This repository contains the public dataset and analysis code used to generate the results and Figure 1 reported in the manuscript examining predictors of home birth among women receiving antenatal care in SMRU clinics.

Predicted probability of home birth by number of antenatal care (ANC) visits, stratified by parity.

SMRU_homebirth/
├── data/
│   └── homebirth_analysis.csv
│
├── scripts/
│   └── analysis_main.R
│
├── docs/
│   └── data_dictionary.md
│
├── output/
│   ├── tables/
│   └── figures/
│
└── README.md

Primary analysis pipeline.

This script:

1. Loads the public dataset
2. Constructs the analysis dataset
3. Fits a mixed-effects logistic regression model with a random intercept for clinic (OR_site)
4. Calculates clustering statistics (variance, MOR, ICC)
5. Generates the main results table (descriptive statistics, UOR, AOR)
6. Produces predicted probability curves for home birth by number of ANC visits (Figure 1)
7. Saves Figure 1 to the output/figures/ directory

Running the script will automatically create output folders if they do not exist.

data/homebirth_analysis.csv

Public analysis dataset containing pseudonymized participant-level observations.

Key identifiers:

• ANC_code – participant identifier
• OR_site – clinic identifier

The dataset contains one row per participant / pregnancy observation.

A full description of dataset variables is available in the data dictionary:

• docs/data_dictionary.md

The analysis was conducted in R using the following packages:

• lme4
• boot
• ggplot2
• scales
• mvtnorm

Install packages if needed:

install.packages(c("lme4","boot","ggplot2","scales","mvtnorm"))

R script for data wrangling and analysis here:

• scripts/analysis_main.R

If using this repository or dataset, please cite the associated manuscript.