Free Tools

Guide · Multimodal registries

Following the Data

For groups planning a registry, or running one that holds more than it uses. Follows EEG, biological samples and every omics layer, EHR extracts, patient reported outcomes, imaging, wearables, and voice recordings from consent through analysis, asking the same questions of each: what the participant agreed to, how it is captured, which standard makes it comparable (OMOP, FHIR, HPO, Phenopackets, BIDS, DICOM), where it is stored, how it links to the rest, which methods fit, and how it is shared. Glossaries of standards and analytic methods, a planning checklist you can download, and a learning library of recorded lectures from government and academic sources. Saves as a PDF to take with you.

Registry governance

The Governance Register

For advocacy groups that run their own registry; articles on governance bodies, consent, US and international regulation, data requests, agreement terms, industry negotiation, and preparing for the things that go wrong. Fill it in as you read, score a live request, then generate a draft governance packet to take to your attorney and Institutional Review Board.

Data sharing documents for a registry

Data Sharing in a Box

For registry teams and advocacy groups getting ready to share data with outside researchers. Start with a readiness check-up of the program, covering the content inventory, governance, infrastructure, team capacity, privacy, and data standards, which ends in a gap list and, for groups heading toward a research network, an OMOP readiness level. Then fill in one stage at a time and the tool writes the pages a researcher expects to find: how to request data and how long review takes, the data use terms with suggested acknowledgment and data availability wording, a page for investigators with federal funding, the Data Access Committee operating procedures and review rubric, the ethics oversight statement, a cohort overview, and a data dictionary that imports a REDCap data dictionary as exported and moves calculated fields into derived fields. Further stages record how data are stored and reached, write a policy for sharing with clinical trial sponsors, and keep a decision record for contributing to an outside repository. The same review time, contact details, and committee name fill every page, so the pages agree with each other. Downloads the selected documents as a Word file, a PDF, Markdown, or a single web page with a menu that you can publish. Draft language for review by your attorney and Institutional Review Board (IRB), with the sources it was adapted from named. One stage shows at a time, it runs entirely in your browser, and it saves to a file that reloads.

Templates, course, and worked examples

The Registry Binder

The practical companion to The Governance Register. The Data Sharing in a Box sample pages are here to read in full, and the tool of the same name, above, fills them in for your registry: a data use agreement, a data access committee charter, an SOP and review rubric, IRB documentation, and a readiness assessment. Registries 101 runs from framing the question through consent, privacy, and network exchange. Plus vendor comparison and meeting worksheets, a data asset inventory, an ecosystem map, and a look at what a participant sees.

Adoption, fidelity, and sustainment

The Implementation Planner

For the group about to launch something other people have to take up: a registry survey, a new consent flow, a result returned to families, a workflow a clinic has to absorb. Whether the program works and whether anyone uses it are separate questions with separate evidence, and this planner is about the second one. Start with a plain description of what you are putting into practice, then answer the RE-AIM questions in writing with a strength rating on each answer, map what helps and what gets in the way across the domains of the Consolidated Framework for Implementation Research, choose strategies from the ERIC compilation with an owner and a target date on each, decide which implementation outcomes you will measure and where those data come from, place yourself in the EPIS phases, and write the sustainment plan while the design can still change. A readiness check-up reads what you entered and says where the plan is thin, and everything assembles into a document you can paste into a protocol, a board packet, or a grant application. Every framework is explained at the point where you use it, with a glossary and the published references behind each one. One stage shows at a time, it runs entirely in your browser, and it saves to a file that reloads.

Data asset register

Data Asset Intake

Catalog every data set, subset, model output, and code release in a research program, with governance and lineage built in. Add multiple assets and export the whole inventory as a spreadsheet or a printable report.

Sharing data for a competition

The Health Data Challenge Planner

For the group whose data would be the material for a hackathon, datathon, or prediction competition. The data and ethics phase asks what participants agreed to and whether that permits this use, which rules apply in each jurisdiction the entrants sit in, how the data would be de-identified and how re-identification risk would be assessed, and which access model fits: open download, credentialed access, or model-to-data inside an enclave where nothing leaves. Alongside it sit a quality and leakage audit, a data dictionary builder, a security plan, a community and ethics review step, and a participant data use agreement drafted from your own terms. The closing phase covers data destruction and access closure once the event is over. Runs entirely in your browser.

Analysis request intake

Data Analysis Request Form

For anyone asking an analyst for work, and for the analyst who has to scope it. Start the project on paper before anyone opens the data: the question being asked, the decision the answer feeds, the data that will be used, the deadline, and who signs off on the result. A completed form is the brief the analyst works from and the record both sides return to when the question drifts.

Part of the Data Analysts Toolkit from Ten Habits.

ETL project worksheet

The OMOP ETL Worksheet

For a site taking a source database into the OMOP CDM. Follows the work from the WhiteRabbit profiling scan and the mapping workshop through semantic mapping, technical architecture, development, infrastructure, the OHDSI tool stack, testing, data quality approval, and documentation, with a checkbox, an owner, and a target date on every step. Each stage also asks for the decision behind the work: which vocabulary release you mapped against, which codes had no standard concept and what you did with them, who approved the mappings and when, and states what has to be true before the next stage starts. Your entries save to a file that reloads, and download as a Markdown project plan or a CSV of tasks for your tracker.

Prompts, mappings, and ETL reference

The OMOP ETL AI Toolkit

For a team writing an OMOP ETL with an AI assistant beside them. Pick the stage you are on, from source profiling through mapping proposals, vocabulary candidates, code, tests, data quality assessment, documentation, peer review, and debugging a failed load, and the page assembles the prompt, with the rules on protected health information and an output contract built into every one so the answer comes back in a shape you can review. The mapping record holds each field-level decision with its rationale, its status, and the reviewer who approved it, and turns itself into a review prompt or exports as Markdown, CSV, or JSON. Alongside it: a prompt library for the work that comes around again, a repository scaffold with the shell script that creates it, a source system profile that becomes a mapping prompt, reference notes on the CDM tables and the vocabularies, and a glossary with the OHDSI documentation every note is drawn from. Where The OMOP ETL Worksheet tracks the project, this tracks the decisions inside it. One stage shows at a time, it runs entirely in your browser, and the session saves to a file that reloads.

OMOP readiness for a registry

Data Sharing in a Box

For a registry deciding whether, and when, to map to the OMOP CDM. The readiness check-up works in layers: whether the data can be shared responsibly, whether another team could interpret them without asking you, and then structural fit, the clinical domains present, vocabulary use, and the practicality of an ETL project. Your answers place the registry on a readiness scale that runs from foundational gaps through terminology mappable to operational, with what to focus on next at each level, and the unchecked items download as a gap list. It is a check of the program and its documentation, and no participant data are entered. When the registry is ready to map, it points to The OMOP ETL Worksheet and The OMOP ETL AI Toolkit. Runs entirely in your browser.

End-to-end study planning

Network Study Paint By Numbers

Take a question all the way to a protocol draft: concept sets, the ATHENA lookups that fill them, cohort entry and exit logic, exposure windows, and the diagnostics that decide whether the study can run across sites. One tab walks through a published OHDSI study, decision by decision, so you can work backwards from something that made it through peer review; the other collects the same decisions for your own study. Every step is shown as ATLAS clicks and as SQL, and your concept sets and cohort definition export as JSON that ATLAS imports.

Research question builder

Analytic Use Case Generator

Turn a clinical idea into a well-formed OHDSI research question, Mad Libs style, then trace every part of it to standard concepts, count it in your own OMOP data with the read-only queries it writes, and gauge its feasibility for a network study. Build a set and download it as a report.

Planning a federated repurposing study

The Repurposing Study Planner

Built around the way an OHDSI federated network study is planned and run, for a question about a drug already on the market. Answer a short set of questions about how the drug is taken and how the outcome appears in a record, and it suggests a design with its caveats and names the analysis package behind it, from a new-user cohort study with an active comparator to a self-controlled design to characterization first. The execution stage follows the sequence a network study moves through, from phenotype evaluation and the study package to the diagnostics gate, site onboarding, and combining results across sites, with an owner and a target date on each. Runs entirely in your browser.

External comparator assessment

RWDExchange

Evaluate whether real world data can serve as an external comparator in a clinical trial. Score each variable across four regulatory frameworks, map it to OMOP concepts, inventory confounders, and export a print-ready feasibility report.

Trial data crosswalk

CDISC SDTM to OMOP Mapper

Variable-level mappings from the SDTM domains to their OMOP CDM v5.4 targets, with notes on where judgment is required and a direction toggle that shows which domains feed each OMOP table. Search across variables, review the vocabulary crosswalks from MedDRA, WHODrug, and CDISC controlled terminology to standard concepts, and read where trial data break the observational mold: partial dates, blinded exposure, and qualifiers with no home. The full spec downloads as a CSV for ETL planning.

FHIR bundle inspector

Bundle Reader

Paste a FHIR bundle and see where every resource would go in the OMOP CDM before you write any code, with the assumptions made visible: imputed dates, invented visit boundaries, inverted statuses, missing units, codes that map to nothing. Built-in exercises, including a real example from the FHIR specification. Reads R4, R5, and R6, runs entirely in your browser, and downloads to a single file for use with real records.

EHR-to-OMOP pipeline

Registry Forge, Enterprise Edition

The full open-source pipeline: pull EHR data via SMART-on-FHIR and bulk exports, parse and map it to OMOP, then explore cohorts, exposures, and devices through ready-made dashboards.

Personal records to cohort

Registry Forge, Patient Edition

Turn health records people download themselves, from MyChart, Blue Button, and other C-CDA exports, into a research-ready cohort data set, with a live demo, EDA reports, and a privacy-first workflow.

Live monitor and OMOP reference

MCH·OMOP Live

Maternal and child health meets the OMOP common data model. A live monitor of articles, preprints, funding, videos, and code, drillable to MCH topics and exportable to CSV, plus the teaching material around it: a starting path for people new to observational research, cheat sheets on vocabulary, cohort design, and treatment pathways, and dated worked examples on menopause and preterm labor that take one clinical idea from a vocabulary search through concept sets, cohorts, and a feasibility assessment. Also a plain-language explorer for the 258 fields of the model.

Guided primer and prompt workbench

Research Data Science: A Guided Tour

For scientists, clinicians, and advocates who are not data scientists. Short modules take you from framing a question through study design, statistics, literature and trial APIs, vocabularies, FHIR, OMOP, real-world data, omics, genetics, GIS, governance, FAIR data, and reproducible analysis, with a rare epilepsy question worked all the way through. A prompt builder assembles a structured brief from your project details, and a prompt library, a filterable tool directory, and a glossary sit alongside. Search every topic, mark modules read, and deep-link to any module. Runs entirely in your browser; nothing you type leaves it.

AI on a regulated data pipeline

The OMOP ETL AI Toolkit

Using a language model on health data has rules, and this tool writes them into every prompt it generates: no protected health information, no small cell counts, no invented concept_ids, and a reviewer named on every mapping before it is implemented. Choose the ETL stage you are on and the page assembles a prompt that states the project context, the OMOP conventions the answer has to respect, and the format the answer comes back in, with the uncertain parts tagged for human review. A short operating policy sets out which parts of the work an assistant can take and which stay with a person, and the mapping record keeps the provenance of each decision, including whether it was AI-assisted. Runs entirely in your browser, so nothing you type is transmitted anywhere.

AI prompts, in your voice

The Nonprofit AI Prompt Studio

AI drafts sound like a nonprofit that does not exist because the model does not know yours. Fill in your organization once, pick a task, from grant narratives to donor letters to board reports, and the Studio assembles a prompt that puts your facts in, keeps your language rules, and forbids invented numbers.

LLM output validation

Verify & Cite

Check what a language model gives you against its sources, step by step: decompose the output into claims, verify every citation and figure, and flag what doesn't hold. Each step cites the method behind it, and you export a defensible evaluation record.

Guided primer and prompt workbench

Research Data Science: A Guided Tour

For scientists, clinicians, and advocates who are not data scientists. Short modules take you from framing a question through study design, statistics, literature and trial APIs, vocabularies, FHIR, OMOP, real-world data, omics, genetics, GIS, governance, FAIR data, and reproducible analysis, with a rare epilepsy question worked all the way through. A prompt builder assembles a structured brief from your project details, and a prompt library, a filterable tool directory, and a glossary sit alongside. Search every topic, mark modules read, and deep-link to any module. Runs entirely in your browser; nothing you type leaves it.

Simulated multi-omics project

Inside the Omics Enclave

For anyone who wants to know what an omics analyst sees inside a data enclave. Walk a synthetic rare epilepsy project one stage at a time, from sign-in through the project folder, the clinical layer of phenopackets and OMOP tables, and then genomics, transcriptomics, proteomics, and metabolomics, ending with what leaves through the airlock and why some outputs are refused. Each stage shows a simulated window with the file browser, the Nextflow run, the notebook cells, and the results, and explains how that step joins to the phenotypes and the clinical record through one linkage table. Read the samplesheets, the inside of a VCF, an mzML, and a feature table, and follow findings that agree across layers and one that turns out to be a medication effect. A tools-by-use-case table and a glossary close it out. Every participant, file, and result is synthetic.

Skills map for clinical data analysts

From Clinical Data to Omics

The companion to Inside the Omics Enclave, for the person who will be doing the work rather than watching it. Rate where you stand, then follow a staged curriculum that starts with the shell and file formats and runs through version control, environments, the compute environment, the path from sample to counts table, pipeline languages, and the statistical traps of high dimensional data. A translator shows tasks you already do in Excel, SAS, SQL, or REDCap on the command line and in R and Python, and a readiness check-up examines the project and the practices around it. Runs entirely in your browser.

Omics reference for team leads

Bioinformatics Shop in a Box

The reference to keep beside Inside the Omics Enclave and From Clinical Data to Omics. Explains the files an omics project produces and what each one can answer, the pipelines behind them, joint genotyping and combining cohorts, the GA4GH standards and how a phenopacket links to a VCF, and where analysis runs when the data cannot leave a cloud region, a cluster, or an enclave. Generators write an onboarding plan, an IT setup request, an analysis protocol, a cohort harmonization plan, a notebook entry, and a transfer manifest. One section shows at a time, and it runs entirely in your browser.

Code translation

Enclave Ready

Paste Stata, SPSS, or SAS code and get Python, R, and SQL equivalents side by side, with flags for commands that conflict with enclave rules, notes on how disclosure review commonly treats each output, and the full crosswalk reference organized by task. Runs entirely in your browser; pasted code goes nowhere.

Simulated multi-omics project

Inside the Omics Enclave

For anyone who wants to know what an omics analyst sees inside a data enclave. Walk a synthetic rare epilepsy project one stage at a time, from sign-in through the project folder, the clinical layer of phenopackets and OMOP tables, and then genomics, transcriptomics, proteomics, and metabolomics, ending with what leaves through the airlock and why some outputs are refused. Each stage shows a simulated window with the file browser, the Nextflow run, the notebook cells, and the results, and explains how that step joins to the phenotypes and the clinical record through one linkage table. Read the samplesheets, the inside of a VCF, an mzML, and a feature table, and follow findings that agree across layers and one that turns out to be a medication effect. A tools-by-use-case table and a glossary close it out. Every participant, file, and result is synthetic.

Skills map for clinical data analysts

From Clinical Data to Omics

The companion to Inside the Omics Enclave, for the person who will be doing the work rather than watching it. Rate where you stand, then follow a staged curriculum that starts with the shell and file formats and runs through version control, environments, the compute environment, the path from sample to counts table, pipeline languages, and the statistical traps of high dimensional data. A translator shows tasks you already do in Excel, SAS, SQL, or REDCap on the command line and in R and Python, and a readiness check-up examines the project and the practices around it. Runs entirely in your browser.

Guide · Multimodal registries

Following the Data

For groups planning a registry, or running one that holds more than it uses. Follows EEG, biological samples and every omics layer, EHR extracts, patient reported outcomes, imaging, wearables, and voice recordings from consent through analysis, asking the same questions of each: what the participant agreed to, how it is captured, which standard makes it comparable (OMOP, FHIR, HPO, Phenopackets, BIDS, DICOM), where it is stored, how it links to the rest, which methods fit, and how it is shared. Glossaries of standards and analytic methods, a planning checklist you can download, and a learning library of recorded lectures from government and academic sources. Saves as a PDF to take with you.

Guided primer and prompt workbench

Research Data Science: A Guided Tour

For scientists, clinicians, and advocates who are not data scientists. Short modules take you from framing a question through study design, statistics, literature and trial APIs, vocabularies, FHIR, OMOP, real-world data, omics, genetics, GIS, governance, FAIR data, and reproducible analysis, with a rare epilepsy question worked all the way through. A prompt builder assembles a structured brief from your project details, and a prompt library, a filterable tool directory, and a glossary sit alongside. Search every topic, mark modules read, and deep-link to any module. Runs entirely in your browser; nothing you type leaves it.

Secondary data catalog

Findable

103 vetted secondary data sources and 270 links that go straight to the files. Health, clinical, genomic, environmental, and social data, each with what is in it, how you get access, and what the license permits. Search it, filter by subject and access, and download the whole catalog as a spreadsheet.

End-to-end study planning

Network Study Paint By Numbers

Take a question all the way to a protocol draft: concept sets, the ATHENA lookups that fill them, cohort entry and exit logic, exposure windows, and the diagnostics that decide whether the study can run across sites. One tab walks through a published OHDSI study, decision by decision, so you can work backwards from something that made it through peer review; the other collects the same decisions for your own study. Every step is shown as ATLAS clicks and as SQL, and your concept sets and cohort definition export as JSON that ATLAS imports.

Planning a manuscript, idea to publication

Manuscript Paint by Numbers

For anyone taking a study from a question to a published paper. Each phase of the work is a numbered region on a canvas that fills in as you finish its checklist, from the question and the literature through authorship, ethics and disclosures, the reporting guideline and journal choice, the outline, each section of the paper, internal review, submission, and peer review and what follows acceptance. Every phase has structured fields, a list of common mistakes, a definition of done, and a worked example from a registry cohort study. The study design you choose selects the reporting guideline, with tickable item lists for STROBE, RECORD, CONSORT, PRISMA 2020, TRIPOD, and STARD. Alongside it sit a timeline planner, copyable availability and disclosure statements, an overclaiming check on your own discussion text, and cover letter and response letter builders. For an OHDSI network study, the phases add data partners, results by database, study diagnostics, and calibration, and a plan from Network Study Paint By Numbers can be linked so its question, concept sets, cohort logic, and time-at-risk are offered in the sections where they are reported. An institutional language library holds the disclaimers and acknowledgments your institution requires and offers them wherever they belong. Runs entirely in your browser, and saves to a file that reloads.

Research question builder

Analytic Use Case Generator

Turn a clinical idea into a well-formed OHDSI research question, Mad Libs style, then trace every part of it to standard concepts, count it in your own OMOP data with the read-only queries it writes, and gauge its feasibility for a network study. Build a set and download it as a report.

External comparator assessment

RWDExchange

Evaluate whether real world data can serve as an external comparator in a clinical trial. Score each variable across four regulatory frameworks, map it to OMOP concepts, inventory confounders, and export a print-ready feasibility report.

Economic evaluation

The HEOR Workbench

For anyone scoping an economic evaluation, or reading one and wanting to argue with it. The study planner collects the decisions a reviewer looks for, following the CHEERS 2022 reporting domains: the decision question, the comparator, the perspective, the time horizon, the discount rate, the outcomes and cost components to capture, the data sources behind them, the uncertainty plan, and where patients and caregivers shape the design. Alongside it, the calculators show what those choices do to the answer: enter costs and effects for two strategies and watch the incremental cost-effectiveness ratio move across the cost-effectiveness plane as you drag the threshold, build quality-adjusted life years from health states with and without discounting, and walk a budget impact funnel from covered lives to net cost for the year. Registry and claims people will recognize the inputs as their own data. Threshold and agency figures are dated and linked to the primary sources. Runs entirely in your browser and saves to a file that reloads.

Digital health technology

The Digital Health Technology Field Guide

A guided reference on wearables, sensors, and digital measures in health research, care, and policy, with the most depth on accelerometer and digital voice data. Choose a role (researcher or data scientist, policymaker or regulator, patient advocate, clinician, or new to the field) and the guide marks a reading path through the history of the field, the signals and their file formats and standards, analysis pipelines and the V3 validation framework, evidence by disease area with live searches of PubMed, Europe PMC, and ClinicalTrials.gov, open data sets, regulation and payment, organizations and people, funding, and the use of these tools in research and in patient-centered care. A planning worksheet exports to Markdown, and a shortlist with notes exports to CSV or a save file that reloads. Dated claims are labeled as a snapshot and linked to their sources. Runs entirely in your browser.

Planning the return of results

Sharing Summary Results with Research Participants

For investigators writing a plan to share results with participants into a protocol or a budget. Choose the options that describe the study (design, size of the analyzed group, how results are broken down, whether findings are exploratory or prognostic, what was captured at consent, how connected the community is, and who delivers the summary) and the page writes the considerations: minimum cell sizes and disclosure review, how to label comparisons that were not planned in advance, preference capture and re-confirmation, who interprets genetic or prognostic content and who pays for that time, and how the plan fits ClinicalTrials.gov reporting, the European Union layperson summary, and the Common Rule consent element. A table compares letters, public posting, webinars, clinician conversations, patient organization channels, video, and a layered approach, with rare disease and pediatric notes on each. Written around the NIH draft policy (NOT-OD-26-113), which is a draft and not a current requirement. Runs entirely in your browser.

Including children in a study

Pediatric Inclusion Commons

For a study team deciding who is eligible and from what age. The regulatory roadmap gathers the inclusion policies, the age category definitions that differ between drug guidance, device law, and funder policy, and the guidance on enrolling adolescents in adult trials, followed by pediatric study plans, extrapolation, external controls, and other approaches used when a large pediatric trial is hard to run. The ethics stage sets out the risk categories for research with children, assent and parental permission, the neonatal provisions, and single Institutional Review Board (IRB) review, and the data stage lists pediatric research networks, neonatal registries, claims, trial registries, and rare disease platforms with how access works and which data model each uses. Standards and methods cover age group definitions, outcome measures for children who cannot self-report, common data elements, and reporting guideline extensions for children. Entries are dated and meant to be confirmed at the source. Runs entirely in your browser.

Non-coding genome data and methods

Reading the Dark Genome

For a team planning omics work that reaches beyond protein-coding genes. The data stages list the public resources by family, from reference sequence and annotation through functional genomics, catalogs of non-coding RNA, repeats, and translated open reading frames, population and clinical variation, variant scoring tools, target knowledge bases, and cohorts with genome sequence, each with its access route and a shortlist you can download. Data forms are set out with the place where non-coding information is commonly lost in each, alongside the identifiers and standards for exchanging findings and a note on genome builds. Approaches to analysis are grouped by the question being asked, with tools in common use, a table of recurring pitfalls, and the statistical points an analyst from clinical data work will want first. A literature review by theme names what remains contested, and the interpretation stage sets out the kinds of evidence used for a non-coding variant. Runs entirely in your browser.

LLM output validation

Verify & Cite

Check what a language model gives you against its sources, step by step: decompose the output into claims, verify every citation and figure, and flag what doesn't hold. Each step cites the method behind it, and you export a defensible evaluation record.

Research question builder

Analytic Use Case Generator

The step the planner sends you to once the question is settled: put it in the standard OHDSI form, trace the drug, the comparator, and the outcome to standard concepts, and run the read-only queries it writes to count how many people at your site have taken the drug for the proposed use. The site feasibility report it produces is what a repurposing plan needs before anyone decides whether the study is possible.

End-to-end study planning

Network Study Paint By Numbers

The mechanics behind the execution stage of a repurposing plan: concept sets, the ATHENA lookups that fill them, cohort entry and exit logic, exposure windows, and the diagnostics that decide whether the study can run across sites, with a published OHDSI study worked backwards alongside your own. Every step is shown as ATLAS clicks and as SQL, and your concept sets and cohort definition export as JSON that ATLAS imports.

The value case for a repurposed drug

The HEOR Workbench

A repurposing result has to survive a coverage decision to reach patients, and a cheap generic can still fail one, because coverage turns on a model rather than on a price. The workbench sets out what that model rests on, which outcomes were counted, whose preferences set the utility weights, which costs the chosen perspective leaves out, and how a threshold turns a ratio into a yes or a no, with a panel on rare disease covering small populations, external comparators drawn from registries, and managed access. The calculators let you test a different perspective or threshold on your own numbers. Runs entirely in your browser.

Grant writing

The Grant Writer's Toolkit

The fillable tools from the Ten Habits companion site, following a proposal from first read of the funding announcement through submission: fit checks, goal worksheets, letters of intent and support, a capacity one-pager, and a project timeline.

Genetic testing beyond the exome

Reading the Dark Genome

For patients, families, and advocates who want to know whether past genetic testing looked in the right places. Explains in plain language what the non-coding genome is, what a gene panel, an exome, a genome, and RNA testing can each see, and why a negative result describes only what that test examined, with the discovery of ReNU syndrome in a non-coding RNA gene as the example. Questions to bring to a genetics appointment can be copied, and a table sets out how an advocacy group can prepare its consent language, samples, registry fields, and research connections so that new discoveries reach its members. The drug development timeline labels every program with the basis of its information, from regulatory approval and peer-reviewed publication to a company press release or website, and says plainly when results have not been peer reviewed and the drug is not approved. Educational and not medical advice. Runs entirely in your browser.

Reading a coverage decision

The HEOR Workbench

For advocates who have been handed a health technology assessment appraisal and told the therapy is not cost-effective. The panels explain what that sentence rests on: which outcomes were counted, whose preferences set the utility weights, which costs the chosen perspective leaves out, including caregiving and lost work, and how a threshold turns a ratio into a yes or a no. The panel on rare disease covers small populations, external comparators drawn from registries, rarity and severity modifiers, and managed access, and names the places where a community's own evidence changes the model's inputs. The calculators let you test what a different perspective or a different threshold does to the same therapy before you write the comment letter. Runs entirely in your browser.

Digital health technology

The Digital Health Technology Field Guide

A guided reference on wearables, sensors, and digital measures in health research, care, and policy, with the most depth on accelerometer and digital voice data. Choose a role (researcher or data scientist, policymaker or regulator, patient advocate, clinician, or new to the field) and the guide marks a reading path through the history of the field, the signals and their file formats and standards, analysis pipelines and the V3 validation framework, evidence by disease area with live searches of PubMed, Europe PMC, and ClinicalTrials.gov, open data sets, regulation and payment, organizations and people, funding, and the use of these tools in research and in patient-centered care. A planning worksheet exports to Markdown, and a shortlist with notes exports to CSV or a save file that reloads. Dated claims are labeled as a snapshot and linked to their sources. Runs entirely in your browser.

Planning workbook and document generators

The Community Chatbot Planning Toolkit

Co-production is the spine of this one. It sets out what sharing decisions with families looks like next to informing and consulting, who belongs on the panel, how to compensate and credit people for their expertise, and the charter terms that hold: written responses to every recommendation, clinical and community sign-off in separate lanes, and a rule that lets any panel member pause the project over a safety concern. Around that sit the roadmap, the vetted knowledge base, the technology in plain language, the safety layer, and testing with personas drawn from the people who arrive frightened and new. The generators produce a readiness check-up, a starter system prompt, a draft first-night response for panel review, a vendor question list, and a planning report for the board. Runs entirely in your browser.

Bringing lived experience into a public comment

Comment Desk

For research teams and patient partners who want a community's experience to reach an agency in a form staff can use. Agencies tend to weigh specific answers with evidence more heavily than general statements, so the desk follows the agency's own question numbering, asks each organization for its points, its sources, and any lived-experience example it has permission to share, and reminds contributors that filed comments are public and should hold nothing that could identify a child or family. A review checklist covers source checking, permissions, length limits, and sign-off, and the group keeps a record of what it filed and what the agency did next. Generic by design: it holds no positions of its own. Runs entirely in your browser.

Prompts, mappings, and ETL reference

The OMOP ETL AI Toolkit

For a team writing an OMOP ETL with an AI assistant beside them. Pick the stage you are on, from source profiling through mapping proposals, vocabulary candidates, code, tests, data quality assessment, documentation, peer review, and debugging a failed load, and the page assembles the prompt, with the rules on protected health information and an output contract built into every one so the answer comes back in a shape you can review. The mapping record holds each field-level decision with its rationale, its status, and the reviewer who approved it, and turns itself into a review prompt or exports as Markdown, CSV, or JSON. Alongside it: a prompt library for the work that comes around again, a repository scaffold with the shell script that creates it, a source system profile that becomes a mapping prompt, reference notes on the CDM tables and the vocabularies, and a glossary with the OHDSI documentation every note is drawn from. Where The OMOP ETL Worksheet tracks the project, this tracks the decisions inside it. One stage shows at a time, it runs entirely in your browser, and the session saves to a file that reloads.

EHR-to-OMOP pipeline

Registry Forge, Enterprise Edition

The full open-source pipeline: pull EHR data via SMART-on-FHIR and bulk exports, parse and map it to OMOP, then explore cohorts, exposures, and devices through ready-made dashboards.

Personal records to cohort

Registry Forge, Patient Edition

Turn health records people download themselves, from MyChart, Blue Button, and other C-CDA exports, into a research-ready cohort data set, with a live demo, EDA reports, and a privacy-first workflow.

FHIR bundle inspector

Bundle Reader

Paste a FHIR bundle and see where every resource would go in the OMOP CDM before you write any code, with the assumptions made visible: imputed dates, invented visit boundaries, inverted statuses, missing units, codes that map to nothing. Built-in exercises, including a real example from the FHIR specification. Reads R4, R5, and R6, runs entirely in your browser, and downloads to a single file for use with real records.

Trial data crosswalk

CDISC SDTM to OMOP Mapper

Variable-level mappings from the SDTM domains to their OMOP CDM v5.4 targets, with notes on where judgment is required and a direction toggle that shows which domains feed each OMOP table. Search across variables, review the vocabulary crosswalks from MedDRA, WHODrug, and CDISC controlled terminology to standard concepts, and read where trial data break the observational mold: partial dates, blinded exposure, and qualifiers with no home. The full spec downloads as a CSV for ETL planning.

Genomics pipelines and infrastructure

Bioinformatics Shop in a Box

For the team building the genomics side of the data. The germline short-variant and RNA-seq pipelines step by step with the tools common at each step, Nextflow, WDL, Snakemake, and CWL compared, validation against Genome in a Bottle, joint genotyping from gVCFs, cloud storage layout and costs, file transfer with checksums, containers and reproducibility, and a command reference for samtools, bcftools, GATK, the cloud command-line tools, Git, and Slurm. Generators write the analysis protocol, the IT setup request, and the transfer manifest. Runs entirely in your browser.

Hospital stay accommodations

Hospital Packet Builder

For families of people who cannot be left alone. Build a one page fact sheet on the accommodations you need, a door sign, packing lists for both of you, questions to settle before you arrive, and ready-to-send records requests. Fill it in, print it, and save it to reuse at the next admission.

Live research monitor

EpilepsyLive

A live research monitor for epilepsy: research, registries, and related tools, courses, and resources, in one place.

When researchers ask for your registry data

Data Sharing in a Box

For the executive director or board member who has been asked for registry data and has no process to point to. In plain language, it sets out what a researcher should submit, who reviews the request and against what, what the researcher's institution signs, and what happens to the data when the project ends, then writes those answers up as pages for your website and procedures for your committee. It explains why a registry run by an advocacy group is often outside HIPAA and what to ask counsel instead, why record-level data about a rare condition usually stay under controlled access even when deidentified, and what changes when the request comes from a company running a clinical trial in your community. Everything it writes is a draft for your attorney and Institutional Review Board (IRB) to review. Downloads as Word, PDF, Markdown, or a web page. Runs entirely in your browser.

Planning workbook and document generators

The Community Chatbot Planning Toolkit

For the board that has been asked to approve a chatbot, and the staff who would have to run it. Written in plain language for people without a technical background: what the project costs in staff effort as much as money, why the vetted content library is the work and the software is not, what to ask a vendor before signing, what the tool must never say to a family, and why an unmaintained medical information tool should come down rather than sit there. The readiness check-up scores your program, not your data, and says plainly when the answer is that a resource page or a helpline would serve your community better. Take the planning report to your board or attach it to a grant application. Runs entirely in your browser.

Challenge prizes and datathons

The Health Data Challenge Planner

For the board asked to approve a prize competition, and the staff who would have to run it. Sets out what the event costs in staff hours as much as prize money, why the data preparation rather than the software decides whether it can happen at all, what a sponsor can and cannot ask for in return, how contest and sweepstakes law and prize tax reporting apply, and how to pay patient partners without putting their benefits at risk. The readiness self-assessment scores your organization, not your data, and the go / no-go record gives the board a written decision to point at, including a decision not to run one. Downloads a budget, a timeline, and a full plan you can attach to a grant application. Runs entirely in your browser.

Writing a public comment

Sharing Summary Results with Research Participants

For patient organizations deciding whether to comment on the NIH draft policy on sharing summary-level study results with participants, with comments due October 26, 2026. Explains in plain language what the draft would require and why a results summary can read as a prognosis in a small, connected rare disease community, especially when it is broken down by genotype or includes findings that were not planned in advance. The comment builder starts from who is writing (a participant, a parent or caregiver, an advocate, a clinician, an investigator), goes through each theme with the case for the ask and its trade-off, takes your own examples, and produces a referenced letter you can edit before downloading. The plain-text version pastes into the NIH comment form. Nothing you type is stored, and a progress file saves your place. Runs entirely in your browser.

Making the case for including children

Pediatric Inclusion Commons

For an advocacy group that wants children included earlier in the research on its condition and needs to know where the rules come from and who else is working on the question. Start from the route written for patient organizations: the policies and guidance to cite when a protocol sets a minimum age of 18 without a stated reason, the incentives and requirements that shape what sponsors study in children, the ways a group can be heard by an agency, and the organizations, young people's advisory networks, consortia, and funders active in pediatric inclusion, with a map. Written in plain language with a glossary, and each entry links to its source. Runs entirely in your browser.

Statements and communication on genomic research

Reading the Dark Genome

For the board or executive director whose community is hearing about whole genome sequencing, non-coding findings, and new RNA-targeted therapies. Lists the society guidance and standards that bear on the topic, then drafts an organizational statement from the positions you choose, such as access to genome sequencing when narrower tests are uninformative, periodic reanalysis of stored data, a plan for recontact, responsible data sharing, careful use of computational predictions, and plain, dated communication about therapies. The ethics stage turns each issue into a policy question, and the drug development timeline shows how to describe a company's program to families: by its stage, by the basis of the information, and with the date it was checked. The draft is a starting point for your board and medical advisors to revise. Runs entirely in your browser.

AI prompts, in your voice

The Nonprofit AI Prompt Studio

AI drafts sound like a nonprofit that does not exist because the model does not know yours. Fill in your organization once, pick a task, from grant narratives to donor letters to board reports, and the Studio assembles a prompt that puts your facts in, keeps your language rules, and forbids invented numbers.

Board governance

The Well-Run Board

For the board that exists on paper and now has to function: a governance readiness self-assessment, a meeting cadence planner, the working minimum of Robert's Rules, a quorum and vote calculator, ready-to-send board emails, and a checklist of questions to bring to a lawyer.

Flashcards and self-quizzing

Omics Flashcards

The vocabulary and the reasoning behind omics measurements, in question and answer cards you can review in short sittings. Topics run from cells, DNA, and the central dogma through blood samples and biobanking, study design, genome sequencing, variants, RNA-seq, proteomics, metabolomics, quality control and statistics, multi-omics and gene-environment analysis, the GA4GH, HPO, and OMOP phenotyping standards, data sharing, and command-line tools. Every card explains its answer rather than asserting it, so the cards work as a reference as well as a drill. Flashcard mode, quiz mode with shuffled options and a running score within a topic, and a mixed review that draws at random across topics. Good beside Bioinformatics Shop in a Box and From Clinical Data to Omics, which cover the same ground at project scale. Sources are listed as links. Runs entirely in your browser.

Teaching implementation science

The Implementation Planner

For students, trainees, and anyone teaching implementation science who wants the frameworks applied to a real program rather than described in a lecture. Each stage explains its framework at the point where it is used: RE-AIM as questions a program team can answer without specialist training, the Consolidated Framework for Implementation Research as a vocabulary for resistance, the Proctor implementation outcomes as endpoints separate from health outcomes, the ERIC compilation as a shared language for what you are going to do about a barrier, and EPIS as the arc the work moves through, backward as often as forward. A glossary defines every term, including the acronyms, and the published papers behind each framework are listed with links. A student finishes with a written plan rather than a set of notes, and a saved file opens in an instructor's own copy. One stage shows at a time, and it runs entirely in your browser.

Teaching manuscript writing

Manuscript Paint by Numbers

For students and early career researchers writing a first paper, and for the mentors and instructors who guide them. Each phase opens with a plain-language explainer, the mistakes reviewers often find, a definition of done, and a worked example from a registry cohort study, and a glossary covers publishing terms in the sense journals use them. The reporting guideline pages show which checklist fits which design, and the overclaiming check flags causal verbs and stacked hedges in a draft discussion so a student can see why a sentence needs to change. A plan saves to a file that a mentor can open in their own copy, and a student reporting an OHDSI network study can link the plan built in Network Study Paint By Numbers and carry it into the write-up. Runs entirely in your browser.

Guided primer and prompt workbench

Research Data Science: A Guided Tour

For scientists, clinicians, and advocates who are not data scientists. Short modules take you from framing a question through study design, statistics, literature and trial APIs, vocabularies, FHIR, OMOP, real-world data, omics, genetics, GIS, governance, FAIR data, and reproducible analysis, with a rare epilepsy question worked all the way through. A prompt builder assembles a structured brief from your project details, and a prompt library, a filterable tool directory, and a glossary sit alongside. Search every topic, mark modules read, and deep-link to any module. Runs entirely in your browser; nothing you type leaves it.

Skills map for clinical data analysts

From Clinical Data to Omics

For analysts and statisticians fluent in SAS, SPSS, Stata, SQL, or R who have been asked to run or lead a genomics or multiomic project without a biology background. It treats study design, governance, provenance, and statistical reasoning as credit you already hold, and the file formats, sequencing vocabulary, and compute environment as the terrain that is new. Rate where you stand and the ratings flag every stage of the curriculum that follows. The translator takes a task you do today, pulling the first rows of a file too large to open, counting distinct patient identifiers, joining a code set to an extract, reshaping wide to long, comparing two versions of an extract, and shows it on the command line and in R and Python against an invented registry extract, with a note on what changes when the file grows. The staged curriculum runs from the shell and file formats through version control, environments, the compute environment, the path from sample to counts table, pipeline languages, matrices and dimension reduction, reading heatmaps and principal component plots, and the statistical traps of high dimensional data, each stage naming what you can do once you have it and linking to freely available sources. A readiness check-up examines the project and the practices around it, not patient data: file naming, write-protected raw files, documented environments, recorded reference versions, and who could reproduce the result if you left. Everything assembles into a roadmap you can print, download as Markdown, or save to a file that reloads. Biology knowledge stays necessary, and the goal is to lead the project and read its output critically rather than to replace a bioinformatician. Runs entirely in your browser.

Self-guided course and calculators

The HEOR Workbench

For anyone who keeps hearing QALY, ICER, and threshold in meetings without a place to look them up: advocates reading a health technology assessment appraisal for the first time, research staff and students, and registry teams whose data sets become the inputs to an economic model. Short panels build from what health economics and outcomes research is, through outcome measures and utility weights, costs and perspective, the incremental cost-effectiveness ratio and the plane it is plotted on, budget impact and the analysis family, and real-world evidence in health technology assessment, each ending with a knowledge check that marks the panel complete. The calculators run the same methods on your own numbers, a glossary spells out every acronym, and the threshold and agency figures are dated snapshots with links to the primary sources. No economics background is assumed, only comfort with arithmetic. One panel shows at a time, it runs entirely in your browser, and the session saves to a file that reloads.

Digital health technology

The Digital Health Technology Field Guide

A guided reference on wearables, sensors, and digital measures in health research, care, and policy, with the most depth on accelerometer and digital voice data. Choose a role (researcher or data scientist, policymaker or regulator, patient advocate, clinician, or new to the field) and the guide marks a reading path through the history of the field, the signals and their file formats and standards, analysis pipelines and the V3 validation framework, evidence by disease area with live searches of PubMed, Europe PMC, and ClinicalTrials.gov, open data sets, regulation and payment, organizations and people, funding, and the use of these tools in research and in patient-centered care. A planning worksheet exports to Markdown, and a shortlist with notes exports to CSV or a save file that reloads. Dated claims are labeled as a snapshot and linked to their sources. Runs entirely in your browser.

GitHub, explained

GitHub, One Step at a Time

For anyone who feels panic when asked to work with GitHub: explain your goal and current status, and the page builds an ordered plan from creating an account through repositories, push and pull, branches, .gitignore, templates, GitHub Pages, MkDocs, and GitHub Actions, in short lessons you check off as you go, each with the point-and-click route and the command-line route. A generator writes the starter files for your project: a .gitignore for your languages, a README skeleton, an MkDocs configuration, a deploy workflow, and the first commands to run.

Command line, explained

The Command Line Starter Kit

For anyone who has never opened the black window. What PowerShell and Command Prompt are, how they differ, how to open each one, and how to read a command so you know what it will do before you press Enter. A generator writes the command for the task you pick, listing every file on a mapped drive, saving the listing as a CSV for Excel, counting files, or finding them by type, size, or date, in both tools, with a picture of what the window will show and a plain explanation of every part. Everything on the page only reads your files.

Statistical software crosswalk

Stata, SPSS, and SAS to Python and R

For anyone moving from Stata, SPSS, or SAS to open-source tools. Describe a data or statistics task in plain language, or type a command you already know, and get the Python and R equivalent as the answer, with the same operation shown below in the language you came from. Each one has a short explanation and a link to the official documentation, and caveat notes flag where a default differs, such as the equal-variance choice in a t-test or the reference category in a regression. Covers reading and inspecting data, tabulation, reshaping, group comparison tests, regression, and survival analysis. Runs entirely in your browser.

Live monitor and OMOP reference

MCH·OMOP Live

Maternal and child health meets the OMOP common data model. A live monitor of articles, preprints, funding, videos, and code, drillable to MCH topics and exportable to CSV, plus the teaching material around it: a starting path for people new to observational research, cheat sheets on vocabulary, cohort design, and treatment pathways, and dated worked examples on menopause and preterm labor that take one clinical idea from a vocabulary search through concept sets, cohorts, and a feasibility assessment. Also a plain-language explorer for the 258 fields of the model.

Course organization

The Course Planner

One place to organize a shared, online, or inherited course: the schedule, the teaching team, live session plans, student due dates alongside grading deadlines, the questions students keep asking, and the materials that still need updating. Download a draft when you are done and upload it next term to keep going.

Searchable study guide

In Your Own Words

Epidemiology, biostatistics, and informatics concepts in plain language: 252 defined terms, the core formulas, and the worksheets, all searchable.

Books & toolkits

Ten Habits

Two short, plainspoken books about data analytics and grant writing drawn from three decades in the field, each paired with a free toolkit of templates, checklists, and resources you can start using today.