense of overseeing the care of other anesthesia providers (e.g., anesthesiology residents and certified registered nurse anesthetists [CRNAs]),15not as related to United States (US) billing nomenclature (i.e., “medical direction” versus “medical supervision”).a We formulated this hypothesis based on a survey of anesthesiologists and CRNAs at one US hospital where the anesthesiologists were involved in fewer than half of inductions and emergences, but 98% of patients’ preoperative and postoperative care and 88% of departmental quality-improvement activities.16
At hospital A, data to test hypothesis 1 were obtained from files recorded in its internal alphanumeric text paging system (Zetron, a subsidiary of JVC Kenwood, Redmond, WA).17 The paging system is accessed by providers in ORs using anesthesia information management system computers that are attached physically to the anesthesia machines. The hospital has 42 anesthetizing locations per day (median 42, 25th percentile 41 and 75th percentile 44). The paging system was implemented in December 2011. Several months later, a quality assessment audit was performed to assess whether the technology was being used appropriately. Pages were analyzed from 60 nonholiday workdays between Monday January 2, 2012 and Friday April 6, 2012. The audit included the time of the page, the sender, the recipient, and the message sent.
At hospital A, the alphanumeric text paging system was not used from the postanesthesia care unit (PACU). Instead, numeric pages were sent using telephone keypads (i.e., pager messages were not captured from the PACU). A resident anesthesiologist was assigned to the PACU on nearly all weekdays and was the initial contact for most issues in this unit. Usually, the attending responsible for covering the PACU was not assigned to supervise the OR(s). Therefore, there were few PACU pages to attending anesthesiologists supervising ORs (i.e., these pages were not included in the analysis). Consequently:
Hypothesis 1A was that slightly less than 50% of the text pages to anesthesiologists would be from outside of ORs.
The subscript “A” indicates that hypothesis 1 was tested at hospital A. The 95% lower confidence limit of the incidence of such pages was calculated using the (exact) method of Blyth-Still-Casella (StatXact-9; Cytel Software Corporation, Cambridge, MA).
At hospital B, data to test hypothesis 1 were obtained by 3 of the authors (BS, FL, RG). Each promptly completed a card with categories after every call received using Vocera B2000 communication badges (Vocera®, San Jose, CA). These authors all occasionally assumed the role of the anesthesia coordinator (“board runner”) at their surgical suites. In this role, in addition to caring for patients, the individual also managed add-on cases, assigned providers, etc. Thus, this would increase the percentage of calls from outside the OR. Consequently:
Hypothesis 1B was that slightly more than 50% of calls would be from outside of ORs.
The lower 95% confidence limit was calculated as described above.
If the principal potential for communications to increase anesthesia productivity was for coordination of anesthesiologists’ activities outside of ORs (hypothesis 1), urgent calls from ORs would consequently be uncommon. Only 7.3% ± 0.7% of 1-minute hypoxemic episodes (pulse oximeter saturation <90%) were unresolved by the anesthesia care provider in the OR after 3 minutes (i.e., by the time when supervising anesthesiologist would reasonably have arrived).18 Approximately 29% of the hypoxemic episodes occurred during induction and emergence (i.e., times when the supervising anesthesiologist would likely be present anyway). The remaining 71% of the unresolved 7.3% was approximately 5%. Some of the 5% of episodes would prompt a message from the provider to an anesthesiologist. Consequently:
Hypothesis 2 was that few (<5%) messages to anesthesiologists would request their urgent presence in the OR.b
Testing this hypothesis was important because other events such as sudden, unexpected blood loss would also prompt urgent messages. The 1-sided P value was calculated using the binomial test comparing the observed incidence with 5%.
At hospital A, a page to an attending was considered requesting presence urgently if the text contained words such as “STAT,” “Help,” “NOW,” “come”, or “immediately.” Each such page was confirmed manually to determine that it did not include a target word in a nonurgent context (e.g., “Please bring me a stat lock,” referring to a STATLOCK® catheter stabilization device). The number of urgent pages was calculated per regular workday. The probability distribution was consistent with Poisson, because there were 47 workdays with 0 such pages, 9 with 1, 4 with 2, and 0 with 3 or more (P = 0.99 by Kolmogorov-Smirnov test). Consequently, pages were pooled, without batching by day.
At hospital B, each author started recording calls on a Monday: March 19 (BS), March 5 (FL), and February 27 (RG). The study size (power analysis) was based on data from hospital A, analyzed on April 8. The percentage of urgent pages was <1.0%. To differentiate between 1.0% and 2.5%, using 2-sided α = 0.05 and β = 0.10, the minimum sample size was 840 calls. This minimum was surpassed at hospital B at the end of the week ending April 27, at which time the study interval ended and the data were analyzed.
Hypothesis 1A was accepted. At hospital A, at least 45% of pages were from outside the ORs, this value representing the lower 95% confidence limit of the percentage (n = 6115/13,368 = 46%). These pages were principally from the holding area (20%; e.g., arrival of the next patient), for deficiencies in documentation (19%; e.g., failure of the anesthesia provider to enter his or her name into the staff table for the case), or from the OR director (5.2%; e.g., moving cases).
Hypothesis 1B was accepted. At hospital B, at least 56% of calls originated from outside the ORs, this value representing the lower 95% confidence limit of the percentage (n = 531/898 = 59%). These calls were principally administrative (30%), from the PACU (18%), or from the holding area (11%).
Hypothesis 2 was accepted. Requests for urgent presence of the anesthesiologist in an OR were at most 0.2% of pages at hospital A, the upper 95% confidence limit of the percentage (n = 17/13,368 = 0.1%), significantly <5% (P < 0.0001). At hospital B, such requests were at most 1.8% of the calls (n = 10/898 = 1.1%), also <5% (P < 0.0001).c
We quantified communications related to anesthesiologists’ supervision of ORs. At both hospitals, approximately half the messages originated from outside of ORs. There were few urgent calls from providers to the supervising anesthesiologists.
Communication systems can increase productivity by coordination of care among ORs and perioperative locations.11 Our results show that there are many such messages (i.e., substantive opportunity for automation to increase productivity, as described by St. Jacques and Rothman12). From the perspective of increasing anesthesia productivity, activities performed by anesthesia providers outside of ORs should be a focus of these systems, not activities related to intraoperative workload. To increase productivity for intraoperative care, focus on matching staffing to the anticipated workload months ahead and adjusting case start times (e.g., staggered first case starts) the day before and the day of surgery.1–7,10,19,20
Our results are important because communication systems cannot substantively increase productivity by facilitating supervising anesthesiologists’ provision of “knowledge input” to providers in ORs. Only 14% ± 0.4% (standard error) of the minutes of critical portions of anesthetics are attributable to physiologic events (i.e., potentially addressable as knowledge activities) versus physical actions (e.g., intubation or turning patient prone).10 Peak anesthesia activity in surgical suites (i.e., total providers needed) occurs at the start of the workday for most days (i.e., when there is near simultaneous activity in multiple ORs) (P < 0.0001).10,21 Organizational decision-making related to first-case starts is sensitive both to lack of scientific knowledge of OR management (e.g., how to increase OR efficiency) and to psychological biases (e.g., knowledge that most cases take less time than scheduled but yet belief that starting a few minutes late results in cases finishing a few minutes late).20,22,23 When the limitation of first-case starts on anesthesia productivity is overcome by the use of staggered first-case starts, breaks become a bottleneck to increasing productivity (see Appendix).
Our results are also important because communication systems cannot substantively increase productivity by coordination of anesthesiologists’ physical presence in ORs.24–28 OR times (“wheels in” to “wheels out”) can be segmented into surgical times (positioning to end of surgery) plus nonsurgical times (i.e., “anesthesia-controlled”). OR times are affected principally by surgical times, not anesthesia-controlled times24–26 (e.g., because mean anesthesia times are brief relative to standard deviations of surgical times).24,29 For joint replacement cases at several community hospitals, there was no association between increases in the physical presence of the anesthesiologist from 0% to 21% of the case and reductions in OR times.27 At an academic “day surgery” center, anesthesia-controlled times were no briefer (<1 minute, P = 0.19) with anesthesiologist supervision of CRNAs than with anesthesiologists practicing alone.28
Half the messages to anesthesiologists originated from outside of ORs, even though most anesthesia work is done inside ORs. This likely occurred because we (appropriately) studied electronic messages, not oral communication with the anesthesiologist when he or she was (already) physically present in the OR. When patients undergoing surgery are unstable or undergoing high-risk portions of procedures, the supervising anesthesiologist would typically be present in the OR, and thus no pages or calls would be made. Consequently, our findings that few (e.g., 0.2%) pages to anesthesiologists request their urgent presence are consistent with the findings of Silber et al.30 that anesthesiologist supervision was associated with greater odds of successful patient “rescue” (i.e., not death) from complications. The use of “computer-enhanced workflow”12 for coordination of care outside of ORs may facilitate increased anesthesiologist supervision of resident physicians, thereby reducing adverse events.d
Our investigation was limited to 2 large, multidisciplinary academic facilities (see Table 1). Our hypotheses were based on previous studies from hospitals.10,16,18 Thus, our results should be considered limited to hospitals, not freestanding outpatient surgery centers. In addition, we did not study communication interactions between pairs of providers, but rather the overall behavior of large populations. Such a design was appropriate for evaluation of the economics of communication systems.
In conclusion, approximately half of messages to supervising anesthesiologists are for activity originating outside the ORs being supervised. To use communication tools to increase anesthesia productivity on the day of surgery, their use should include a focus on care coordination outside ORs (e.g., PACU) and among ORs (e.g., at the control desk).
APPENDIX: INFLUENCE OF BREAKS ON ANESTHESIA PRODUCTIVITY
During an 8-hour workday plus 30 unpaid minutes for lunch, the percentage of time requiring an additional provider for breaks (e.g., lunch) would be 0.5 hours/8.5 hours = 6%. The percentage may be <6% because some breaks could be taken during prolonged turnovers or while waiting for add-on cases. We hypothesized that approximately 5% of the anesthesia time would be attributable to an anesthesia provider giving a break in an OR.
We tested the hypothesis using previously published data10 from May 3, 2010 through April 29, 2011, 7:00 AM to 2:59 PM, excluding holidays, weekends, and Thursdays. The Thursdays were excluded because the ORs started an hour later than normal and our analysis was based on paging frequencies during each hour of the day.10 Breaks were identified from the anesthesia information management system. The incidences were conservative (i.e., underestimates of actual percentages of OR time), because breaks were given sometimes without explicit documentation. The percentages of OR time31,32 were calculated separately for each of the n = 13 four-week periods. Each period contained >2015 hours of cases. The results were reported as the mean percentage among the 13 four-week periods ± SEM. The percentage of OR time during which there was a provider (anesthesiologist or CRNA) documented in the anesthesia information management system as giving a break was 5.1% ± 0.2% of anesthesia minutes. The concordance with the expected percentage from the preceding paragraph highlights validity in using anesthesia information management system data to measure workload.10,33
The influence of breaks on increases in anesthesia productivity has not been reported previously. Two articles formally modeled the costs of different anesthesia workforce models.34,35 Consider salaried professionals working at a hospital at which nearly every OR runs for 8 hours to 10 hours daily. Let the annual cost of a CRNA for regular workdays equal CCRNA and that of an anesthesiologist equal CMD. Hogan et al.35 considered the annual cost to staff 12 ORs with 1 CRNA in each OR and each anesthesiologist supervising 2 ORs to equal 12 CCRNA + 6 CMD. The annual cost to staff 12 ORs with 1 CRNA in each OR and each anesthesiologist supervising 4 ORs was considered to equal 12 CCRNA + 3 CMD. The economic savings in using 1:4 rather than 1:2 supervision was considered to equal 3 CMD, where 3 CMD = 6 CMD – 3 CMD. However, more than 12 CRNAs would be needed to staff the 12 ORs when anesthesiologists are supervising 4 ORs. To provide OR anesthesia care, (1) there is substantial activity performed outside of the ORs (hypothesis 1), and (2) breaks (e.g., lunch relief) need to be provided. If anesthesiologists do not perform these activities, and the workday is long (see below), additional CRNAs are required. The appropriate incremental cost reduction between 1:2 and 1:4 supervision, to be compared with patient outcomes, is not 3 CMD but 3 (CMD – CCRNA).e This is not taking into account total hours worked (e.g., call) and for practices in which anesthesiologists work more total hours.f Then, the difference would be even smaller.
Our conclusions may be limited to facilities with full (e.g., >7-hour) workdays. Consider a facility at which (1) requests for presence of anesthesiologists outside the OR could wait until the anesthesiologist is available, and (2) breaks do not need to be provided because there is sufficient time between cases or there are few hours of cases in each OR. For example, the facility may have few hours of cases per anesthetizing location (i.e., low anesthesia group productivity2,27,36–38 and/or prolonged turnovers in the middle of the day from gaps between surgeons2,39–41). The most cost-effective use of CRNAs at such sites would be to perform each of the tasks sequentially. Although this may result in longer turnover times than if more CRNAs and available ORs were provided and work was done in parallel, turnover times are not limiting the surgeon at such facilities, because otherwise more cases would have been scheduled.38
Franklin Dexter is the Statistical Editor and Section Editor for Economics, Education, and Policy for the Journal. This manuscript was handled by Dwayne R. Westenskow, Section Editor of Technology, Computing, and Simulation, and Dr. Dexter was not involved in any way with the editorial process or decision.
We appreciate the assistance of Tracey Cartland, in the SUNY Upstate Medical University Information Technology Department, for providing logs of Vocera calls.
Name: Bettina Smallman, Dr Med.
Contribution: This author helped design the study, conduct the study, and write the manuscript.
Attestation: Bettina Smallman has approved the final manuscript.
Name: Franklin Dexter, MD, PhD.
Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.
Attestation: Franklin Dexter has approved the final manuscript.
Name: Danielle Masursky, PhD.
Contribution: This author helped design the study, conduct the study, and write the manuscript. This author is the archival author for data from State University of New York Upstate University Hospital.
Attestation: Danielle Masursky has approved the final manuscript.
Name: Fenghua Li, MD.
Contribution: This author helped conduct the study.
Attestation: Fenghua Li has approved the final manuscript.
Name: Reza Gorji, MD.
Contribution: This author helped conduct the study.
Attestation: Reza Gorji has approved the final manuscript.
Name: Dave George, MBA.
Contribution: This author helped conduct the study.
Attestation: Dave George has approved the final manuscript.
Name: Richard H. Epstein, MD, CPHIMS.
Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript. This author is the archival author for data from Thomas Jefferson University Hospital.
Attestation: Richard Epstein approved the final manuscript.
This manuscript was handled by: Dwayne R. Westenskow.
a We consider “supervision” in the generic sense of overseeing the care of other anesthesia providers. However, for readers in the US interested in the immediate, practical implications of our report, the US Centers for Medicare & Medicaid Services allows an exception to the limitations on medical direction for receiving new patients into the preoperative area, discharging patients from the postoperative areas, and coordinating case scheduling.
b Another potential role of communication systems was to answer questions remotely. We found grouping of questions into categories including or not including a request for presence to be unreliable. In addition, there was heterogeneity among providers in whether they responded by going to ORs, perhaps reflecting difference among procedures. For example, consider receiving the question “Should I administer a unit of erythrocytes?” Interpretation depends on the context of the patient, procedure, and provider. If transfusion was planned, a phone call might be sufficient without need to visit the OR. However, if the need for transfusion was unexpected, a visit to the OR might be made to assess the situation (e.g., to evaluate the need for more IV access).
c The percentage was larger for hospital B than hospital A (P < 0.0001) even though the 3 anesthesiologists recording messages at hospital B knew that their supervision was being evaluated prospectively, whereas the decision (and need) to assess hospital A’s pages was made retrospectively. The hospital B messages were voice calls, not alphanumeric text pages to be typed, and a vital sign may have been spoken instead of asking “come.” Communication of a vital sign or statement that a vital sign was abnormal (e.g., “the patient is hypoxic”) was included in at most 1.7% of calls (1.0%, n = 9/898). When all calls regarding vital signs were counted, either from an OR or from another site (e.g., PACU), the total was at most 1.9% of the calls (1.2%, n = 11/898). Requests for urgent presence and either including a vital sign or indicating an abnormal vital sign were at most 0.7% of calls (0.3%, n = 2/898). The latter gives the maximal percentage potential enhancement of communications by a remote monitoring system using physiologic monitors.
d De Oliveira GS, Fitzgerald P, McCarthy R. Anesthesia resident supervision and its implication to patient safety. American Society of Anesthesiologists’ Annual Meeting 2011:A1723.
e As emphasized in the Methods section, we use the term “supervision” in its generic sense, not as related to US billing nomenclature (i.e., “medical direction” versus “medical supervision”). If “supervision” at a hospital was principally an activity of responding to urgent requests for assistance, as our Results show did not hold at the hospitals we studied, then the cost term would be the paper’s original 3 CMD.
f Becker’s ASC Review. ASA President Offers 7 Reasons to Question Anesthesia Cost-Containment Study. Available at: www.BeckersASC.com/anesthesia/ASA-President-offers-7-reasons-to-question-anesthesia-cost-containment-study.html. Accessed May 6, 2012; Abouleish AE, Stead SW, Cohen NA. Myth or fact, nurse anesthetists cost less than anesthesiologists. ASA Newslett 2010;74:30–4, 51.
1. Abouleish AE, Dexter F, Epstein RH, Lubarsky DA, Whitten CW, Prough DS. Labor costs incurred by anesthesiology groups because of operating rooms not being allocated and cases not being scheduled to maximize operating room efficiency. Anesth Analg. 2003;96:1109–13
2. McIntosh C, Dexter F, Epstein RH. The impact of service-specific staffing, case scheduling, turnovers, and first-case starts on anesthesia group and operating room productivity: a tutorial using data from an Australian hospital. Anesth Analg. 2006;103:1499–516
3. Epstein RH, Dexter F. Statistical power analysis to estimate how many months of data are required to identify operating room staffing solutions to reduce labor costs and increase productivity. Anesth Analg. 2002;94:640–3
4. Dexter F, Traub RD. How to schedule elective surgical cases into specific operating rooms to maximize the efficiency of use of operating room time. Anesth Analg. 2002;94:933–42
5. Pandit JJ, Dexter F. Lack of sensitivity of staffing for 8-hour sessions to standard deviation in daily actual hours of operating room time used for surgeons with long queues. Anesth Analg. 2009;108:1910–5
6. Wachtel RE, Dexter F. Review of behavioral operations experimental studies of news vendor problems for operating room management. Anesth Analg. 2010;110:1698–710
7. Dexter F, Wachtel RE, Epstein RH. Event-based knowledge elicitation of operating room management decision-making using scenarios adapted from information systems data. BMC Med Inform Decis Mak. 2011;11:2
8. Posner KL, Freund PR. Trends in quality of anesthesia care associated with changing staffing patterns, productivity, and concurrency of case supervision in a teaching hospital. Anesthesiology. 1999;91:839–47
9. Freund PR, Posner KL. Sustained increases in productivity with maintenance of quality in an academic anesthesia practice. Anesth Analg. 2003;96:1104–8
10. Epstein RH, Dexter F. Influence of supervision ratios by anesthesiologists on first-case starts and critical portions of anesthetics. Anesthesiology. 2012;116:683–91
11. Dexter F, Willemsen-Dunlap A, Lee JD. Operating room managerial decision-making on the day of surgery with and without computer recommendations and status displays. Anesth Analg. 2007;105:419–29
12. St. Jacques P, Rothman B. Enhancing point of care vigilance using computers. Anesthesiol Clin. 2011;29:505–19
13. Xiao Y, Dexter F, Hu P, Dutton RP. Usage of distributed displays of operating room video when real-time occupancy status was available. Anesth Analg. 2008;106:554–60
14. Larsson J, Holmström I, Rosenqvist U. Professional artist, Good Samaritan, servant and co-ordinator: four ways of understanding the anaesthetist’s work. Acta Anaesthesiol Scand. 2003;47:787–93
15. de Oliveira Filho GR, Dal Mago AJ, Garcia JH, Goldschmidt R. An instrument designed for faculty supervision evaluation by anesthesia residents and its psychometric properties. Anesth Analg. 2008;107:1316–22
16. Fassett S, Calmes SH. Perceptions by an anesthesia care team on the need for medical direction. AANA J. 1995;63:117–23
17. Epstein RH, Dexter F, Rothman B. Communication latencies of wireless devices suitable for time-critical messaging within an anesthesia decision support system. Anesth Analg (in press).
18. Epstein RH, Dexter F. Implications of resolved hypoxemia on the utility of desaturation alerts sent from an anesthesia decision support system to supervising anesthesiologists. Anesth Analg. 2012;115:929–33
19. Dexter F, Shi P, Epstein RH. Descriptive study of case scheduling and cancellations within 1 week of the day of surgery. Anesth Analg. 2012;115:1188–95
20. Dexter EU, Dexter F, Masursky D, Garver MP, Nussmeier NA. Both bias and lack of knowledge influence organizational focus on first case of the day starts. Anesth Analg. 2009;108:1257–61
21. Paoletti X, Marty J. Consequences of running more operating theatres than anaesthetists to staff them: a stochastic simulation study. Br J Anaesth. 2007;98:462–9
22. Dexter F, Xiao Y, Dow AJ, Strader MM, Ho D, Wachtel RE. Coordination of appointments for anesthesia care outside of operating rooms using an enterprise-wide scheduling system. Anesth Analg. 2007;105:1701–10
23. Pandit JJ, Abbott T, Pandit M, Kapila A, Abraham R. . Is ‘starting on time’ useful (or useless) as a surrogate measure for ‘surgical theatre efficiency’? Anaesthesia. 2012;67:823–32
24. Dexter F, Coffin S, Tinker JH. Decreases in anesthesia-controlled time cannot permit one additional surgical operation to be reliably scheduled during the workday. Anesth Analg. 1995;81:1263–8
25. Abouleish AE, Dexter F, Whitten CW, Zavaleta JR, Prough DS. Quantifying net staffing costs due to longer-than-average surgical case durations. Anesthesiology. 2004;100:403–12
26. Silber JH, Rosenbaum PR, Zhang X, Even-Shoshan O. Estimating anesthesia and surgical procedure times from medicare anesthesia claims. Anesthesiology. 2007;106:346–55
27. Dexter F, Weih LS, Gustafson RK, Stegura LF, Oldenkamp MJ, Wachtel RE. Observational study of operating room times for knee and hip replacement surgery at nine U.S. community hospitals. Health Care Manag Sci. 2006;9:325–39
28. Urman RD, Sarin P, Mitani A, Philip B, Eappen S. Presence of anesthesia resident trainees in day surgery unit has mixed effects on operating room efficiency measures. Ochsner J. 2012;12:25–9
29. Dexter F, Abouleish AE, Epstein RH, Whitten CW, Lubarsky DA. Use of operating room information system data to predict the impact of reducing turnover times on staffing costs. Anesth Analg. 2003;97:1119–26
30. Silber JH, Kennedy SK, Even-Shoshan O, Chen W, Koziol LF, Showan AM, Longnecker DE. Anesthesiologist direction and patient outcomes. Anesthesiology. 2000;93:152–63
31. Xiao Y, Hu P, Hu H, Ho D, Dexter F, Mackenzie CF, Seagull FJ, Dutton RP. An algorithm for processing vital sign monitoring data to remotely identify operating room occupancy in real-time. Anesth Analg. 2005;101:823–9
32. Epstein RH, Dexter F, Piotrowski E. Automated correction of room location errors in anesthesia information management systems. Anesth Analg. 2008;107:965–71
33. Epstein RH, Dexter F. Mediated interruptions of anaesthesia providers using predictions of workload from anaesthesia information management system data. Anaesth Intensive Care. 2012;40:803–12
34. Glance LG. The cost effectiveness of anesthesia workforce models: a simulation approach using decision-analysis modeling. Anesth Analg. 2000;90:584–92
35. Hogan PF, Seifert RF, Moore CS, Simonson BE. Cost effectiveness analysis of anesthesia providers. Nurs Econ. 2010;28:159–69
36. Abouleish AE, Prough DS, Whitten CW, Zornow MH, Lockhart A, Conlay LA, Abate JJ. Comparing clinical productivity of anesthesiology groups. Anesthesiology. 2002;97:608–15
37. Berry M, Berry-Stölzle T, Schleppers A. Operating room management and operating room productivity: the case of Germany. Health Care Manag Sci. 2008;11:228–39
38. Sulecki L, Dexter F, Zura A, Saager L, Epstein RH. Lack of value of scheduling processes to move cases from a heavily used main campus to other facilities within a health care system. Anesth Analg. 2012;115:395–401
39. Dexter F, Epstein RH, Marcon E, Ledolter J. Estimating the incidence of prolonged turnover times and delays by time of day. Anesthesiology. 2005;102:1242–8
40. Wachtel RE, Dexter F. Reducing tardiness from scheduled start times by making adjustments to the operating room schedule. Anesth Analg. 2009;108:1902–9
41. Masursky D, Dexter F, Isaacson SA, Nussmeier NA. Surgeons’ and anesthesiologists’ perceptions of turnover times. Anesth Analg. 2011;112:440–4