A robust estimation model for surgery durations with temporal, operational, and surgery team effects

Quantifying the impact of surgical teams on each stage of the operating room process

Introduction

Operating room (OR) efficiency is a key factor in determining surgical healthcare costs. To enable targeted changes for improving OR efficiency, a comprehensive quantification of the underlying sources of variability contributing to OR efficiency is needed. Previous literature has focused on select stages of the OR process or on aggregate process times influencing efficiency. This study proposes to analyze the OR process in more fine-grained stages to better localize and quantify the impact of important factors.

Methods

Data spanning from 2019-2023 were obtained from a surgery center at a large academic hospital. Linear mixed models were developed to quantify the sources of variability in the OR process. The primary factors analyzed in this study included the primary surgeon, responsible anesthesia provider, primary circulating nurse, and procedure type. The OR process was segmented into eight stages that quantify eight process times, e.g., procedure duration and procedure start time delay. Model selection was performed to identify the key factors in each stage and to quantify variability.

Results

Procedure type accounted for the most variability in three process times and for 44.2% and 45.5% of variability, respectively, in procedure duration and OR time (defined as the total time the patient spent in the OR). Primary surgeon, however, accounted for the most variability in five of the eight process times and accounted for as much as 21.1% of variability. The primary circulating nurse was also found to be significant for all eight process times.

Discussion

The key findings of this study include the following. (1) It is crucial to segment the OR process into smaller, more homogeneous stages to more accurately assess the underlying sources of variability. (2) Variability in the aggregate quantity of OR time appears to mostly reflect the variability in procedure duration, which is a subinterval of OR time. (3) Primary surgeon has a larger effect on OR efficiency than previously reported in the literature and is an important factor throughout the entire OR process. (4) Primary circulating nurse is significant for all stages of the OR process, albeit their effect is small.

Dissatisfaction-considered waiting time prediction for outpatients with interpretable machine learning

Long waiting time in outpatient departments is a crucial factor in patient dissatisfaction. We aim to analytically interpret the waiting times predicted by machine learning models and provide patients with an explanation of the expected waiting time. Here, underestimating waiting times can cause patient dissatisfaction, so preventing this in predictive models is necessary. To address this issue, we propose a framework considering dissatisfaction for estimating the waiting time in an outpatient department. In our framework, we leverage asymmetric loss functions to ensure robustness against underestimation. We also propose a dissatisfaction-aware asymmetric error score (DAES) to determine an appropriate model by considering the trade-off between underestimation and accuracy. Finally, Shapley additive explanation (SHAP) is applied to interpret the relationship trained by the model, enabling decision makers to use this information for improving outpatient service operations. We apply our framework in the endocrinology metabolism department and neurosurgery department in one of the largest hospitals in South Korea. The use of asymmetric functions prevents underestimation in the model, and with the proposed DAES, we can strike a balance in selecting the best model. By using SHAP, we can analytically interpret the waiting time in outpatient service (e.g., the length of the queue affects the waiting time the most) and provide explanations about the expected waiting time to patients. The proposed framework aids in improving operations, considering practical application in hospitals for real-time patient notification and minimizing patient dissatisfaction. Given the significance of managing hospital operations from the perspective of patients, this work is expected to contribute to operations improvement in health service practices.

A novel approach to forecast surgery durations using machine learning techniques

This study presents a methodology for predicting the duration of surgical procedures using Machine Learning (ML). The methodology incorporates a new set of predictors emphasizing the significance of surgical team dynamics and composition, including experience, familiarity, social behavior, and gender diversity. By applying ML techniques to a comprehensive dataset of over 77,000 surgeries, we achieved a 24% improvement in the mean absolute error (MAE) over a model that mimics the current approach of the decision maker. Our results also underscore the critical role of surgeon experience and team composition dynamics in enhancing prediction accuracy. These advancements can lead to more efficient operational planning and resource allocation in hospitals, potentially reducing downtime in operating rooms and improving healthcare delivery.

Predicting operative time for metabolic and bariatric surgery using machine learning models: a retrospective observational study

BACKGROUND

Predicting operative time is essential for scheduling surgery and managing the operating room. This study aimed to develop machine learning (ML) models to predict the operative time for metabolic and bariatric surgery (MBS) and to compare each model.

METHODS

The authors used the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program database between 2016 and 2020 to develop ML models, including linear regression, random forest, support vector machine, gradient-boosted tree, and XGBoost model. Patient characteristics and surgical features were included as variables in the model. The authors used the mean absolute error, root mean square error, and R 2 score to evaluate model performance. The authors identified the 10 most important variables in the best-performing model using the Shapley Additive exPlanations algorithm.

RESULTS

In total, 668 723 patients were included in the study. The XGBoost model outperformed the other ML models, with the lowest root mean square error and highest R 2 score. Random forest performed better than linear regression. The relative performance of the ML algorithms remained consistent across the models, regardless of the surgery type. The surgery type and surgical approach were the most important features to predict the operative time; specifically, sleeve gastrectomy (vs. Roux-en-Y gastric bypass) and the laparoscopic approach (vs. robotic-assisted approach) were associated with a shorter operative time.

CONCLUSIONS

The XGBoost model best predicted the operative time for MBS among the ML models examined. Our findings can be useful in managing the operating room scheduling and in developing software tools to predict the operative times of MBS in clinical settings.

Improving preoperative prediction of surgery duration

BACKGROUND

Operating rooms (ORs) are one of the costliest units in a hospital, therefore the cumulative consequences of any kind of inefficiency in OR management lead to a significant loss of revenue for the hospital, staff dissatisfaction, and patient care disruption. One of the possible solutions to improving OR efficiency is knowing a reliable estimate of the duration of operations. The literature suggests that the current methods used in hospitals, e.g., a surgeon's estimate for the given surgery or taking the average of only five previous records of the same procedure, have room for improvement.

METHODS

We used over 4 years of elective surgery records (n = 52,171) from one of the major metropolitan hospitals in Australia. We developed robust Machine Learning (ML) approaches to provide a more accurate prediction of operation duration, especially in the absence of surgeon's estimation. Individual patient characteristics and historic surgery information attributed to medical records were used to train predictive models. A wide range of algorithms such as Extreme Gradient Boosting (XGBoost) and Random Forest (RF) were tested for predicting operation duration.

RESULTS

The results show that the XGBoost model provided statistically significantly less error than other compared ML models. The XGBoost model also reduced the total absolute error by 6854 min (i.e., about 114 h) compared to the current hospital methods.

CONCLUSION

The results indicate the potential of using ML methods for reaching a more accurate estimation of operation duration compared to current methods used in the hospital. In addition, using a set of realistic features in the ML models that are available at the point of OR scheduling enabled the potential deployment of the proposed approach.

Accurate Operative Time Prediction in Thyroid Surgery: A Rural Tertiary Care Facility Experience

OBJECTIVE

To determine whether surgeons can estimate thyroid operative time more accurately than a system-generated average time estimate.

METHODS

Four otolaryngologists at a single institution with extensive endocrine surgery experience were asked to predict their operative times for all eligible thyroid surgeries. These estimates were compared to system-generated operative time predications based on averaging the surgeon's previous 10 cases with the same Current Procedural Terminology code. The surgeon-generated estimations and system-generated estimations were then compared to each other and the actual operative time.

RESULTS

A final sample of 73 cases was used for all analyses. Average age was 51 years old and the majority of patients were female. Surgeon-generated operative time estimates were significantly more accurate than system-generated estimates based on time averaging (P < .001). These findings were consistent across each surgeon individually and within each procedure type (hemithyroidectomy and total thyroidectomy). These findings had a power of over 99% based on mean differences.

CONCLUSION

As the financial center of modern hospitals, an efficient operating room is integral to economic success. Improving the precision of operative time estimation reduces costly unplanned staff overtime, canceled cases, and underutilization. Our research at a rural tertiary care center shows that experienced thyroid surgeons can substantially reduce the error of estimating thyroid operative times by considering individual patient characteristics. Although no objective variables have so far been identified to correlate with thyroid operative time, surgeon-generated operative time estimation is significantly more accurate than a generic system approach of averaging previous operative times.

Surgical procedure prediction using medical ontological information

BACKGROUND AND OBJECTIVE

Predicting the duration of surgical procedures is an important step in scheduling operating rooms. Many factors have been shown to influence the duration of a procedure, in this research we aim to use medical ontological information to improve the predictions.

METHODS

This paper presents two methods for incorporating the medical information about a surgical procedure into the prediction of the duration of the procedure. The first method uses the Systematised Nomenclature of Medicine Clinical Terms to relate different procedures to each other. The second uses simple text fragments. The relationships between types of procedures are included in a regression model for the procedure duration. These methods are applied to data from New Zealand healthcare facilities and the accuracy of the estimations of the durations is compared. In addition a simulation of scheduling the procedures in an operating room is performed.

RESULTS

It is shown that both of the methods provide an improvement in the prediction of procedure durations. When compared to a traditional categorical encoding, the ontological information provides an improvement in the continuous ranked probability scores of the prediction of procedure durations from 18.4 min to 17.1 min, and from 25.3 to 21.3 min for types of procedures that are not performed very often.

CONCLUSIONS

Different methods for encoding medical ontological information in surgery procedure duration predictions are presented, and show an improvement over traditional models. The improvement in duration prediction is shown to improve the efficiency of scheduling in a simple simulation.

Estimation of Surgery Durations Using Machine Learning Methods-A Cross-Country Multi-Site Collaborative Study

The scheduling of operating room (OR) slots requires the accurate prediction of surgery duration. We evaluated the performance of existing Moving Average (MA) based estimates with novel machine learning (ML)-based models of surgery durations across two sites in the US and Singapore. We used the Duke Protected Analytics Computing Environment (PACE) to facilitate data-sharing and big data analytics across the US and Singapore. Data from all colorectal surgery patients between 1 January 2012 and 31 December 2017 in Singapore and, 1 January 2015 to 31 December 2019 in the US were used, and 7585 cases and 3597 single and multiple procedure cases from Singapore and US were included. The ML models were based on categorical gradient boosting (CatBoost) models trained on common data fields shared by both institutions. The procedure codes were based on the Table of Surgical Procedure (TOSP) (Singapore) and the Current Procedural Terminology (CPT) codes (US). The two types of codes were mapped by surgical experts. The CPT codes were then transformed into the relative value unit (RVU). The ML models outperformed the baseline MA models. The MA, scheduled durations and procedure codes were found to have higher loadings as compared to surgeon factors. We further demonstrated the use of the Duke PACE in facilitating data-sharing and big data analytics.

Improving intraoperative handoffs for ambulatory anesthesia: challenges and solutions for the anesthesiologist

Permanent transitions of care from one anesthesia provider to another are associated with adverse events and mortality. There are currently no available data on how to mitigate these poor patient outcomes other than to reduce the occurrence of such handoffs. We used data from an ambulatory surgery center to demonstrate the steps that can be taken to achieve this goal. First, perform statistical forecasting using many months of historical data to create optimal, as opposed to arbitrary shift durations. Second, consider assigning the anesthesia providers designated to work late, if necessary, to the ORs estimated to finish the earliest, rather than latest. We performed multiple analyses showing the quantitative advantage of this strategy for the ambulatory surgery center with multiple brief cases. Third, sequence the cases in the 1 or 2 ORs with the latest scheduled end times so that the briefest cases are finished last. If a supervising anesthesiologist needs to be relieved early for administrative duties (eg, head of the group to meet with administrators or surgeons), assign the anesthesiologist to an OR that finishes with several brief cases. The rationale for these recommendations is that such strategies provide multiple opportunities for a different anesthesia provider to assume responsibility for the patients between cases, thus avoiding a handoff altogether.

RSDNet: Learning to Predict Remaining Surgery Duration from Laparoscopic Videos Without Manual Annotations

Accurate surgery duration estimation is necessary for optimal OR planning, which plays an important role in patient comfort and safety as well as resource optimization. It is, however, challenging to preoperatively predict surgery duration since it varies significantly depending on the patient condition, surgeon skills, and intraoperative situation. In this paper, we propose a deep learning pipeline, referred to as RSDNet, which automatically estimates the remaining surgery duration (RSD) intraoperatively by using only visual information from laparoscopic videos. The previous state-of-the-art approaches for RSD prediction are dependent on manual annotation, whose generation requires expensive expert knowledge and is time-consuming, especially considering the numerous types of surgeries performed in a hospital and the large number of laparoscopic videos available. A crucial feature of RSDNet is that it does not depend on any manual annotation during training, making it easily scalable to many kinds of surgeries. The generalizability of our approach is demonstrated by testing the pipeline on two large datasets containing different types of surgeries: 120 cholecystectomy and 170 gastric bypass videos. The experimental results also show that the proposed network significantly outperforms a traditional method of estimating RSD without utilizing manual annotation. Further, this paper provides a deeper insight into the deep learning network through visualization and interpretation of the features that are automatically learned.

Assessment of operative times of multiple surgical specialties in a public university hospital

Objective

To evaluate the indicators duration of anesthesia, operative time and time patients stay in the operating rooms of different surgical specialties at a public university hospital.

Methods

It was done by a descriptive cross-sectional study based on the operating room database. The following stages were measured: duration of anesthesia, procedure time and patient length of stay in the room of the various specialties. We included surgeries carried out in sequence in the same room, between 7:00 a.m. and 5 p.m., either elective or emergency. We calculated the 80th percentile of the stages, where 80% of procedures were below this value.

Results

The study measured 8,337 operations of 12 surgical specialties performed within one year. The overall mean duration of anesthesia of all specialties was 178.12±110.46 minutes, and the 80th percentile was 252 minutes. The mean operative time was 130.45±97.23 minutes, and the 80th percentile was 195 minutes. The mean total time of the patient in the operating room was 197.30±113.71 minutes, and the 80th percentile was 285 minutes. Thus, the variation of the overall mean compared to the 80th percentile was 41% for anesthesia, 49% for surgeries and 44% for operating room time. In average, anesthesia took up 88% of the operating room period, and surgery, 61%.

Conclusion

This study identified patterns in the duration of surgery stages. The mean values of the specialties can assist with operating room planning and reduce delays.

Effect of Individual Surgeons and Anesthesiologists on Operating Room Time

BACKGROUND

Variability in operating room (OR) time causes overutilization and underutilization of the available ORs. There is evidence that for a given type of procedure, the surgeon is the major source of variability in OR time. The primary aim was to quantify the variability between surgeons and anesthesiologists. As illustration, the value of modeling the individual surgeons and anesthesiologist for OR time prediction was estimated.

METHODS

OR data containing 16,480 cases were obtained from a general surgery department. The total amount of variability in OR time accounted for by the type of procedure, first and second surgeon, and the anesthesiologist was determined with the use of linear mixed models. The effect on OR time prediction was evaluated as reduction in overtime and idle time per case.

RESULTS

Differences between first surgeons can account for only 2.9% (2.0%-4.2%) of the variability in OR time. Differences between anesthesiologists can account for 0.1% (0.0%-0.3%) of the variability in OR time. Incorporating the individual surgeons and anesthesiologists led to an average reduction of overtime and idle time of 1.8 (95% confidence interval, 1.7-2.0, 10.5% reduction) minutes and 3.0 (95% confidence interval, 2.8%-3.2, 17.0% reduction) minutes, respectively.

CONCLUSIONS

In comparison with the type of procedure, differences between surgeons account for a small part of OR time variability. The impact of differences between anesthesiologists on OR time is negligible. A prediction model incorporating the individual surgeons and anesthesiologists has an increased precision, but improvements are likely too marginal to have practical consequences for OR scheduling.