Association of surgeons’ gender with elective surgical lists in the State of Florida is explained by differences in mean operative caseloads

Abstract

Background

A recent publication reported that at three hospitals within one academic health system, female surgeons received less surgical block time than male surgeons, suggesting potential gender-based bias in operating room scheduling. We examined this observation’s generalizability.

Methods

Our cross-sectional retrospective cohort study of State of Florida administrative data included all 4,176,551 ambulatory procedural encounters and inpatient elective surgical cases performed January 2017 through December 2019 by 8875 surgeons (1830 female) at all 609 non-federal hospitals and ambulatory surgery centers. There were 1,509,190 lists of cases (i.e., combinations of the same surgeon, facility, and date). Logistic regression adjusted for covariables of decile of surgeon’s quarterly cases, surgeon’s specialty, quarter, and facility.

Results

Selecting randomly a male and a female surgeons’ quarter, for 66% of selections, the male surgeon performed more cases (P < .0001). Without adjustment for quarterly caseloads, lists comprised one case for 44.2% of male and 54.6% of female surgeons (difference 10.4%, P < .0001). A similar result held for lists with one or two cases (difference 9.1%, P < .0001). However, incorporating quarterly operative caseloads, the direction of the observed difference between male and female surgeons was reversed both for case lists with one (-2.1%, P = .03) or one or two cases (-1.8%, P = .05).

Conclusions

Our results confirm the aforementioned single university health system results but show that the differences between male and female surgeons in their lists were not due to systematic bias in operating room scheduling (e.g., completing three brief elective cases in a week on three different workdays) but in their total case numbers. The finding that surgeons performing lists comprising a single case were more often female than male provides a previously unrecognized reason why operating room managers should help facilitate the workload of surgeons performing only one case on operative (anesthesia) workdays.

Introduction

Managerial epidemiological studies of anesthesia and surgery in the USA have consistently found low productivity of most hospital surgical suites and ambulatory surgery centers (i.e., there is substantial unused time) [1–6]. Many surgical facilities nationwide have far fewer than 8 hours of cases per anesthetizing location per workday [1–6], especially at anesthetizing locations that are not operating rooms [5, 6]. The many surgeons who perform one case on days they operate and those who perform, on average, two or fewer cases per week, have reduced productivity that results from waiting on the day of surgery for completion of preceding surgical cases that take more time than originally estimated [7]. There are many such surgeons, given that previous studies have shown that most surgeons in both Florida and Iowa perform only one or two cases on the days they operate [8–10]. This finding is unaltered by whether the surgeon principally cares for adult or pediatric patients [8–10]. Frequently, operating room managers do not plan convenient access to operating room time for low-caseload surgeons. This deficiency has been shown in multiple studies with different populations to result principally from lack of scientific knowledge of mathematically optimal surgical scheduling strategies among the decision-makers [11–16]. Typically, managers reserve too much time for surgeons with high daily caseloads and too little time, collectively, for surgeons with low daily caseloads [11, 17]. Courses in the relevant management science significantly increase trust in the relevant mathematical techniques and improve managerial skill in their application [14, 15, 18–20].

An earlier survey study by Yesantharao et al. suggested a novel additional mechanism for why surgeons performing single cases on operative days may have insufficient access to operating room time [21]. They found, at three affiliated teaching hospitals, that female surgeons were less likely than male surgeons to have adequate operating room block time compared to their reported hourly workloads (multivariable-adjusted P = .004) [21]. In other words, implicit or explicit gender-based bias of the operating room managers could be a contributing factor.

We performed a managerial epidemiology study to examine the generalizability of this observation of possible gender bias concerning surgeon access to operating room time across all ambulatory surgery centers and (non-federal) hospitals in Florida [22]. We compared lists of cases (i.e., combinations of a surgeon, date, and facility) between female and male surgeons. We tested whether female surgeons more often than male surgeons performed only one case on days where they did at least one elective surgical case (i.e., had lists of one case). Following Yesantharo [21], our hypothesis #1 is that, overall, female surgeons would have a greater fraction of their lists with one case. This comparison of lists with one case is measurable because many surgeons perform just one elective case on a regular (non-weekend, non-holiday) workday: 47.2% of 175,751 lists among hospitals in Iowa [8], 47.4% of 1,114,860 lists among hospitals and ambulatory surgery centers in Florida [9], and 47.9% of 27,557 lists among pediatric hospitals [10]. This comparison is relevant because observations of multiple populations have found no surgical specialty to have average case durations sufficiently long that single cases would usually fill an operating room for the workday [23–26]. Hypothesis #1 is important because although these low-caseload surgeons account for most surgical growth from year to year [10, 25, 26], their individual percentage utilization of operating room time, either adjusted (including turnover times) or raw (without turnover times) [27], cannot be measured accurately [23, 28]. Such limitations associated with measuring workload of low-caseload surgeons is an example of the type of knowledge necessary for teams responsible for operating room time to make the best possible decisions [11, 13–15].

If there were differences in the percentages of lists with one case between female and male surgeons, there would be different work experiences of female and male surgeons in surgical suites [7, 21, 29]. Understanding the causes of gender-related differences in case scheduling would give insight to operating room managers on how to address potential gender-based concerns related to inequality of access to operating room time. For example, the raw differences between genders might be an artifact from including surgeons’ specialty as a confounder, but not considering surgeons’ quarterly operative caseloads [22]. Our hypothesis #2 was that gender differences would remain despite incorporating confounders, most specifically, quarterly caseloads. If hypothesis #2 were rejected, an implication would be that single-case lists are not related to gender bias, but rather to other factors, such as smaller quarterly caseloads.

Materials and methods

The Institutional Review Board of the University of Florida (IRB202002442) approved this research as exempt from patient consent. The Institutional Review Boards of the University of Iowa (October 20, 2021) and the University of Miami (October 21, 2021) determined that the current analyses of de-identified data do not meet the regulatory definition of human subjects research.

Cross-sectional retrospective cohort study using State of Florida data

We used publicly available data from the Agency for Health Care Administration (AHCA) for patients receiving care at non-federal hospitals and ambulatory surgery centers in Florida from January 1, 2017, through December 31, 2019 [22]. Following approval by AHCA, these data were supplemented with the date of each encounter (for ambulatory patients) and the date of each hospital admission (for inpatients). Dates were necessary to obtain our primary endpoint, the percentage of daily lists of cases (i.e., combinations of a surgeon, facility, and date of surgery) per quarter that included one case. Such discharge abstract data do not include case durations or surgical times, but even if they did, percentage utilizations could not be estimated accurately for these surgeons [23, 28]. Data use agreements were executed between the University of Florida and AHCA, and between the University of Florida and the University of Miami. AHCA disclaims responsibility for the results and conclusions of the study. Sharing of these data is precluded by those data use agreements; readers interested in analyzing the raw data will need to apply for their use with the AHCA as done by the authors.

Table 1 shows the exclusion criteria among the 10,589,761 outpatient encounters and inpatient elective surgical admissions in Florida included in this study [22]. The resulting 4,176,551 cases were performed during the 12 included quarters by 8875 surgeons, with female (21%) or male gender determinable for all surgeons from the National Provider Identifier database (Table 2) [30]. Throughout our paper, we use the terms “male” and “female” because non-binary options for gender are not provided in the study database.

[Figure omitted. See PDF.]

[Figure omitted. See PDF.]

Statistical analyses

Full statistical output, including commands and comments, is provided in the supplemental content. These are listed in the same sequence of the Methods and Results for readers interested in more detail or who want to replicate our work with other state or provincial datasets. Stata v17.0 was used for all analyses (StataCorp, College Station, Texas).

Univariate analyses were performed to compare surgeons based on gender. Exact binomial confidence intervals for the area under the receiver operating characteristic curves (c-statistic) of each surgeon’s quarterly operative caseload to predict gender [31, 32] were calculated. The 99% confidence intervals for the proportion of lists that comprised one case were calculated using the delta method.

Hypothesis #1 assessed the unadjusted association between surgeon gender and whether the list contained one case. Logistic regression was used, P < .01 treated as significant.

Hypothesis #2 assessed adjusted analyses. Logistic regression was used to control for deciles of cases per quarter, specialty, quarter, and interactions between gender and deciles (Tables 2 and 3). We applied clustering by facility when calculating the standard errors. The supplemental content includes rationales and details related to the calculations, and the associated regression coefficients from the logistic regressions. P < .01 was treated as significant.

[Figure omitted. See PDF.]

Both hypotheses were formulated in terms of differences of proportions. (Although odds ratios were calculated, they grossly overstate relative risks because the prevalence of surgeons having lists of one case, or one or two cases, was approximately 50%, not small [e.g., 5%] or large [e.g., 95%] (Table 3). Therefore, we report the odds ratios only in supplemental content.) Contrasts of predictive probabilities between genders were calculated using the STATA margins command. Numerical variables were handled in the regression analyses by using their average estimates over all observations. Confidence intervals were calculated using the delta method to apply the regression model’s variance estimates. These contrasts are reported in the paper because they combine the coefficients of gender alone and the interactions in the nonlinear (logistic) regression.

Several sensitivity analyses were performed. First, individual independent variables (e.g., cases per quarter) were removed from the models to understand what adjustments were influencing results. Second, the dependent variable was changed from one case per list to one or two cases per list (Table 3). Third, we examined sensitivity of confidence interval width (i.e., statistical power) to the numbers of lists (i.e., hypothetical benefit of adding years of data) versus the numbers of facilities (i.e., would need to add more US state[s]).

Regarding study size, in addition to the preceding sensitivity analysis, we also evaluated our decision to proceed (i.e., test our hypothesis #1) after reading the report by Yesantharao from three hospitals [21]. To judge minimum differences that are managerially important, the University of Iowa Department of Anesthesia routinely holds meetings addressing requirements to provide anesthesia services for procedural based physicians who will do 1 or 2 cases in a procedural suite every two weeks (e.g., pain medicine physicians) [33]. Because such proceduralists generally can fill an operating room for half-day, managers regularly treat issues related to 1 out of 20 lists as appropriately calling for their attention [33], where 20 = (10 days in 2 weeks) × (2 of these brief lists per workday per anesthesia practitioner). Being conservative, statistically, if the 99% confidence interval (P < .01) for the unadjusted absolute difference between female and male surgeons excluded 5% (i.e., 1 out of 20 lists), then the study size would be sufficiently large to detect a managerially important difference at the University of Iowa. Such considerations underestimate value because there also are organizational and societal consequences for policies being biased (e.g., based on gender).

Results

We analyzed 1,509,190 lists of cases performed by 8875 surgeons (Tables 1 and 2).

While male surgeons performed a median of 29 cases per quarter (interquartile range 12 to 61), female surgeons performed a median of 13 cases per quarter (interquartile range 7 to 30) (Table 3). Male surgeons’ mode was the 10^th decile, with a left-skewed distribution, while female surgeons’ mode was the 2^nd decile, with a right-skewed distribution. Suppose that a male and a female surgeons’ quarter were selected randomly. Then, for 66% of such selections (i.e., the c-statistic), the male surgeon would have performed more cases (99% confidence interval 65% to 66%, P < .0001).

While 44.2% of male surgeons’ lists of cases (i.e., surgeon-date-facility combination) had one case, 54.6% of female surgeons’ lists had one case (Tables 3 and 4, P < .0001; Fig 1, top row). The value of 54.6% was significantly greater than half (P < .0001). While 65.8% of male surgeons’ lists had one or two cases, 74.8% of female surgeons’ lists had one or two cases (Tables 3 and 4, P < .0001).

[Figure omitted. See PDF.]

The pooled rows include all surgeon-quarter combinations with at least one case. Those pooled rows show the findings of Hypothesis 1. While 44.2% of male surgeons’ lists of cases had one case, 54.6% of female surgeons’ lists had one case, significantly more (P < .0001). The next 10 rows show the findings of Hypothesis 2. With adjustment for quarterly operative caseloads, the gender association (male < surgeon) is mitigated fully, because male surgeons’ mode was the 10^th decile while female surgeons’ mode was the 2^nd decile (i.e., they performed fewer cases). Put another way, select at random a male surgeon-quarter and a female surgeon-quarter, both in the same decile of quarterly caseload. The paired surgeons perform surgery on average for the same numbers of workdays to complete those cases. That is what the operating room manager controls and that influences surgeon productivity. For inferential results controlling for specialty see Table 4 and Fig 2. The potential interaction included in the statistical model is shown by the medians of male and female surgeons matching for the lower (e.g., 1^st and 2^nd) deciles but not for the upper (e.g., 9^th and 10^th) deciles.

[Figure omitted. See PDF.]

Hypothesis 1: Female surgeons have a greater fraction of their lists with only one case

More lists of female surgeons than male surgeons comprised one case (10.4%, P < .0001; Fig 1 top row). Therefore, hypothesis #1 was not rejected. The same result held for lists with one or two cases (9.1%, P < .0001). Both two-sided lower 99% confidence limits exceeded 5.0%, showing managerially important effect sizes.

Hypothesis 2: Gender differences in fractions of lists with only one case persist after incorporating confounders

After adjustment for covariates (e.g., quarterly operative caseloads and specialty), there were no significant differences between female and male surgeons in the absolute percentage differences of lists of cases that comprised one case (-2.1%, P = .03; Fig 1). Therefore, hypothesis #2 was rejected. The same result held for lists with one or two cases (-1.8%, P = .05) (Table 4 and Fig 2). The reversals of signs of the estimates from differences suggesting less access of female versus male surgeons (10.4% and 9.1%) to the opposite (-2.1% and -1.8%) were attributable to the differences in quarterly operative caseloads between female and male surgeons (Table 4 and Figs 1 and 2).

[Figure omitted. See PDF.]

N = 1,509,190 lists. Error bars show two-sided 99% confidence intervals for the differences in these expected values. The Stata commands and output of these logistic regression models are in the supplemental content. The unadjusted models used gender alone as the independent variable, with robust variance estimation to be consistent with the other models. The full adjusted model included gender, decile of surgeon’s cases during the quarter, interaction between gender and caseload, surgeon’s specialty, and quarter (Tables 2 and 3), and was estimated with clustering by facility. The model with adjustment for quarterly operative caseload only excluded the interaction but was estimated with clustering by facility (Table 4). When estimated without clustering the point estimate was the same but the confidence intervals are narrower, ± 0.2% (Table 4). All variables other than gender were handled by using their average estimates over all observations. The figure shows that unadjusted differences between male and female surgeons vanishes when accounting for the effect of the confounding variables, most importantly caseload. See supplemental content for these averages.

Discussion

Operating room managers have a responsibility to ensure that surgeons and other proceduralists have access to operating room time that is based fairly on the medical requirements of their patients and their quarterly caseloads (workloads). Neither incentive programs nor surgical scheduling should be biased, either directly or indirectly, with respect to protected characteristics (e.g., gender, age) of employees and medical staff [22, 34]. The operating room manager has at most negligible influence over a surgeon’s workload during a quarter but does affect how and when those surgical cases are performed. Our managerial epidemiology study shows the generalizability of Yesantharao et al.’s earlier study of three hospitals to the hundreds of hospitals and ambulatory surgery centers throughout a state [21]. Just like they found, our results (hypothesis #1) showed that surgeon gender was associated with differences in surgical case scheduling [21]. Because of significant differences in quarterly caseloads between male and female surgeons [22], hospitals should expect female surgeons to have substantively (>5%) greater incidences of elective surgical days with only one case, or with one or two cases, than male surgeons. Our results (hypothesis #2) also show that such differences between male and female surgeons in their lists (Fig 1 top row) were not due to systematic bias in operating room scheduling (Fig 1 next ten rows). By this, we mean that when total quarterly workload averages three cases per week, whether that would be one day with three cases or three days each with one case does not significantly differ based on surgeon gender. The frequency of surgical lists including one case was explained principally by the surgeons’ quarterly caseloads, a novel finding because examined among thousands of surgeons.

Addressing needs of low caseload surgeons (i.e., the majority of female surgeons)

Managers need to recognize that female versus male surgeons’ experiences with case scheduling may differ substantively, because surgeons with single cases (i.e., more often female surgeons) less often have first case starts and therefore have reduced personal productivity from greater tardiness of waiting for preceding late running cases to finish (Fig 1) [7, 21, 29]. There are essentially no conditions wherein these surgeons performing single cases should have been allocated individual block time, because they could not fully fill the workday and partial use of a day cannot be estimated sufficiently accurately [23, 28]. Nevertheless, there is much that operating room managers can do to facilitate case scheduling for these low caseload surgeons, often performing a single “to-follow” case [7, 29]. Such interventions may be country dependent, with prior studies being principally from the USA. Managers can ensure there are processes such that those surgeons with one case to be performed can get their case on the operating room schedule into their service’s shared (allocated) time several weeks in advance when the case is without resource constraints (e.g., only can be performed in a specific room). That can be done probabilistically, not assigning the case to a specific room, but rather providing a confirmed date with an expectation that there will be sufficient capacity for the case, but not with a specified start time [33, 35, 36]. If the service’s time fills, and the low-caseload surgeon has a case to schedule, the manager can release the time of another service forecasted to have substantial unused time [33, 37–39]. When feasible, plan a brief gap (e.g., 30 min) in addition to the average turnover time between the estimated end time of the preceding surgeon in the operating theatre and the start time of the surgeon with the single case [40–42]. Adding a brief gap is especially useful if the preceding cases have a substantial probability of finishing at least 1-hour later than scheduled [42–44]. Such a process reduces the amount of time that a surgeon would be in the facility waiting idly for an operating room to be available. Preceding surgical cases finishing late and thereby reducing the surgeon’s productivity happens far more often than delays in the availability of surgeons due to travel disruptions when coming from clinics or from other hospitals [45].

Limitations

Our finding of absence of gender-based bias in how operating room cases were scheduled does not negate the fundamentally different question as to why female surgeons performed far fewer cases per quarter. For example, throughout Ontario, female surgeons received fewer procedural referrals than male surgeons, principally because male physicians more often referred patients to male surgeons than female surgeons [46]. That we cannot address this different question is mitigated by the fact that factors influencing surgeon referrals rarely are under the purview of operating room managers (i.e., our study addresses systematic bias in operating room scheduling, not systemic gender-based bias such as due to referral patterns) [22].

In addition to surgeons operating in the state of Florida [9, 26], the observation of overall one or two elective cases per week (Table 2) has similarly been found in multiple previous managerial epidemiology studies from the USA, including three University hospitals [47–49], a large community hospital [50], a city wide health system [29], statewide in Iowa [8, 25], and statewide in Florida among pain medicine physicians’ operative cases [33]. Although suggesting generalizability and the importance of studying surgeons performing single cases on their operative days, our results show that findings may differ among countries depending on surgeons’ average caseloads. Because we studied major therapeutic and major diagnostic procedures (i.e., cases requiring general anesthesia or major conduction blocks) (Table 1), results may differ for countries with surgeons performing proportionately less time evaluating patients, performing office-based procedures, and/or performing minor therapeutic and diagnostic procedures.

We used two years of data (ending just before the start of the COVID-19 pandemic) to achieve a contemporaneous cross-sectional analysis. Had we increased the months of data (or even, potentially, the numbers of surgeons), neither would have substantively reduced our confidence interval widths because these were limited, principally, by the numbers of facilities (Table 4 and Fig 1). In Florida, the number of hospitals does not change markedly from year to year. We studied all non-federal hospitals and all ambulatory surgery centers statewide in Florida. Therefore, greater precision would be obtained by adding more state(s). By doing so, it might be possible to determine if the marginal P-values suggesting possible bias favoring female surgeons is a reliable finding. However, we doubt that such knowledge would be of substantive value because, in retrospect, the practical use would be for the above guidance to the individual operating room manager at single facilities, not to inform statewide or national policy. Furthermore, studying subsets of the Florida data (e.g., by size of facility) to assess potential contributors to bias would be uninformative because of the wide confidence interval widths even when including all data. Thus, we recommend that the focus of additional research related to potential gender bias affecting operating room management should be the referral of operative versus non-operative patients to surgeons differing by gender, as reported from Canada [46]. In other words, based on our findings, we suggest future study to understand better why female surgeons are performing fewer cases per quarter, rather than focusing on how the cases are being scheduled.

The National Provider Identifier database currently only allows providers to self-report a gender of “female” or “male” [30]. Thus, potential influence of other gender identities could not be assessed. However, the prevalence of non-binary surgeons [51] likely is too small to affect our conclusions.

Finally, our paper was limited to data related to individual surgeon productivity and related total growth within operating room budgets [8, 23, 25, 26, 33]. Thus, our paper and results should not be interpreted as having anything to do with operating room nursing or anesthetist efficiency, productivity, utilization, scheduling, or assignment [35–39]. The US administrative state level databases lack case duration data, as needed to analyze such endpoints. However, the latter was not a limitation for our tested hypotheses, because with most surgeons performing one or two cases per day, their individual percentage utilizations could not be measured accurately anyway [23, 28].

Conclusions

Surgeons’ frequencies of performing one case on operative days were highly dependent on their quarterly caseloads. The differences between male and female surgeons in their lists were not due to systematic bias in operating room scheduling. Thus, health policy planners looking to reduce gender bias related to surgeons’ caseloads should not be focusing on operating room scheduling and managers as potential sources, but rather to external factors such as gender-bias in referrals of cases to surgeons. We found that surgeons performing only one case on their operative (anesthesia) workdays more often were female than male surgeons. This is a previously unrecognized reason why it is important for operating room managers to ease the burden for low caseload surgeons. Improving those surgeons’ access will support better access of female surgeons.

Supporting information

S1 File. Full statistical output from Stata v17.0, including commands and comments, is provided in the supplemental content.

These are listed in the same sequence of the Methods and Results for readers interested in more detail or who want to replicate our work with other state or provincial datasets.

https://doi.org/10.1371/journal.pone.0283033.s001

(PDF)

Citation: Dexter F, Epstein RH, Fahy BG (2023) Association of surgeons’ gender with elective surgical lists in the State of Florida is explained by differences in mean operative caseloads. PLoS ONE 18(3): e0283033. https://doi.org/10.1371/journal.pone.0283033

About the Authors:

Franklin Dexter

Roles: Conceptualization, Formal analysis, Methodology, Visualization, Writing – original draft, Writing – review & editing

Affiliation: Division of Management Consulting, Department of Anesthesia, University of Iowa, Iowa City, Iowa, United States of America

ORICD: https://orcid.org/0000-0001-5897-2484

Richard H. Epstein

Roles: Data curation, Resources, Software, Validation, Writing – original draft, Writing – review & editing

E-mail: [email protected]

Affiliation: Department of Anesthesiology, Perioperative Medicine & Pain Management, Miller School of Medicine, University of Miami, Miami, Florida

ORICD: https://orcid.org/0000-0001-8466-3845

Brenda G. Fahy

Roles: Investigation, Resources, Writing – review & editing

Affiliation: Department of Anesthesiology, University of Florida, Gainesville, Florida

References

1. Abouleish AE, Prough DS, Whitten CW, Zornow MH, Lockhart A, Conlay LA, et al. Comparing clinical productivity of anesthesiology groups. Anesthesiology. 2002;97:608–615. pmid:12218527

2. 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 US community hospitals. Health Care Manag Sci. 2006;9:325–339.

3. Dexter F, Dutton RP, Kordylewski H, Epstein RH. Anesthesia workload nationally during regular workdays and weekends. Anesth Analg. 2015;121:1600–1603. pmid:25923436

4. Dexter F, Epstein RH, Rodriguez LI. Throughout the United States, pediatric patients undergoing ambulatory surgery enter the operating room and are discharged earlier in the day than are adults. Periop Care Oper Room Manag. 2019;16:100076.

5. Gabriel RA, Burton BN, Tsai MH, Ehrenfeld JM, Dutton RP, Urman RD. After-hour versus daytime shifts in non-operating room anesthesia environments: national distribution of case volume, patient characteristics, and procedures. J Med Syst. 2017; 41:140. pmid:28776233

6. Wong T, Tsai MH, Urman RD. The expansion of non-operating room anesthesia services. Periop Care Oper Room Manag. 2017; 9:39–42.

7. Wachtel RE, Dexter F. Influence of the operating room schedule on tardiness from scheduled start times. Anesth Analg. 2009;108:1889–1901. pmid:19448219

8. Dexter F, Jarvie C, Epstein RH. At most hospitals in the State of Iowa, most surgeons’ daily lists of elective cases include only 1 or 2 cases: individual surgeons’ percentage operating room utilization is a consistently unreliable metric. J Clin Anesth. 2017;42:88–92. pmid:28843944

9. Epstein RH, Dexter F, Fahy BG, Diez C. Most surgeons’ daily elective lists in Florida comprise only 1 or 2 elective cases, making percent utilization unreliable for planning individual surgeons’ block time. J Clin Anesth. 2021;75:110432. pmid:34280684

10. Epstein RH, Dexter F, Diez C, Fahy BG. Similarities between pediatric and general hospitals based on fundamental attributes of surgery including cases per surgeon per workday. Cureus. 2022;14: e21736. pmid:35251808

11. Wachtel RE, Dexter F. Review of behavioral operations experimental studies of newsvendor problems for operating room management. Anesth Analg. 2010;110:1698–1710.

12. Wachtel RE, Dexter F. Difficulties and challenges associated with literature searches in operating room management, complete with recommendations. Anesth Analg. 2013;117:1460–1479. pmid:24257396

13. Prahl A, Dexter F, Braun MT, Van Swol L. Review of experimental studies in social psychology of small groups when an optimal choice exists and application to operating room management decision-making. Anesth Analg. 2013;117:1221–1229. pmid:24108254

14. Dexter F, Van Swol L. Influence of data and formulas on trust in information from journal articles in an operating room management course. A&A Case Rep. 2016;6:329–334.

15. Ahn PH, Dexter F, Fahy BG, Van Swol LM. Demonstrability of analytics solutions and shared knowledge of statistics and operating room management improves expected performance of small teams in correctly solving problems and making good decisions. Periop Care Oper Room Manag. 2020;19:100090.

16. Elhakim M, Dexter F, Fahy BG. Changes in current employment positions after taking an operating room management course content by physicians and non-physicians and potential use of the content. Periop Care Oper Room Manag. 2020;20:100097.

17. Bai M, Pasupathy KS, Sir MY. Pattern-based strategic surgical capacity allocation. J Biomed Inform. 2019;94:103170.

18. Wachtel RE, Dexter F. Curriculum providing cognitive knowledge and problem-solving skills for anesthesia systems-based practice. J Grad Med Educ. 2010;2:624–632.

19. Dexter F, Epstein RH, Fahy BG, Van Swol LM. With directed study before a 4-day operating room management course, trust in the content did not change progressively during the classroom time. J Clin Anesth. 2017;42:57–62. pmid:28822884

20. Vasilopoulos T, Dexter F, Van Swol LM, Fahy BG. Trust improves during one-day resident operating room management course preceded by directed study of required statistical content. J Clin Anesth. 2019;55:43–49. pmid:30594680

21. Yesantharao P, Lee E, Kraenzlin F, Persing S, Chopra K, Shetty PN, et al. Surgical block time satisfaction: A multi-institutional experience across twelve surgical disciplines. Periop Care Oper Room Manag. 2020;21:100128.

22. Dexter F, Epstein RH, Ledolter J, Pearson AC, Maga J, Fahy BG. Benchmarking surgeons’ gender and year of medical school graduation associated with monthly operative workdays for multispecialty groups. Cureus 2022;14:e25054.

23. Dexter F, Macario A, Traub RD, Hopwood M, Lubarsky DA. An operating room scheduling strategy to maximize the use of operating room block time: Computer simulation of patient scheduling and survey of patients’ preferences for surgical waiting time. Anesth Analg. 1999;89:7-20.

24. Costa AS Jr. Assessment of operative times of multiple surgical specialties in a public university hospital. Einstein (São Paulo). 2017;15:200–205. pmid:28767919

25. Dexter F, Jarvie C, Epstein RH. Lack of generalizability of observational studies’ findings for turnover time reduction and growth in surgery based on the State of Iowa, where from one year to the next, most growth was attributable to surgeons performing only a few cases per week. J Clin Anesth. 2018;44:107–113. pmid:29175752

26. Epstein RH, Dexter F, Diez C, Fahy BG. Elective surgery growth at Florida hospitals accrues mostly from surgeons averaging 2 or fewer cases per week: a retrospective cohort study. J Clin Anesth. 2022;78:1106492022.

27. Boggs SD, Tsai MH, Urman RD. The Association of Anesthesia Clinical Directors (AACD) glossary of times used for scheduling and monitoring of diagnostic and therapeutic procedures. J Med Syst. 2018;42:171. pmid:30097795

28. Dexter F, Macario A, Traub RD, Lubarsky DA: Operating room utilization alone is not an accurate metric for the allocation of operating room block time to individual surgeons with low caseloads. Anesthesiology. 2003;98:1243–1249. pmid:12717148

29. 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 healthcare system. Anesth Analg. 2012; 115:395–401. pmid:22610848

30. Provider data catalog: doctors and clinicians data dictionary. Available from: https://data.cms.gov/provider-data/sites/default/files/data_dictionaries/DOC_Data_Dictionary.pdf. Accessed November 3, 2021.

31. Divine G, Norton HJ, Hunt R, Dienemann J. Statistical grand rounds: a review of analysis and sample size calculation considerations for Wilcoxon tests. Anesth Analg. 2013;117:699–710. pmid:23456667

32. Dexter F. Wilcoxon-Mann-Whitney test used for data that are not normally distributed. Anesth Analg. 2013;117:537–538. pmid:23966647

33. Dexter F, Epstein RH, Podgorski EM III, Pearson AC. Appropriate operating room time allocations and half-day block time for low caseload proceduralists, including anesthesiologist pain medicine physicians in the State of Florida. J Clin Anesth. 2020;64:109817. pmid:32353806

34. U.S. Equal Employment Opportunity Commission. Prohibited employment policies/ practices. Available from: https://www.eeoc.gov/prohibited-employment-policiespractices. Accessed November 4, 2021

35. Strum DP, Vargas LG, May JH, Bashein G. Surgical suite utilization and capacity planning: a minimal cost analysis model. J Med Syst. 1997;21:309–322. pmid:9507407

36. 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: tutorial using data from an Australian hospital. Anesth Analg. 2006;103:1499–1516.

37. Dexter F, Traub RD, Macario A. How to release allocated operating room time to increase efficiency: predicting which surgical service will have the most underutilized operating room time. Anesth Analg. 2003;96:507–512. pmid:12538204

38. Dexter F, Macario A. When to release allocated operating room time to increase operating room efficiency. Anesth Analg. 2004;98:758–762. pmid:14980933

39. Dexter F, Shi P, Epstein RH. Descriptive study of case scheduling and cancellations within 1 week of the day of surgery. Anesth Analg. 2012;115:1188–1195. pmid:23011558

40. Wachtel RE, Dexter F. Reducing tardiness from scheduled start times by making adjustments to the operating room schedule. Anesth Analg. 2009;108:1902–1909. pmid:19448220

41. Dexter F, Epstein RH, Schwenk ES. Tardiness of starts of surgical cases is not substantively greater when the preceding surgeon in an operating room is of a different versus the same specialty. J Clin Anesth. 2019;53:20–26. pmid:30290278

42. Dexter F, Traub RD, Lebowitz P. Scheduling a delay between different surgeons’ cases in the same operating room on the same day using upper prediction bounds for case durations. Anesth Analg. 2001;92:943–946. pmid:11273931

43. 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;105:1701–1710. pmid:18042870

44. Dexter F, Bayman EO, Pattillo JCS, Schwenk ES, Epstein RH. Influence of parameter uncertainty on the tardiness of the start of a surgical case following a preceding surgical case performed by a different surgeon. Periop Care Oper Room Manag. 2018;13:12–17.

45. Epstein RH, Dexter F, Smaka TJ. Obtaining and modeling variability in travel times from off site satellite clinics to hospitals and surgery centers for surgeons and proceduralists seeing office patients in the morning and performing a to-follow list of cases in the afternoon. Anesth Analg. 2020;131:228–238. pmid:30998561

46. Dossa F, Zeltzer D, Sutradhar R, Simpson AN, Baxter NN. Sex differences in the pattern of patient referrals to male and female surgeons. JAMA Surg. 2021;157:95–103.

47. Masursky D, Dexter F, Isaacson SA, Nussmeier NA. Surgeons’ and anesthesiologists’ perceptions of turnover times. Anesth Analg. 2011;112:440–444. pmid:21212255

48. Epstein RH, Dexter F. Rescheduling of previously cancelled surgical cases does not increase variability in operating room workload when cases are scheduled based on maximizing efficiency of use of operating room time. Anesth Analg. 2013;117:995–1002. pmid:24023019

49. Dexter F, Masursky D, Ledolter J, Wachtel RE, Smallman B. Monitoring changes in individual surgeon’s workloads using anesthesia data. Can J Anesth. 2012;59:571–577. pmid:22431148

50. Day R, Garfinkel R, Thompson S. Integrated block sharing: a win–win strategy for hospitals and surgeons. Manuf Serv Oper Manag. 2012;14:567–583.

51. Association of American Medical Colleges. Matriculating student questionnaire, 2018 all schools summary report. Available from: https://www.aamc.org/media/9641/download. Accessed 23 January 2022.