Raw data preview
Detected STEPS columns

Rows (raw)

Rows (clean)

Rows dropped

Variables

Derived variable distributions

Heaping severity

Overall completeness

Implausible values

Weight CV




Design Summary

Design Type

Unweighted N

Weighted N

Strata

PSUs

Design Elements

Domains computed

Key indicators

Generation Status

Report Summary

Report Status

Output Files

stepssurvey

WHO STEPS NCD Survey Analysis Pipeline

Version 0.1.0

About the Package

The WHO STEPwise Approach to NCD Risk Factor Surveillance (STEPS) is the global standard for monitoring non-communicable disease risk factors. Adopted by over 130 countries, it provides a standardised framework for collecting behavioural, physical, and biochemical data across three steps: questionnaire-based assessment of tobacco, alcohol, diet, and physical activity (Step 1); physical measurements including blood pressure and anthropometry (Step 2); and biochemical measurements of blood glucose and lipids (Step 3).

Despite well-defined WHO protocols for data collection, the analysis of STEPS data remains a significant bottleneck. Country teams often rely on manual Epi Info or SPSS scripts that require substantial statistical expertise, are error-prone, and produce inconsistent outputs. This is particularly challenging for low- and middle-income countries (LMICs) where trained biostatisticians are scarce and surveillance cycles are infrequent.

stepssurvey addresses this gap by providing a complete, end-to-end analysis pipeline in R that automates the entire workflow from raw survey data to publication-ready reports, faithfully implementing WHO-standard methodology for all ten headline NCD risk factor indicators. It supports CSV, Excel, Stata, and SPSS formats; implements WHO-standard cleaning rules (smk_cln/smkless_cln, GPAQ P_clean); handles complex survey designs with step-specific weights; and produces formatted tables, WHO-branded visualisations, and three Word report templates.

The package has been validated against published WHO fact sheets from nine countries across six WHO regions, achieving 96% concordance (123 of 128 indicators within 1 percentage point of published values) with 100% overlapping 95% confidence intervals.

About AIIMS Bhopal

The All India Institute of Medical Sciences (AIIMS), Bhopal is an Institute of National Importance established in 2012, serving as a centre of excellence in medical education, research, and tertiary healthcare for central India. The Department of Community and Family Medicine has extensive experience in population-based NCD surveillance, having conducted STEPS-aligned surveys and large-scale epidemiological studies across Madhya Pradesh.

Authors

Dr Abhijit Pakhare — Professor, Department of Community and Family Medicine, AIIMS Bhopal

Dr Ankur Joshi — Additional Professor, Department of Community and Family Medicine, AIIMS Bhopal

Dr Lena Charlette — Assistant Professor, Department of Community and Family Medicine, AIIMS Madurai

AI-Assisted Development: Substantial portions of this package were developed with the assistance of Claude (Anthropic), an AI coding assistant. This includes the Shiny application, automated test suite, visualisation functions, report templates, and the multi-country validation pipeline. The domain expertise, epidemiological methodology, and analytical decisions are entirely the authors'; Claude served as a development accelerator that enabled three researchers to build and validate a tool that would otherwise require a dedicated software engineering team. All AI-generated code was reviewed, tested, and validated by the authors against published WHO outputs.

Citation

If you use stepssurvey in your research or surveillance work, please cite:

Pakhare A, Joshi A, Charlette L (2026). stepssurvey: Analyse WHO STEPS Survey Data. R package version 0.1.0. https://github.com/drpakhare/stepssurvey

License: MIT | Source: GitHub | Docs: Package website | Issues: Bug reports