Decision tree model to assess consequences and costs associated with therapy administration pathways for patients with HER2+ breast cancer in Italian oncological centers
Data files
Jul 06, 2026 version files 2.73 MB
-
Background.csv
1.15 KB
-
Cover.csv
414 B
-
Foglio1.csv
294 B
-
Input_Parameters-_cost.csv
891 B
-
Input_Parameters.csv
7.26 KB
-
Model_submission.xlsm
2.02 MB
-
OWSA.csv
97.13 KB
-
Parameters_Sensitivity.csv
24.85 KB
-
PSA_Simulation_-_1Cycle.csv
280 KB
-
PSA_Simulation_-_NCycle.csv
280.35 KB
-
README_Dryad_cleaning_notes.txt
997 B
-
README.md
8.68 KB
-
Results_Deterministisc_A.csv
2.96 KB
-
Results_OWSA.csv
186 B
-
Results_Probabilistic_A.csv
2.70 KB
-
Simulation_1Cycle.csv
3.06 KB
-
Simulation_NCycle.csv
3.04 KB
-
worksheet_export_summary.csv
725 B
Abstract
Human Epidermal Growth Factor Receptor 2-positive (HER2+) breast cancer poses significant therapeutic challenges, particularly concerning treatment administration pathways and their associated costs. This study evaluates the managerial and economic impacts of different therapeutic administration scenarios for HER2-positive breast cancer patients, focusing on optimizing hospital workflows, resource utilization, and patient outcomes in Italian oncological centers. A decision tree model was developed to simulate and compare five treatment administration pathways: Standard, Drug-Change, Drug Day, Dedicated Ambulatory, and Optimal Pathway scenarios. The model integrates patient and healthcare professional (HCP) activity and waiting times, infusion chair occupation, and direct and indirect costs. Sensitivity analyses assessed variability in model outcomes. Switching from endovenous (EV) to subcutaneous (SC) administration substantially reduced patient throughput times and HCP workloads. The Optimal Pathway scenario yielded the highest resource optimisation, reducing HCP activity time by up to 48 hours, infusional chair occupational time by up to 150 hours, and patients’ total time by up to 753 hours per 100 patients monthly. Cost analyses indicated significant savings in both direct and indirect cost for all the proposed scenarios in comparison to the Standard one. The adoption of SC formulations and innovative pathway optimizations enhances treatment organizational efficiency and reduces both direct and indirect costs. These findings underscore the value of tailored approaches to administration based on the structural and organizational characteristics of individual oncology centers, aligning with current Italian healthcare reforms.
Dataset DOI: 10.5061/dryad.mpg4f4rgm
Description of the data and file structure
This dataset contains a Microsoft Excel macro-enabled decision tree model developed to assess the organizational consequences, time outcomes, and costs associated with different therapy administration pathways for patients with HER2-positive breast cancer treated with pertuzumab and trastuzumab.
The model compares five alternative administration schemes:
-
Standard scheme
-
Drug-change scheme
-
Drug-day scheme
-
Dedicated ambulatory scheme
-
Optimal pathway scheme
The model includes clinical, organizational, time, and cost inputs, and estimates patient time, healthcare professional active time, infusion chair occupation time, caregiver time, catheter use, adverse events, direct healthcare costs, indirect costs, and total costs.
The workbook includes deterministic analyses, probabilistic sensitivity analyses, and one-way sensitivity analyses. Input parameters can be modified in the dedicated input sheets, while the results sheets automatically report the consequences and costs associated with each scenario.
The dataset does not contain individual patient-level data or identifiable personal information. It consists of aggregated model parameters, assumptions, calculations, and model outputs.
Files and variables
File: Model_submission.xlsm
Description:
The dataset includes one Excel macro-enabled workbook:
Model_submission.xlsm
The workbook contains the following sheets:
-
Cover: title page and authorship information.
-
Background: description of the workbook structure and purpose of each sheet.
-
Decision tree: graphical representation of the simulation model.
-
Input Parameters: editable clinical, organizational, and time parameters used in the model, including number of patients, number of treatment accesses, proportion of patients treated with intravenous therapy, proportion requiring blood sample exams, patient weight, catheter use, adverse event risks, working status, caregiver presence, travel time, and phase-specific activity/waiting times.
-
Input Parameters - cost: editable cost parameters, including healthcare professional hourly costs, infusion chair costs, drug acquisition costs, adverse event costs, and productivity loss inputs.
-
Results_Deterministic A: deterministic results for model outcomes and costs.
-
Results_Probabilistic A: probabilistic analysis results, reporting uncertainty around model outcomes and costs.
-
Results_OWSA: results of the one-way sensitivity analysis.
-
Parameters_Sensitivity: parameters used for sensitivity analyses.
-
Simulation_1Cycle: deterministic simulation for one therapeutic cycle.
-
Simulation_NCycle: deterministic simulation for multiple therapeutic cycles.
-
OWSA: one-way sensitivity analysis calculations.
-
PSA Simulation - 1Cycle: probabilistic sensitivity analysis simulation for one therapeutic cycle.
-
PSA Simulation - NCycle: probabilistic sensitivity analysis simulation for multiple therapeutic cycles.
-
Foglio1: support sheet containing lookup values and model settings.
Main input variables include:
-
Number of patients per day
-
Number of treatment accesses
-
Percentage of patients receiving intravenous therapy
-
Percentage of patients requiring blood sample exams
-
Mean patient body weight
-
Catheter use by administration type
-
Risk of catheter-related bloodstream infection or pocket infection
-
Risk of thrombotic events
-
Percentage of working patients
-
Percentage of patients with a caregiver
-
Percentage of working caregivers
-
Travel time
-
Patient active time
-
Patient waiting time
-
Healthcare professional active time
-
Infusion chair occupation time
-
Healthcare professional hourly costs
-
Infusion chair cost per minute
-
Drug acquisition costs
-
Adverse event costs
-
Working cost per hour
Main output variables include:
-
Patient active time
-
Patient waiting time
-
Patient inpatient time
-
Patient total time
-
Caregiver time
-
Healthcare professional active time
-
Infusion chair occupation time
-
Number of patients with catheter
-
Number of adverse events
-
Healthcare professional costs
-
Infusion chair costs
-
Treatment costs
-
Adverse event costs
-
Direct healthcare costs
-
Indirect productivity costs
-
Total costs
Code/software
The model was developed in Microsoft Excel as a macro-enabled workbook (.xlsm). All calculations are embedded within the workbook through Excel formulas and macros.
To reproduce the analyses:
-
Open the file Model_submission.xlsm in Microsoft Excel.
-
Enable macros when prompted.
-
Review or modify the editable input cells in the sheets Input Parameters and Input Parameters - cost.
-
The deterministic results are reported in Results_Deterministic A.
-
The probabilistic sensitivity analysis results are reported in Results_Probabilistic A and calculated through the PSA simulation sheets.
-
The one-way sensitivity analysis results are reported in Results_OWSA and calculated through the OWSA-related sheets.
The workbook was designed for use in Microsoft Excel. Compatibility with other spreadsheet software, such as LibreOffice or Google Sheets, is not guaranteed because the file contains macros and Excel-specific formulas.
Access information
Other publicly accessible locations of the data:
- None. The dataset is made publicly available through this Dryad repository.
Data was derived from the following sources:
- The model structure, assumptions, and input parameters were derived from published literature, publicly available sources, and expert elicitation, as described in the associated manuscript.
- No individual patient-level data or identifiable personal health information were used.
- The dataset does not include third-party datasets requiring separate access permissions.
- Where applicable, source references for model inputs are reported in the associated manuscript.
- The submitted Excel workbook contains aggregated parameters, assumptions, calculations, and model outputs generated for the purposes of the analysis.
Files included in this dataset
Model_submission.xlsm
Formatted Excel version of the model. The file has been retained because the original spreadsheet functions as a structured model/tool and the formatting supports interpretation and navigation of the model. Figures/images have been removed from the Excel file.
Background.csv
Unformatted CSV export of the “Background” worksheet from the Excel model.
Cover.csv
Unformatted CSV export of the “Cover” worksheet from the Excel model.
Foglio1.csv
Unformatted CSV export of the “Foglio1” worksheet from the Excel model.
Input_Parameters.csv
Unformatted CSV export of the “Input Parameters” worksheet from the Excel model, containing model input parameters.
Input_Parameters-_cost.csv
Unformatted CSV export of the cost-related input parameters worksheet from the Excel model.
OWSA.csv
Unformatted CSV export of the one-way sensitivity analysis worksheet.
Parameters_Sensitivity.csv
Unformatted CSV export of the sensitivity analysis parameters worksheet.
Results_Deterministisc_A.csv
Unformatted CSV export of the deterministic analysis results worksheet. The filename is kept as uploaded.
Results_OWSA.csv
Unformatted CSV export of the one-way sensitivity analysis results worksheet.
Results_Probabilistic_A.csv
Unformatted CSV export of the probabilistic analysis results worksheet.
Simulation_1Cycle.csv
Unformatted CSV export of the one-cycle simulation worksheet.
Simulation_NCycle.csv
Unformatted CSV export of the N-cycle simulation worksheet.
PSA_Simulation_-_1Cycle.csv
Unformatted CSV export of the one-cycle probabilistic sensitivity analysis simulation worksheet.
PSA_Simulation_-_NCycle.csv
Unformatted CSV export of the N-cycle probabilistic sensitivity analysis simulation worksheet.
worksheet_export_summary.csv
Technical summary file documenting the worksheet-to-CSV export process and the cleaning steps applied to the CSV files.
README_Dryad_cleaning_notes.txt
Technical note describing the cleaning performed on the original Excel file and CSV exports, including removal of images/figures, unnecessary formatting, blank rows, and header issues.