[Paper Review] Clinical and Non-clinical Effects on Surgery Duration: Statistical Modeling and Analysis

Jin Wang, Ph. D.+10 more|arXiv (Cornell University)|Jan 12, 2018

Healthcare Operations and Scheduling Optimization13 references3 citations

TL;DR

This study investigates how non-clinical factors—surgeon workload, operating room (OR) workload, and surgery position in a sequence—affect surgical procedure duration (the third phase of surgery). Using statistical modeling on two years of Chinese hospital data, it finds that surgery duration decreases by ~10 minutes per additional surgery per surgeon, but increases with OR workload beyond four surgeries per day. The study reveals significant interactions between surgery type, position, and surgeon, highlighting the need to incorporate these factors into OR scheduling for improved efficiency.

ABSTRACT

Surgery duration is usually used as an input to the operation room (OR) allocation and surgery scheduling problems. A good estimation of surgery duration benefits the operation planning in ORs. In contrast, we would like to investigate whether the allocation decisions in turn influence surgery duration. Using almost two years of data from a large hospital in China, we find evidence in support of our conjecture. Surgery duration decreases with the number of surgeries a surgeon performs in a day. Numerically, surgery duration will decrease by 10 minutes on average if a surgeon performs one more surgery. Furthermore, we find a non-linear relationship between surgery duration and the number of surgeries allocated to an OR. Also, a surgery's duration is affected by its position in a sequence of surgeries performed by one surgeon. In addition, surgeons exhibit different patterns on the effects of surgery type and position. Since the findings are obtained from a particular data set, We do not claim the generalizability. Instead, the analysis in this paper provides insights into surgery duration study in ORs.

Motivation & Objective

• To examine the influence of non-clinical factors—surgeon workload, OR workload, and surgery position—on surgical procedure duration.
• To assess whether scheduling and allocation decisions affect actual surgery duration, challenging the assumption that duration is a fixed input.
• To investigate interactions between surgery type, position in sequence, and surgeon performance on duration.
• To develop a statistical modeling framework that captures complex, non-linear relationships and interactions in surgery duration data.
• To provide actionable insights for improving operating room planning and scheduling by incorporating workload and sequencing effects.

Proposed method

• The study uses a large, real-world dataset of nearly two years of surgery records from a Chinese hospital, focusing on the surgical procedure phase (third phase of surgery).
• A mixed-effects linear regression model is applied to estimate the impact of surgeon workload, OR workload, and surgery position on duration, with random effects for surgeons and surgery types.
• Interaction terms between surgery type and position are included to detect differential effects across surgeons and procedures.
• A regression tree is used to identify non-linear patterns and potential interaction structures among predictors.
• The model’s fit is evaluated using log-likelihood and adjusted R², with significance assessed via p-values and standard errors.
• The analysis is restricted to the surgical procedure duration (from anesthesia end to procedure end), excluding preparation and wake-up times.

Experimental results

Research questions

• RQ1 How does the number of surgeries a surgeon performs in a day affect the duration of individual surgical procedures?
• RQ2 What is the relationship between the number of surgeries scheduled in an operating room and the duration of individual procedures?
• RQ3 Does the position of a surgery within a sequence of surgeries performed by a surgeon influence its duration?
• RQ4 How do interactions between surgery type, position in sequence, and surgeon identity affect surgical duration?
• RQ5 To what extent do non-clinical factors such as workload and sequencing impact surgery duration beyond clinical factors?

Key findings

• Surgery duration decreases by an average of 10 minutes for each additional surgery performed by a surgeon in a day.
• When OR workload exceeds four surgeries per day, surgery duration increases by approximately 5 minutes per additional surgery, indicating a non-linear relationship.
• Surgery duration is significantly influenced by its position in a surgeon’s daily sequence, with durations varying depending on order (e.g., earlier vs. later surgeries).
• Surgeons exhibit distinct performance patterns: some, like Surgeons B and C, show greater sensitivity to surgery position, indicating variability in workload adaptation.
• Significant interactions exist between surgery type and position—e.g., Surgery Type 11 shows a 1.47-minute reduction in duration when moved from position 3 to 5, while Surgery Type 10 shows a non-significant 0.42-minute decrease.
• The inclusion of interaction terms improves model fit, with adjusted R² increasing to 0.224 and log-likelihood improving, confirming the importance of modeling these complex relationships.