Automated survey-weighted analysis, publication-ready figures, and manuscript generation. Transform months of epidemiological research into a reproducible, one-click pipeline.
From research question to submission-ready manuscript in one reproducible workflow
Upload a Word document or enter your research question. PICO/PECO framework automatically extracted using NLP.
Intelligent mapping from your concepts to 150+ NHANES variables across demographics, labs, questionnaires, and mortality data.
Automated XPT file download from CDC NHANES FTP, with local parquet caching. Supports all 10 cycles from 1999-2018.
Clean missing codes, recode variables, merge DEMO+LAB+Q datasets. Automatic survey weight adjustment for multi-cycle analysis.
Survey-weighted descriptive stats, Rao-Scott chi-square, logistic/linear regression, subgroup analysis with proper SE estimation.
Lancet-standard Table 1 (baseline), regression tables, forest plots, correlation heatmaps, Kaplan-Meier curves. 300 DPI output.
Real-time PubMed search via NCBI E-utilities. Auto-retrieves related studies, formats Vancouver-style citations for your manuscript.
Generate a Lancet-format paper: structured abstract, methods with STROBE compliance, results with embedded tables, and discussion.
Purpose-built for epidemiological research, not a generic data tool
Proper NHANES complex survey design with MEC/interview weights, PSU, and strata. Not just “add weights to regression” — Taylor linearization SEs, Rao-Scott chi-square, and weight adjustment for multi-cycle pooling.
300 DPI figures in Lancet color palette. Table 1 with means/SD or n/% by group. Forest plots with HR/OR and 95% CI. Correlation heatmaps. STROBE checklist compliance verification.
Curated database of NHANES variables across 27 categories: demographics, body measures, blood pressure, labs, questionnaires, diet, physical activity, sleep, depression (PHQ-9), and mortality follow-up.
DeepSeek LLM integration for Lancet-format manuscript generation. Structured abstract, methods section with exact statistical parameters, results narrative, and discussion with strengths/limitations.
Pre-configured exposure/outcome/covariate sets for: obesity, diabetes, hypertension, dyslipidemia, CVD, smoking, depression, CKD, diet quality, and sleep disorders. One-click study setup.
Every analysis produces a complete ZIP package: cleaned dataset, analysis scripts (Python + R), all tables/figures, generated manuscript, and a manifest with version info and parameters.
NHANES analysis requires specialized knowledge. We encode that knowledge into software.
| Capability | NHANES to Lancet | Generic Stats Tools | Manual Analysis |
|---|---|---|---|
| Automatic CDC data download | ✓ | — | — |
| Survey-weighted analysis | ✓ | Partial | ✓ |
| NHANES variable knowledge base | ✓ | — | — |
| Lancet-format manuscript generation | ✓ | — | — |
| STROBE compliance check | ✓ | — | Manual |
| PubMed literature integration | ✓ | — | — |
| Reproducible pipeline | ✓ | Partial | — |
| Time to results | ~10 minutes | Hours | Weeks |
Real epidemiological research scenarios, not toy examples
First NHANES paper without spending months learning survey design, SAS/STATA code, and Lancet formatting conventions.
Rapid hypothesis testing across multiple NHANES cycles. Batch-process several research questions with consistent methodology.
Standardized, auditable analysis pipeline for population health surveillance. Reproducible reports for policy stakeholders.
Choose the plan that fits your research workflow
For individual researchers exploring NHANES
For research teams and active labs
For institutions and CROs
Join researchers using NHANES to Lancet for faster, reproducible epidemiological analysis.