This documentation covers a fully reproducible statistical analysis pipeline for investigating the global prevalence of inappropriate acid suppressing agent (ASA) use — including proton pump inhibitors (PPIs) and H2-receptor antagonists (H2 blockers). Built with theDocumentation Index
Fetch the complete documentation index at: https://mintlify.com/namakala/inappropriate-acid-suppressor-agent-use/llms.txt
Use this file to discover all available pages before exploring further.
targets framework in R, the pipeline takes a curated study-level dataset and produces pooled prevalence estimates, subgroup analyses, meta-regression models, and publication bias assessments, all rendered into an automated Quarto report.
Research background
Inappropriate prescribing of acid suppressing agents is a recognized global clinical problem. PPIs and H2 blockers are among the most widely dispensed drug classes worldwide, yet a substantial proportion of prescriptions lack a valid clinical indication. Overuse is associated with avoidable adverse effects, unnecessary healthcare costs, and antimicrobial resistance through gut microbiome disruption. This pipeline systematically quantifies the prevalence of inappropriate ASA use by pooling evidence from primary studies through random-effects meta-analysis. It applies the Freeman-Tukey double arcsine transformation to stabilize the variance of proportion data, and uses the Copas selection model to assess sensitivity to publication bias.Research questions
The pipeline is designed to answer the following questions:- What is the pooled global prevalence of inappropriate ASA prescribing?
- Does prevalence differ by geographic region (continent)?
- Does prevalence differ by clinical setting (inpatient, outpatient, ICU)?
- How does the use of explicit prescribing guidelines affect reported prevalence?
- What study-level characteristics (year, sample size, quality) explain between-study heterogeneity?
Key variables
Each included study contributes the following variables to the analysis dataset:| Variable | Description |
|---|---|
Author | Study author(s) |
Sample_size | Number of patients or prescriptions assessed |
Inappropriate_indication | Count of inappropriate prescriptions identified |
Prevalence | Proportion of prescriptions deemed inappropriate |
Year | Publication year |
Continent | Geographic region of the study |
Setting | Clinical setting (e.g., inpatient, outpatient) |
JBI_Classification | Study quality rating per JBI critical appraisal |
use_guideline | Whether the study applied a named prescribing guideline |
Explore the pipeline
Quickstart
Clone the repository, restore the R environment, and run the full pipeline in a few commands.
Pipeline overview
Understand how
targets orchestrates every step from raw data ingestion to the final report.Analysis methods
Learn the statistical methods: random-effects models, subgroup analyses, and meta-regression.
Results summary
Review the pooled prevalence estimates and key findings from the completed analysis.
Data availability
Raw data are not included in this repository. You must export the study-level dataset as
data.csv and place it at data/raw/data.csv before running the pipeline. See the Quickstart for the required directory structure.