Data from: Characterizing surgeon workload with electronic health record data to predict time interval between surgeries and postoperative care delivery
Data files
Jun 22, 2026 version files 26.76 KB
-
dryad_data.xlsx
17.18 KB
-
README.md
9.58 KB
Abstract
Objective: This study positions surgeon gap time, defined as the interval between consecutive surgeries performed by the same surgeon, as a surgeon-level metric of efficiency. Understanding gap time requires accounting for a surgeon’s operative workload, yet no objective electronic health record (EHR)-derived measure exists. We conceptualize surgical case demand as an EHR-derived surrogate for operative workload and examine its association with surgeon gap time.
Results: Clustering identified three demand types with distinct post-operative profiles. High-level recovery was required for 5.9%, 6.5%, and 17.2% of Demand Types 1, 2, and 3 cases respectively (p < 0.001), with median LOS of 0.09, 0.97, and 1.80 days (p < 0.001). Demand type, case priority, surgical specialty, and surgical care location were significant predictors of gap time.
Discussion: Surgical case demand serves as an EHR-derived surrogate for operative workload, enabling structured analysis of surgeon gap time and the factors associated with it.
Conclusion: Surgeon gap time is an objective metric that can be operationalized from EHR data, providing insights for scheduling, resource allocation, and overall health system efficiency.
Overview
This repository contains summarized data underlying the findings of the manuscript above. The raw electronic health record (EHR) dataset is not publicly available due to data privacy restrictions. The source data consists of 86,480 surgical cases performed at a U.S. teaching hospital over three years, spanning 14 surgical specialties.
Contents
A single Excel workbook, dryad_data.xlsx, with three sheets at increasing levels of granularity:
| Sheet | Rows | Description |
|---|---|---|
| Overall | one row per variable/level | Dataset-wide descriptive statistics (corresponds to Table 1 in the manuscript): counts, percentages, means, and standard deviations for patient, surgery, and outcome variables. |
| By Demand Type | 3 (one per demand type) | Full feature and outcome profile of each surgical case demand type. Each demand type's distribution across patient, surgery, and clinical-outcome variables. |
| By Specialty | 14 (one per specialty) | The same feature and outcome profile, broken out by surgical specialty, including each specialty's distribution across the three demand types. |
Column dictionary
Sheet 1 — Overall (long format)
| Column | Description |
|---|---|
category |
Grouping of the variable (e.g., Patient demographics, Surgery characteristics) |
variable |
Variable name |
level |
Category level for categorical variables; "n/a" for continuous variables |
n |
Number of cases (categorical variables) |
percent |
Percent of the analytic dataset (categorical variables) |
mean |
Mean (continuous variables) |
sd |
Standard deviation (continuous variables) |
Sheets 2 & 3 — wide format
The first column identifies the group (demand_type on Sheet 2; surgical_specialty on Sheet 3). Remaining columns:
| Column | Description |
|---|---|
n_cases |
Number of surgical cases in the group |
pct_of_total |
Percent of the full analytic dataset (denominator = 86,480) |
mean_age, sd_age |
Patient age (years) |
pct_female, pct_male |
Patient sex distribution (%) |
pct_asa_1 … pct_asa_5E |
ASA class distribution (%) across the 10 levels |
pct_case_basic, pct_case_complex, pct_case_major_complex |
Case level distribution (%) |
pct_elective, pct_emergency |
Schedule priority distribution (%) |
pct_on_hours, pct_off_hours |
Work hours distribution (%) |
pct_demand_type_1, pct_demand_type_2, pct_demand_type_3 |
Demand type distribution (%) — Sheet 3 (By Specialty) only |
mean_surgery_duration, sd_surgery_duration, median_surgery_duration |
Surgery duration (hours) |
mean_total_or_time, sd_total_or_time |
Total OR time (hours) |
pct_high_recovery |
Percent of cases routed to high-level (ICU) postoperative recovery |
pct_low_recovery |
Percent of cases routed to low-level postoperative recovery |
mean_postop_los, sd_postop_los, median_postop_los |
Postoperative length of stay (days) |
On Sheets 2 and 3, percentage columns are computed within each group (each group's ASA columns, for example, sum to ~100%).
Variable definitions
| Variable | Type | Description |
|---|---|---|
| ASA class | Ordinal | American Society of Anesthesiologists physical status classification (1, 1E, 2, 2E, 3, 3E, 4, 4E, 5, 5E); higher numbers denote greater preoperative illness severity; "E" denotes emergency status |
| Case level | Ordinal | Institutional classification of surgical complexity and expected resource use (Basic, Complex, Major complex) |
| Schedule priority | Nominal | Elective vs. Emergency |
| Surgical specialty | Nominal | Specialty of the procedure |
| Work hours | Nominal | On-hours (07:30 – 16:00) vs. Off-hours |
| Surgery duration | Continuous (hours) | Time from surgical incision to procedure completion |
| Total OR time | Continuous (hours) | Total period the patient occupied the operating room (pre-incision setup through post-procedure cleanup) |
| Postoperative recovery location | Nominal | High level recovery (ICU) vs. low level recovery (PACU, general ward, home) |
| Postoperative LOS | Continuous (days) | Days from surgery end date to hospital discharge date |
| Demand type | Nominal | Case demand cluster (Demand Type 1, 2, or 3) from k-prototypes clustering on ASA class, case level, surgery duration, and work hours |
Non-applicable values
Cells marked "n/a" in this workbook denote not applicable i.e. the column does not apply to the variable in that row. This appears only in the Overall sheet, which combines categorical variables (where n and percent are reported) and continuous variables (where mean and sd are reported) in the same long-format table:
- For categorical rows (e.g., Sex, ASA class, Case level): the
meanandsdcolumns are marked "n/a". - For continuous rows (e.g., Age, Surgery duration, LOS): the
level,n, andpercentcolumns are marked "n/a". - For the Surgeon ID count row: all columns other than
nare marked "n/a".
Contact
For questions about this dataset, please contact:
- Muge Capan, PhD (corresponding author) — mcapan@umass.edu
Human subjects data
The data underlying this submission were derived from a retrospective analysis of electronic health record (EHR) data from a single U.S. teaching hospital. The study was reviewed by the Institutional Review Board (IRB) of the University of Massachusetts, Amherst, which granted a waiver of informed consent given the retrospective use of de-identified records.
We analyzed 86,480 surgeries in 14 specialties performed between 2020 and 2023 at a U.S. Medical Center. Surgical case demand was operationalized using patient and surgery features and clustered into demand types. Clinical implications were assessed by comparing postoperative care location and length of stay (LOS) across demand types. A classification tree was developed to assign demand types for future cases. Regression models were used to identify predictors of gap time.