A public reproducibility package for the study "Delayed umbilical cord clamping and postpartum maternal outcomes: a calendar-time-adjusted observational cohort study." It contains the publication-ready de-identified analysis dataset, the full R statistical analysis script, archived analysis outputs, and supporting documentation needed to review and reproduce the analyses.
The study evaluates whether guideline-concordant delayed umbilical cord clamping — defined as clamping at 60 seconds or later — is associated with early postpartum maternal haematological outcomes (postpartum anaemia, haemoglobin drop, postpartum haemorrhage and transfusion) in a single-centre retrospective cohort. The analytic approach combines calendar-time adjustment via natural cubic splines with four complementary regression and propensity-score methods (covariate-adjusted modified Poisson regression, inverse-probability-of-treatment weighting, overlap weighting and 1:1 propensity-score matching) to address implementation-era confounding.
| Path | Contents | Purpose |
|---|---|---|
data/PPH_CordClamping.csv |
Publication-ready de-identified analysis dataset (4,644 delivery episodes, 68 variables). Includes day-resolution delivery dates. | Source dataset for reproducing all analyses. |
analysis/PPH_CordClamping_statistical_script.R |
Full R statistical analysis script (v16.4). | Reproduces the regression, propensity-score, balance, sensitivity and output-generation workflow. |
analysis_outputs/analysis_outputs_cord_clamping_v16_4_complete.zip |
Archived outputs from the v16.4 statistical run. | Tables, figures, model summaries and quality-control outputs generated by the analysis pipeline. |
documents/03_Main_Manuscript.docx |
Main manuscript. | Article text. |
documents/06_Supplementary_Materials.docx |
Supplementary material. | Supplementary methods, tables and figures. |
documents/09_STROBE_Checklist.docx |
Completed STROBE checklist. | Reporting-guideline documentation for the observational cohort study. |
DATA_DICTIONARY.md |
Variable-level data dictionary. | Documents variable definitions, types, missingness and representative values. |
REPRODUCIBILITY_NOTE.md |
Reproducibility note. | Documents the v16.4 analysis script changes, PCC_04411 date correction, and instructions for reproducing the analysis. |
metadata/ |
Dataset profile, archive inventory and repository manifest. | Transparent file-level and dataset-level metadata. |
LICENSE |
MIT License. | Applies to all code in analysis/ and any scripts in this repository. |
LICENSE-data |
Creative Commons Attribution 4.0 International (CC-BY-4.0). | Applies to the dataset in data/ and the data dictionary. |
Editorial correspondence (cover letters, reviewer rebuttals, journal-specific submission materials) is intentionally not included in this public repository.
The dataset is publication-ready and de-identified. It contains a repository-specific study identifier and analysis variables; it does not include names, telephone numbers, e-mail addresses, street addresses, medical-record numbers, raw free-text clinical notes, or personal identifiers. Calendar time is represented via the delivery_date_iso field (M/D/YYYY format) and the derived delivery_year variable. One record (PCC_04411) carries a known source-level date typo (year 2002 instead of 2022), which is documented and corrected in the analysis script. Raw free-text fields for comorbidities, pregnancy complications and caesarean indications were either removed or represented as standardised categorical variables with quality-control flags.
A suggested data-availability statement for any manuscript drawing on this repository:
The de-identified analysis dataset, statistical code and reproducibility materials that support the findings of this study are openly available on Zenodo at
https://doi.org/10.5281/zenodo.20315727. The archived package contains a de-identified analysis dataset with day-resolution delivery dates, the statistical analysis code (v16.4), a variable-level data dictionary, archived analysis outputs, and the linked public GitHub repository for version-controlled code review.
This is an observational cohort study reported in accordance with the STROBE guideline; the completed checklist is included in documents/09_STROBE_Checklist.docx.
The analysis script targets R (≥4.2 recommended) and uses the following packages: tidyverse, readr, readxl, janitor, lubridate, broom, tableone, WeightIt, cobalt, survey, sandwich, lmtest, MatchIt, logistf, EValue, splines, stringr and forcats. The script includes package-installation checks, but pre-installing the dependencies in a controlled research-computing environment is recommended.
To re-run the analysis from a fresh clone, from the repository root:
Rscript analysis/PPH_CordClamping_statistical_script.R
The script resolves the dataset at data/PPH_CordClamping.csv when launched from the repository root, and at ../data/PPH_CordClamping.csv when launched from inside analysis/. Outputs are written to the path configured at the top of the script.
The dataset includes day-resolution delivery dates, enabling exact calendar-time adjustment. The analysis script (v16.4) handles one known date typo (PCC_04411: 2002 corrected to 2022) and excludes 12 records with missing delivery dates from calendar-adjusted models. See REPRODUCIBILITY_NOTE.md for details on the v16.4 changes and reproducibility instructions.
If you use this repository, please cite the manuscript together with the archived reproducibility package:
Aynaoğlu Yıldız G, Goc G, Süt H, Turgay B, Yıldırım Baydemir B, Öztürk E, Atabekoğlu CS. Delayed umbilical cord clamping and postpartum maternal outcomes: a calendar-time-adjusted observational cohort study. Reproducibility repository, 2026.
https://doi.org/10.5281/zenodo.20315727.
A dataset-style citation:
[dataset] Aynaoğlu Yıldız G, Goc G, Süt H, Turgay B, Yıldırım Baydemir B, Öztürk E, Atabekoğlu CS. 2026. Delayed umbilical cord clamping and postpartum maternal outcomes: de-identified analysis dataset and reproducibility package. Zenodo.
https://doi.org/10.5281/zenodo.20315727.
A machine-readable citation block is provided in CITATION.cff.
Code in this repository is released under the MIT License. The dataset and data dictionary are released under Creative Commons Attribution 4.0 International (CC-BY-4.0).
Reuse is permitted under those terms with appropriate attribution. Contributors must not commit unredacted clinical exports, raw free-text fields, personal identifiers, signed disclosure forms, or private editorial correspondence to this repository.
For data, code or documentation issues, open an issue on the repository tracker. For questions related to the underlying study or institutional approvals, contact the corresponding author listed in CITATION.cff.