Increasing costs of medical care have caused healthcare executives to search for new and creative ways to keep their hospitals in business. A 2018 survey of hospital executives identified cost control, expense reduction, exploration of diversified and innovative revenue streams, and increasing outpatient procedural market share as the top four concerns for their medical systems.1 Owing to an increased patient load, higher expenses, and fewer elective and outpatient procedures, a 2021 third-quarter projection indicated a 54-billion-dollar loss from hospitals nationwide—even when factoring in the massive government aid given in 2021.2

Boggs et al described potential solutions such as overbooking patients, block optimization, 24-hour staffing of operating rooms, and development of specific quality measures and standards that could increase hospital efficiency.3 Lex et al found that a dedicated orthopaedic trauma room increases hospital efficiency and improves finances.4,5 Similar conclusions regarding efficiency were made in the article by Martin et al., describing the use of dedicated spine team members for posterior spinal fusion procedures.6 Beyond a dedicated staff and space to better use the time and resources of hospitals, other studies have looked at the use of surgeon-preferred staff. Cousins et al found a decrease in operating room time when surgeons worked with preferred anesthesiologists or surgical technicians but an increase with preferred vendors.7 Staff turnover during the procedure also markedly increased operating room time.8 The current literature has investigated OR efficiency and logistical management in spine,6,8,9 knee and hip,7,10 and hand1,4,11 surgeries, with each study showing hope for either better financial and/or logistical outcomes. Chohan et al concluded that patients who were assigned to a high-efficiency OR reported similar patient-reported outcomes compared with patients in standard ORs,12 establishing that increased efficiency is both a safe and practical way to decrease costs in a healthcare system.

However, the existing literature lacks investigation of these principles of efficiency (defined here as the length of operating room time) as they relate to shoulder surgeries and quantitative patient outcomes (such as complication rates). Our study investigates this relationship, with a focus on reverse total shoulder arthroplasties (rTSAs) and anatomic total shoulder arthroplasties (aTSAs). With the steady economic pressure faced by hospitals and the promising possibilities of increased OR efficiency, this study provides a comprehensive look at the effect of use of surgeon-preferred staff, staff turnover, and presence of residents on operating room time and quantitative patient outcomes as it relates to rTSAs and anatomic total shoulder arthroplasties. We hypothesize that the use of surgeon-preferred staff will decrease operating room time and patient 90-day complication rates will be negatively affected by the presence of nonpreferred staff and high amounts of staff turnover in a given case.

Methods

This is a single-center, IRB-approved, retrospective study focused on determining the effects of staffing on operating room time and efficiency in total shoulder arthroplasty. The study included patients who underwent total shoulder arthroplasty (anatomic and reverse) by a single fellowship-trained orthopaedic surgeon from January 1, 2018, to January 1, 2023. Surgical information such as staff present in the operating room and total operating time and 90-day complication rates were collected through electronic medical records. Ninety-day complications included both shoulder-specific and non–shoulder-specific complications. Dichotomous variables, OR time > 60 minutes and ≥ 1 90-day complication, were used to run relative risk statistics. Sixty minutes was chosen as the cutoff time point for analysis because the mean OR time (for all TSA cases combined) was 69.56 ± 16.45 minutes, and 60 minutes is within one standard deviation below the mean. This way a target of 60 minutes would be within reason, yet there would still be an overall reduction in operating time for both rTSA and aTSA cases. OR time (as a continuous variable) was defined as the time from incision to closure. Data analysis was conducted using IBM SPSS (IBM Corp. Released 2021. IBM SPSS Statistics for Macintosh, Version 28.0: IBM Corp) to run frequency, relative risk, and multiple linear regression calculations. The chosen statistical analyses were verified as appropriate with the aid of Laerd Statistics, an SPSS statistics tutorial and software guide.13 Both anatomic and reverse total shoulder arthroplasties were included in the study; however, they were analyzed independently of each other.

Results

Four hundred twenty-three patients were included in the study from August 2018 to April 2023, 264 of which were rTSA and 159 were anatomic total shoulder arthroplasty (aTSA).

In the rTSA group, 63.63% of patients were female. The average age was 69.45 ± 8.78 years (Table 1). Frequencies of staff present during surgeries and surgical outcome variables (total operating room time and 90-day complication rate) are listed in Tables 2 and 3, respectively. The observed 90-day complications included surgical failure requiring revision arthroplasty, postoperative carpal tunnel, incisional abscess, cellulitis around the surgical site, deep surgical infection, nerve damage (postoperative numbness and paresthesias in nerve distribution), periprosthetic fracture, lymphedema, pulmonary infection or decompensation, and small bowel obstruction. Operating room time of greater than 60 minutes occurred in 58.3% of rTSA cases. 9.1% of rTSA patients had at least one complication at 90 days after surgery (Table 3). Continuous rTSA variable means are listed in Table 1. Relative risk ratios of staff present during rTSA cases on operating room time and 90-day complications are summarized in Table 4, including 95% confidence intervals. The presence of a surgeon-preferred circulatory nurse, surgical technician, and surgical assistant markedly reduced the risk of OR time being greater than 60 minutes by 28.8%, 41.4%, and 42.3%, respectively. If three of four surgeon-preferred staff were present during the case, risk of OR time > 60 minutes was reduced by 38.6%. By contrast, having both PGY3 and PGY5 residents in surgery together, more than one circulatory nurse, more than one surgical technician, and more than one staff turnover/break increased the risk of OR time being greater than 60 minutes by 30.8%, 22.9%, 25.9%, and 62.1%, respectively. No relative risk ratios for 90-day complications after rTSA were statistically significant (Table 4). To determine independent predictors of rTSA outcomes, multiple linear regression was used. The first model (Table 5) investigated rTSA OR time (P < 0.001). Both variables (number of circulators and number of residents present in OR) added statistically significantly to the prediction (P < 0.001). The following equation offers a partial prediction of the specific surgeon's rTSA OR time for a patient. However, it only captures a portion of the variability, as indicated by the R-squared value of 0.11: Reverse TSA OR time = 49.142 + 5.119 (number of circulators) + 7.018 (number of residents). Regression coefficients, standard errors, and R2 are provided in Table 5. The second model (Table 6) investigated rTSA 90-day complications (P = 0.007). Both variables (patient age and number of staff turnover/breaks) added statistically significantly to the prediction (P < 0.05). The provided equation offers a partial prediction of the specific surgeon's 90-day complication rate after rTSA for a patient. However, it only captures a portion of the variability, as indicated by the modest R-squared value of 0.037: Reverse TSA number of 90-day complications = 0.442−(0.005 × patient age in years) + 0.054 (number of staff turnovers/breaks during surgery). Regression coefficients, standard errors, and R2 are presented in Table 6. Although the equations may mitigate a lower variance in the independent variable, they still yield valuable insights. Each individual independent variable in the multiple linear regression model elicits a specific change (as indicated by the unstandardized B value) in the dependent variable for every one-unit adjustment in the independent variable.

Table 1 - Reverse Total Shoulder Arthroplasty Continuous Variable Means and Standard Deviations Variable Mean and Standard Deviation Per Surgical Case Patient age (years) 69.45 ± 8.78 Residents 1.25 ± 0.60 Operating room time (minutes) 65.46 ± 16.35 Circulatory nurses 1.47 ± 0.35 Surgical technicians 1.21 ± 0.45 Surgical assistants 0.93 ± 0.54 Staff turnover/breaks 0.42 ± 0.79 90-day complications 0.09 ± 0.34
Table 2 - Frequencies of Staff Present During Reverse Total Shoulder Arthroplasty Cases Staff Present During Surgery Frequency of Surgical Cases % of Total Surgical Cases Any resident (regardless of year) 241 91.3 PGY3 resident 214 81.1 PGY5 resident 112 42.4 PGY3 and PGY5 residents 85 32.2 Surgeon-preferred circulatory nurse 95 36 Surgeon-preferred surgical technician 101 38.3 Surgeon-preferred surgical assistant 118 44.7 Surgeon-preferred industry representative 124 47 All 4 surgeon-preferred staff 32 12.1 3 of 4 surgeon-preferred staff 54 20.5 2 of 4 surgeon-preferred staff 42 15.9 1 of 4 surgeon-preferred staff 64 24.2 > 1 circulatory nurse 110 41.7 > 1 surgical technician 53 20.1 > 1 surgical assistant 30 11.4 > 1 staff turnover/break 27 10.2 Total 264 100

PGY3 = postgraduate year 3, PGY5 = postgraduate year five

Table 3 - Frequencies of Surgical Outcome Variables in Reverse Total Shoulder Arthroplasty Cases Surgical Outcome Variables Frequency of Surgical Cases % of Total Surgical Cases Surgical time > 60 minutes 154 58.3 ≥ 1 90-day complication 24 9.1 Total 264 100
Table 4 - Relative Risk Findings of Staff Present During Reverse Total Shoulder Arthroplasty on Operating Room Time and 90-Day Complications Staff Present During Surgery Operating Room Time > 60 Minutes ≥ 1 90-Day Complications RRR 95% CI % risk RRR 95% CI % risk Any resident 1.506 0.922, 2.461 ↑50.6 0.732 0.235, 2.284 ↓26.8 PGY3 resident 1.331 0.968, 1.830 ↑33.1 0.467 0.212, 1.031 ↓53.3 PGY5 resident 1.223 1.000, 1.496 ↑22.3 1.148 0.534, 2.468 ↑14.8 PGY3 and PGY5 residents 1.308a 1.074, 1.593 ↑30.8 0.702 0.289, 1.704 ↓29.8 Surgeon-preferred circulatory nurse 0.712a 0.558, 0.907 ↓28.8 1.067 0.486, 2.345 ↑6.7 Surgeon-preferred surgical technician 0.586a 0.453, 0.757 ↓41.4 0.807 0.358, 1.817 ↓19.3 Surgeon-preferred surgical assistant 0.577a 0.456, 0.732 ↓42.3 0.509 0.219, 1.187 ↓49.1 Surgeon-preferred industry representative 0.941 0.766, 1.155 ↓5.9 1.334 0.620, 2.869 ↑33.4 All 4 surgeon-preferred staff 0.725 0.483, 1.089 ↓27.5 1.036 0.327, 3.277 ↑3.6 3 of 4 surgeon-preferred staff 0.614a 0.433, 0.871 ↓38.6 0.778 0.277, 2.181 ↓22.2 2 of 4 surgeon-preferred staff 0.835 0.605, 1.150 ↓16.5 0.755 0.236, 2.418 ↓24.5 1 of 4 surgeon-preferred staff 1.211 0.980, 1.496 ↑21.1 1.287 0.559, 2.963 ↑28.7 > 1 circulatory nurse 1.229a 1.006, 1.503 ↑22.9 1.655 0.770, 3.555 ↑65.5 > 1 surgical technician 1.259a 1.016, 1.560 ↑25.9 1.991 0.900, 4.401 ↑99.1 > 1 surgical assistant 0.968 0.695, 1.348 ↓3.2 1.560 0.572, 4.257 ↑56.0 > 1 staff turnover/break 1.621a 1.358, 1.933 ↑62.1 0.798 0.198, 3.209 ↓20.2

RRR = relative risk ratio, CI = confidence interval, ↑: increased, ↓: decreased

aDenotes statistically significant values, P < 0.05.

Table 5 - Multiple Linear Regression—Prediction of OR Time in Reverse Total Shoulder Arthroplasty Model Unstandardized B Coefficients Standard Error Standardized Coefficients Beta t P Constant 49.142 3.125 15.724 <0.001 Number of circulatory nurses 5.119 1.604 0.187 3.192 0.002 Number of residents 7.018 1.585 0.259 4.428 <0.001 R2 of model 0.11

Dependent variable: operating room time (minutes).

Equation: Reverse TSA OR time = 49.142 + 5.119 (number of circulators) + 7.018 (number of residents).

Table 6 - Multiple Linear Regression—Prediction of 90-Day Complications in Reverse Total Shoulder Arthroplasty Model Unstandardized B Coefficients Standard Error Standardized Coefficients Beta t P Constant 0.442 0.167 2.647 0.009 Patient age (years) −0.005 0.002 −0.137 −2.248 0.025 Number of staff turnovers/breaks 0.054 0.026 0.124 2.039 0.042 R2 of model 0.037

Dependent variable: number of complications in 90 days

Equation: Reverse TSA number of 90-day complications = 0.442−(0.005 × patient age in years) + 0.054 (number of staff turnovers/breaks during surgery).

In the aTSA group, 54.43% of patients were female. The average age was 63.51 ± 8.97 years (Table 7). Frequencies of staff present during aTSA surgeries and surgical outcome variables (total operating room time and 90-day complication rate) are listed in Tables 8 and 9, respectively. Operating room time of greater than 60 minutes occurred in 88% of aTSA cases. 11.4% of aTSA patients had at least one complication at 90 days after surgery (Table 9). Continuous aTSA variable means are listed in Table 9. Relative risk ratios of staff present during aTSA cases on operating room time and 90-day complications are summarized in Table 10, including 95% confidence intervals. The presence of surgeon-preferred circulatory nurse and surgical assistant markedly reduced the risk of OR time > 60 minutes by 15.5% and 15.3%, respectively. By contrast, having a PGY5 resident, both PGY3 and PGY5 residents in surgery together, and more than one surgical assistant increased the risk of OR time > 60 minutes by 13.7%, 18.7%, and 15.1%, respectively. No relative risk ratios for 90-day complications after aTSA were statistically significant (Table 10). To determine independent predictors of aTSA outcomes, multiple linear regression was used. The model statistically significantly predicted aTSA OR time (P < 0.001). All three variables (number of residents in the OR, number of circulators in the OR, and if surgeon-preferred circulator was in OR) added statistically significantly to the prediction (P < 0.001). The provided equation offers a partial prediction of the specific surgeon's aTSA OR time for a patient. However, it only captures a portion of the variability, as indicated by the modest R-squared value of 0.11: OR time = 58.018 + (6.058 × number of residents in surgery) + (3.990 × number of circulators present)−(5.220 × preferred circulator present*), *presence is measured by either present (1) or not present (0). Regression coefficients, standard errors, and R2 are presented in Table 11. Again, this equation, despite addressing lower variance in the independent variable, still provides valuable insights. Each independent variable in the multiple linear regression model induces a specific change in the dependent variable with a one-unit adjustment.

Table 7 - Anatomic Total Shoulder Arthroplasty Continuous Variable Means and Standard Deviations Variable Mean and Standard Deviations Per Surgical Case Patient age (years) 63.51 ± 8.97 Residents 1.23 ± 0.65 Operating room time (minutes) 76.42 ± 14.24 Circulatory nurses 1.56 ± 0.70 Surgical technicians 1.25 ± 0.48 Surgical assistants 0.99 ± 0.42 Staff turnover/breaks 0.50 ± 0.92 90-day complications 0.11 ± 0.41
Table 8 - Frequencies of Staff Present During Anatomic Total Shoulder Arthroplasty Cases Staff Present During Surgery Frequency of Surgical Cases % of Total Surgical Cases Any resident (regardless of year) 138 87.3 PGY3 resident 123 77.8 PGY5 resident 70 44.3 PGY3 and PGY5 residents 54 34.2 Surgeon-preferred circulatory nurse 69 43.7 Surgeon-preferred surgical technician 74 46.8 Surgeon-preferred surgical assistant 77 48.7 Surgeon-preferred industry representative 81 51.3 All 4 surgeon-preferred staff 28 17.7 3 of 4 surgeon-preferred staff 35 22.2 2 of 4 surgeon-preferred staff 27 17.1 1 of 4 surgeon-preferred staff 30 19.0 > 1 circulatory nurse 75 47.5 > 1 surgical technician 37 23.4 > 1 surgical assistant 13 8.2 > 1 staff turnover/break 21 13.3 Total 159 100

PGY3 = postgraduate year 3, PGY5 = postgraduate year 5

Table 9 - Frequencies of Surgical Outcome Variables in Anatomic Total Shoulder Arthroplasty Cases Surgical Outcome Variables Frequency of Surgical Cases % of Total Surgical Cases Surgical time > 60 minutes 139 88 ≥ 1 90-day complication 18 11.4 Total 159 100
Table 10 - Relative Risk Findings of Staff Present During Anatomic Total Shoulder Arthroplasty on Operating Room Time and 90-Day Complications Staff Present During Surgery Operating Room Time > 60 minutes ≥ 1 90-Day Complication RRR 95% CI % risk RRR 95% CI % risk Any resident 0.974 0.831, 1.142 ↓6.6 0.725 0.230, 2.283 ↓27.5 PGY3 resident 1.034 0.890, 1.200 ↑3.4 0.996 0.350, 2.834 ↑99.6 PGY5 resident 1.137a 1.017, 1.270 ↑13.7 0.629 0.248, 1.590 ↑62.9 PGY3 and PGY5 residents 1.187a 1.079, 1.306 ↑18.7 0.741 0.279, 1.969 ↓25.9 Surgeon-preferred circulatory nurse 0.845a 0.742, 0.961 ↓15.5 1.612 0.672, 3.867 ↑61.2 Surgeon-preferred surgical technician 0.914 0.811, 1.030 ↓8.6 1.135 0.476, 2.708 ↑13.5 Surgeon-preferred surgical assistant 0.847a 0.751, 0.956 ↓15.3 0.526 0.208, 1.332 ↓47.4 Surgeon-preferred industry representative 0.993 0.885, 1.114 ↓0.7 0.760 0.317, 1.826 ↓24.0 All 4 surgeon-preferred staff 0.921 0.767, 1.105 ↓7.9 0.929 0.288, 2.993 ↓7.1 3 of 4 surgeon-preferred staff 0.967 0.833, 1.123 ↓3.3 0.703 0.216, 2.290 ↓29.7 2 of 4 surgeon-preferred staff 0.815 0.648, 1.026 ↓18.5 1.866 0.726, 4.799 ↑86.6 1 of 4 surgeon-preferred staff 1.029 0.898, 1.178 ↑2.9 0.533 0.130, 2.196 ↓46.7 > 1 circulatory nurse 0.972 0.865, 1.092 ↓2.8 0.885 0.369, 2.125 ↓11.5 > 1 surgical technician 0.978 0.848, 1.128 ↓2.2 0.654 0.200, 2.136 ↓34.6 > 1 surgical assistant 1.151a 1.080, 1.226 ↑15.1 0.656 0.095, 4.544 ↓34.4 > 1 staff turnover/break