HCUP NIS Cohort Analysis
Budget: $250 – $750 USD
I need an analyst who knows the HCUP Nationwide Inpatient Sample inside out. Using Stata, please take the raw NIS files I provide, clean and merge them, and then build diagnosis-specific cohorts from my ICD-9/ICD-10 lists. Once the cohorts are ready, I want weighted national estimates, descriptive tables, and the multivariable models I have already outlined, all properly accounting for the survey design (strata, cluster, and discharge weights).
Because this work will feed straight into a manuscript, transparency and reproducibility matter as much as the numbers themselves. Comment your code thoroughly, keep each data-processing step modular, and document every assumption so another analyst could rerun the study end-to-end without guesswork.
Deliverables
• Cleaned, analysis-ready NIS dataset (subject to HCUP data-use restrictions)
• Fully annotated Stata do-files for data cleaning, cohort creation, descriptive stats, and multivariable modelling
• Publication-quality tables/figures (Word or Excel plus .png/.eps originals)
• Draft methods paragraph suitable for a peer-reviewed journal
• README explaining file structure, required Stata version/packages, and exact commands to reproduce results
Acceptance criteria
1. Survey weights applied correctly (results must match HCUP documentation examples).
2. Cohort counts and key demographics returned within ±2 % of my benchmark numbers.
3. All code runs start-to-finish on a fresh machine with only the stated Stata packages installed.
If this sounds routine to you, let’s get started—I’m ready to share the data access credentials and study protocol as soon as we agree on the timeline.
Because this work will feed straight into a manuscript, transparency and reproducibility matter as much as the numbers themselves. Comment your code thoroughly, keep each data-processing step modular, and document every assumption so another analyst could rerun the study end-to-end without guesswork.
Deliverables
• Cleaned, analysis-ready NIS dataset (subject to HCUP data-use restrictions)
• Fully annotated Stata do-files for data cleaning, cohort creation, descriptive stats, and multivariable modelling
• Publication-quality tables/figures (Word or Excel plus .png/.eps originals)
• Draft methods paragraph suitable for a peer-reviewed journal
• README explaining file structure, required Stata version/packages, and exact commands to reproduce results
Acceptance criteria
1. Survey weights applied correctly (results must match HCUP documentation examples).
2. Cohort counts and key demographics returned within ±2 % of my benchmark numbers.
3. All code runs start-to-finish on a fresh machine with only the stated Stata packages installed.
If this sounds routine to you, let’s get started—I’m ready to share the data access credentials and study protocol as soon as we agree on the timeline.