The nature and sources of variability in pediatric surgical case duration

Determining pre-procedure fasting alert time using procedural and scheduling data

Before a medical procedure requiring anesthesia, patients are required to not eat or drink non-clear fluids for 6 h and not drink clear fluids for 2 h. Fasting durations in standard practice far exceed these minimum thresholds due to uncertainties in procedure start time. The aim of this retrospective, observational study was to compare fasting durations arising from standard practice with different approaches for calculating the timepoint at which patients are instructed to stop eating and drinking. Scheduling data for procedures performed in the cardiac catheterization laboratory of an academic hospital in Canada (January 2020 to April 2022) were used. Four approaches utilizing machine learning (ML) and simulation were used to predict procedure start times and calculate when patients should be instructed to start fasting. Median fasting duration for standard practice was 10.08 h (IQR 3.5) for both food and clear fluids intake. The best performing alternative approach, using tree-based ML models to predict procedure start time, reduced median fasting from food/non-clear fluids to 7.7 h (IQR 2) and clear liquids fasting to 3.7 h (IQR 2.4). 97.3% met the minimum fasting duration requirements (95% CI 96.9% to 97.6%). Further studies are required to determine the effectiveness of operationalizing this approach as an automated fasting alert system.

A Comparison of Surgeon Estimated Times and Actual Operative Times in Pediatric Dental Rehabilitation under General Anesthesia. A Retrospective Study

This retrospective study aimed to compare the accuracy of the pediatric dental surgeon's estimated operative times for dental rehabilitation under general anesthesia (DRGA) in pediatric patients. This study population included 674 pediatric patients who underwent DRGA at the study facility between January 2022 and December 2022, using convenience sampling to select patients who met our inclusion criteria. Data were collected from electronic medical and anesthesia records based on several factors, including patient-related factors such as age and gender, surgeon-related factors such as rank and experience, and anesthesia-related factors such as induction and recovery time (in minutes). This study highlights a significant difference between the surgeon's estimated time (SET) and actual operative time (AOT) for pediatric DRGA procedures, with a mean difference of 19.28 min (SD = 43.17, p < 0.0001), indicating a tendency for surgeons to overestimate surgery time. Surgical procedure time was the strongest predictor of this discrepancy, with an R square value of 0.427 and a significant p-value of 0.000. Experience with surgeons, anesthesia induction, and recovery time were also significant predictors. Meanwhile, age, gender, and rank of surgeons did not significantly predict the difference between SET and AOT. Therefore, the study suggests that surgeons should adjust their estimates for pediatric DRGA procedures, specifically emphasizing a more accurate estimation of surgery time, to ensure adequate resource allocation and patient outcomes.

Predicting surgical department occupancy and patient length of stay in a paediatric hospital setting using machine learning: a pilot study

Objective

Early and accurate prediction of hospital surgical-unit occupancy is critical for improving scheduling, staffing and resource planning. Previous studies on occupancy prediction have focused primarily on adult healthcare settings, we sought to develop occupancy prediction models specifically tailored to the needs and characteristics of paediatric surgical settings.

Materials and methods

We conducted a single-centre retrospective cohort study at a surgical unit in a tertiary-care paediatric hospital in Boston, Massachusetts, USA. We developed a hierarchical modelling framework for predicting next-day census using multiple types of data—from bottom-up patient-specific orders and procedures to top-down temporal variables and departmental admission statistics.

Results

The model predicted upcoming admissions and discharges with a median error of 17%–21% (2–3 patients per day), and next-day census with a median error of 7% (n=3). The primary factors driving these predictions included day of week and scheduled surgeries, as well as procedure duration, procedure type and days since admission. We found that paediatric surgical procedure duration was highly predictive of postoperative length of stay.

Discussion

Our hierarchical modelling framework provides an overview of the factors driving capacity issues in the paediatric surgical unit, highlighting the importance of both top-down temporal features (eg, day of week) as well as bottom-up electronic health records (EHR)derived features (eg, orders for patient) for predicting next-day census. In the practice, this framework can be implemented stepwise, from top to bottom, making it easier to adopt.

Conclusion

Modelling frameworks combining top-down and bottom-up features can provide accurate predictions of next-day census in a paediatric surgical setting.

Omuz bölgesi yağ kalınlığının rotator manşet operasyonlarının süresine etkisi var mıdır?

Objective: Accurate estimation of operation time will reduce operating room costs and increase patient satisfaction. In recent studies, authors have found that thicker adipose tissue at the operation site is associated with higher rate of complications. However, there is no study in the literature investigating the effect on operation time of an increase in adipose tissue thickness. This present study hypothesised that thicker adipose tissue in the shoulder surgery would prolong the operation time, so the study was planned accordingly. Material and Methods: Preoperative magnetic resonance images of patients applied with rotator cuff repair between 2015 and 2020 were independently evaluated by two observers. The acromial fat thickness was measured as the fat thickness of the operation area, and the scapular fat tissue thickness as the fat thickness of the region relatively far from the operation area. The data obtained were evaluated using multivariate analysis and a binary logistic regression model. Results: Evaluation was made of a total of 106 patients. The mean total operation time was 89±33 mins. The mean acromial fat thickness was 12.2±4.89 mm and mean scapular fat thickness was 27.9±12.5mm. The increase in acromial fat thickness was determined to have extended the operation time (OR=5.75, 29.21, p

Combining adult with pediatric patient data to develop a clinical decision support tool intended for children: leveraging machine learning to model heterogeneity

Background

Clinical decision support (CDS) tools built using adult data do not typically perform well for children. We explored how best to leverage adult data to improve the performance of such tools. This study assesses whether it is better to build CDS tools for children using data from children alone or to use combined data from both adults and children.

Methods

Retrospective cohort using data from 2017 to 2020. Participants include all individuals (adults and children) receiving an elective surgery at a large academic medical center that provides adult and pediatric services. We predicted need for mechanical ventilation or admission to the intensive care unit (ICU). Predictor variables included demographic, clinical, and service utilization factors known prior to surgery. We compared predictive models built using machine learning to regression-based methods that used a pediatric or combined adult-pediatric cohort. We compared model performance based on Area Under the Receiver Operator Characteristic.

Results

While we found that adults and children have different risk factors, machine learning methods are able to appropriately model the underlying heterogeneity of each population and produce equally accurate predictive models whether using data only from pediatric patients or combined data from both children and adults. Results from regression-based methods were improved by the use of pediatric-specific data.

Conclusions

CDS tools for children can successfully use combined data from adults and children if the model accounts for underlying heterogeneity, as in machine learning models.

Average and longest expected treatment times for ultraviolet light disinfection of rooms

BACKGROUND

Planning Ultraviolet-C (UV-C) disinfection of operating rooms (ORs) is equivalent to scheduling brief OR cases. The study purpose was evaluation of methods for predicting surgical case duration applied to treatment times for ORs and hospital rooms.

METHODS

Data used were disinfection times with a 3-tower UV-C disinfection system in N=700 rooms each with ≥100 completed treatments.

RESULTS

The coefficient of variation of mean treatment duration among rooms was 19.6% (99% confidence interval [CI] 18.2%-21.0%); pooled mean 18.3 minutes among the 133,927 treatments. The 50^th percentile of coefficients of variation among treatments of the same room was 27.3% (CI 26.3%-28.4%), comparable to variabilities in durations of surgical procedures. The ratios of the 90^th percentile to mean differed among rooms. Log-normal distributions had poor fits for 33% of rooms. Combining results, we calculated 90% upper prediction limits for treatment times by room using a distribution-free method (e.g., third longest of preceding 29 durations). This approach was suitable because, once UV-C disinfection started, the median difference between the duration estimated by the system and actual time was 1 second.

CONCLUSIONS

Times for disinfection should be listed as treatment of a specific room (e.g., "UV-C main OR16"), not generically (e.g., "UV-C"). For estimating disinfection time after single surgical cases, use distribution-free upper prediction limits, because of considerable proportional variabilities in duration.

Similarities Between Pediatric and General Hospitals Based on Fundamental Attributes of Surgery Including Cases Per Surgeon Per Workday

Introduction

Operating room (OR) management decision-making at both pediatric and adult hospitals is determined, in large part, by the same fundamental attributes of surgery and other considerations related to case duration prediction. These include the non-preemptive nature of surgeries, wide prediction limits for case duration, and constraints to moving or resequencing cases on the day of surgery. Another attribute fundamentally affecting OR management is the median number of cases a surgeon performs on their OR days. Most adult surgeons have short lists of cases (i.e., one or two cases per day). Similarly, at adult hospitals, growth in caseloads is mostly due to the subset of those surgeons who also operate just once or twice per week. It is unknown if these characteristics of surgery apply to pediatric surgeons and pediatric hospitals as well.

Methods

Our retrospective cohort study included all elective surgical cases performed at the six pediatric hospitals in Florida during 2018 and 2019 (n = 71,340 cases). We calculated the percentages of combinations of surgeon, date, and hospital (lists) comprising one or two cases, or just one case, and determined if the values were statistically >50% (i.e., indicative of “most”). We determined if most of the growth in caseload and intraoperative work relative value units (wRVUs) at the pediatric hospitals between 2018 and 2019 accrued from low-caseload surgeons. Results are reported as mean ± standard error of the mean.

Results

Averaging among the six pediatric hospitals, the non-holiday weekday lists of most surgeons at each facility had just one or two elective cases, inpatient and/or ambulatory (68.1%; p = 0.016 vs. 50%, n = 27,557 lists). Growth in surgical caseloads from 2018 to 2019 was mostly attributable to surgeons who in 2018 averaged ≤2.0 cases per week (76.3% ± 5.4%, p = 0.0085 vs. 50%). Similarly, growth in wRVUs was mostly attributable to these low-caseload surgeons (73.8% ± 5.4%, p = 0.017 vs. 50%).

Conclusions

Like adult hospitals, most pediatric surgeons’ lists of cases consist of only one or two cases per day, with many lists containing a single case. Similarly, growth at pediatric hospitals accrued from low-caseload surgeons who performed one or two cases per week in the preceding year. These findings indicate that hospitals desiring to increase their surgical caseload should ensure that low-caseload surgeons are provided access to the OR schedule. Additionally, since percent-adjusted utilization and raw utilization cannot be accurately measured for low-caseload surgeons, neither metric should be used to allocate OR time to individual surgeons. Since most adult and pediatric surgeons have low caseloads, this is a fundamental attribute of surgery.

Percentages of Cases in Operating Rooms of Sufficient Duration to Accommodate a 30-Minute Breast Milk Pumping Session by Anesthesia Residents or Nurse Anesthetists

INTRODUCTION

Accommodating breast milk pumping sessions is required by US federal statute, but fulfillment is challenging for US anesthesia providers (e.g., anesthesia residents and nurse anesthetists). Considerations of good anesthesia practices (e.g., being present for critical portions of cases, including induction and emergence) create limits on which procedures are suitable for such relief. Our objective was to quantify the minimum percentages of cases for which there could reliably (≥ 95%) be at least 30 minutes during the surgical time when the anesthesia provider could receive such breaks.

METHODS

We studied all surgical cases performed at an anesthesia department over four years, including its inpatient surgical suite, pediatric hospital, and ambulatory surgery center. The 5% lower prediction bounds of surgical times (surgery or procedure start to end) were calculated from three years of historical data (October 1, 2016, to September 30, 2019) based on two-parameter lognormal distributions. The prediction bounds were compared to actual surgical start times during the next one year (October 1, 2019, to September 30, 2020). We considered the interval available for a breast milk pumping session during a case to be from 15 minutes after the start of the surgical time (to allow completion of initial documentation, other activities, and hand-off to the relieving anesthesia provider) until the end of the surgical time.

RESULTS

The lower prediction bounds were accurate, with 4.9% (4.6% - 5.2%) of future cases' surgical times being briefer, matching the nominal 5.0% rate. Applying these bounds, approximately 39% of cases (99% confidence interval 39% - 40%) were reliably of sufficient duration for the anesthesia provider delivering care in that one operating room to receive a 30-minute break for breast milk pumping session between 15 minutes after the start of surgery and procedure end. This percentage (39%) was substantially less than the 72% of the surgical times that were observed, in retrospect, to be sufficiently long because the lower 5% prediction bounds accounted correctly for the uncertainty in the duration of each case. The observed 39% prevalence was significantly fewer than half the cases (P < 0.0001 vs. 50%) suitable for such relief.

CONCLUSIONS

Individuals making operating room assignments for anesthesia providers need to consider the 5% lower prediction bounds of surgical times for cases in the room when making such assignments for women who require time for breast milk pumping sessions. Such considerations will generally result in assignments to rooms with one or more long-duration cases. Such a strategy may involve changes in preferred assignments for the anesthesia providers and alteration in the order of rotations for anesthesia residents (e.g., palliative care rotation rather than transition to practice at a pediatric ambulatory surgery center). When making room assignments for anesthesia providers who are breastfeeding, our results show that the 5% lower prediction bounds of surgical times need to be calculated; relying on the average surgical times for procedures is insufficient. Our paper also shows how to perform the mathematics using a spreadsheet program or equivalent.

Probabilistic forecasting of surgical case duration using machine learning: model development and validation

OBJECTIVE

Accurate estimations of surgical case durations can lead to the cost-effective utilization of operating rooms. We developed a novel machine learning approach, using both structured and unstructured features as input, to predict a continuous probability distribution of surgical case durations.

MATERIALS AND METHODS

The data set consisted of 53 783 surgical cases performed over 4 years at a tertiary-care pediatric hospital. Features extracted included categorical (American Society of Anesthesiologists [ASA] Physical Status, inpatient status, day of week), continuous (scheduled surgery duration, patient age), and unstructured text (procedure name, surgical diagnosis) variables. A mixture density network (MDN) was trained and compared to multiple tree-based methods and a Bayesian statistical method. A continuous ranked probability score (CRPS), a generalized extension of mean absolute error, was the primary performance measure. Pinball loss (PL) was calculated to assess accuracy at specific quantiles. Performance measures were additionally evaluated on common and rare surgical procedures. Permutation feature importance was measured for the best performing model.

RESULTS

MDN had the best performance, with a CRPS of 18.1 minutes, compared to tree-based methods (19.5-22.1 minutes) and the Bayesian method (21.2 minutes). MDN had the best PL at all quantiles, and the best CRPS and PL for both common and rare procedures. Scheduled duration and procedure name were the most important features in the MDN.

CONCLUSIONS

Using natural language processing of surgical descriptors, we demonstrated the use of ML approaches to predict the continuous probability distribution of surgical case durations. The more discerning forecast of the ML-based MDN approach affords opportunities for guiding intelligent schedule design and day-of-surgery operational decisions.

Operating room efficiency

The operating room is the financial hub of any hospital, and maximizing operating room efficiency has important implications for cost savings, patient satisfaction, and medical team morale. Over the past decade, manufacturing principles and processes such as Lean and Six Sigma have been applied to various aspects of healthcare including the operating room. Although time consuming, process mapping and deep examinations of each step of the patient journey from pre-operative visit to post-operative discharge can have multiplicative benefits that extend from cost savings to maintaining the focus on improving quality and patient safety.

Predictive Model of Surgical Time for Revision Total Hip Arthroplasty

BACKGROUND

Maximizing operating room utilization in orthopedic and other surgeries relies on accurate estimates of surgical control time (SCT). A variety of case and patient-specific variables can influence the duration of surgical time during revision total hip arthroplasty (THA). We hypothesized that these variables are better predictors of actual SCT (aSCT) than a surgeon's own prediction (pSCT).

METHODS

All revision THAs from October 2008 to September 2014 from one institution were accessed. Variables for each case included aSCT, pSCT, patient age, gender, body mass index, American Society of Anesthesiologists Physical Status class, active infection, periprosthetic fracture, bone loss, heterotopic ossification, and implantation/explantation of a well-fixed acetabular/femoral component. These were incorporated in a stepwise fashion into a multivariate regression model for aSCT with a significant cutoff of 0.15. This was compared to a univariate regression model of aSCT that only used pSCT.

RESULTS

In total, 516 revision THAs were analyzed. After stepwise selection, patient age and American Society of Anesthesiologists Physical Status were excluded from the model. The most significant increase in aSCT was seen with implantation of a new femoral component (24.0 min), followed by explantation of a well-fixed femoral component (18.7 min) and significant bone loss (15.0 min). Overall, the multivariate model had an improved r² of 0.49, compared to 0.16 from only using pSCT.

CONCLUSION

A multivariate regression model can assist surgeons in more accurately predicting the duration of revision THAs. The strongest predictors of increased aSCT are explantation of a well-fixed femoral component, placement of an entirely new femoral component, and presence of significant bone loss.

Use of Historical Surgical Times to Predict Duration of Primary Total Knee Arthroplasty

BACKGROUND

Primary total knee arthroplasty (TKA) is one of the most commonly performed procedures at US hospitals. Surgeons are typically asked to estimate surgical control time (SCT) needed for the procedure. Here, we compare the performance of a surgeon's prediction against a potentially more accurate method of using historical averages over the last 3, 5, 10, and 20 cases.

METHODS

Data were collected on all scheduled primary TKAs done at one institution from October 2008 to September 2014. For each case, actual SCT (aSCT) and predicted SCT were obtained. Historical SCTs were calculated based on the mean of the last 3, 5, 10, and 20 aSCTs of the same surgeon. Estimation biases were calculated based on the difference between aSCT and predicted SCT or between aSCT and historical estimates. Values were compared using Kruskal-Wallis analysis of variance and Steel-Dwass pairwise comparisons.

RESULTS

A total of 2539 primary TKAs were evaluated across 9 surgeons. Surgeons overestimated SCT by an average of 18.1 minutes. Using 3-20 cases in the historical average reduced mean estimation bias to a range of -0.1 to -0.3 minutes (P < .001). None of the historical estimations were significantly different from each other, demonstrating a lack of improvement with additional cases (P < .001).

CONCLUSION

Historical averages of procedure times appear to be a promising method of estimating surgical time for primary TKAs. Here, we show that even a small number of cases (eg, 3) can reduce estimation biases compared to solely using surgeons' estimates alone.

The Impact of Overestimations of Surgical Control Times Across Multiple Specialties on Medical Systems

Optimization of operating room (OR) resources and costs require accurate estimations of procedure times. In some institutions, surgeons are asked to predict their surgical control time (SCT), which typically constitutes the majority of the total procedure time. Here, we examine differences in predicted versus actual SCT and variations by specialty. Data included all scheduled surgical procedures at one academic institution from October 2008 - September 2014. Exclusion criteria consisted of missing values for SCT as well as estimated SCTs that fell outside of 5 standard deviations of any given procedure's mean. Differences in estimation were calculated by subtracting estimated SCT from actual SCT and compared against a null hypothesis of 0 with a two-tailed t-test. Differences between specialties were examined using analysis of variance and Games-Howell tests. Between 2008 and 2014, 119,410 scheduled procedures were performed. After exclusion, 116,599 cases were analyzed. On average, SCT was overestimated by 12.9 min (p < 0.0001). Overestimations persisted when divided by specialty (p < 0.0001). With thoracic surgery as a control, all other specialties except for cardiac surgery had overestimations of SCT. The greatest time differences were seen in dental (37.6 min, p < 0.0001), cardiology (33.0 min, p < 0.0001), and neurosurgery (29.7 min, p < 0.0001). Overall, SCTs are overestimated at this institution across many specialties. Depending on the methodology by which a hospital chooses to allocate OR time, SCT estimations could potentially be reduced in certain specialties to allow for better allocation of OR resources.