Clinical & Research Statistics
- Study design & power analysis
- Bayesian & frequentist modeling
- Interim monitoring & adaptive designs
- Manuscript-ready results & interpretation
Statistical Consulting · Research Software · Education
Rigorous statistics.
Deployable software.
Research that matters.
I partner with clinical researchers, medical teams, and labs to blend statistical rigor with production-quality software - so your decisions hold up under scrutiny and your systems scale without heroics.
- Projects delivered
- 140+
- Years in applied statistics
- 10+
- Focus
- Clinical · Software · Strategy
Trusted by
Cedars ResearchAtlas TrialsNorthwind LabsHelix HealthSummit Analytics
- Role
- Statistician & Software Engineer
- Specialty
- Clinical Research & Biostatistics
- Lab
- Thornton Statistical Research Lab
- Stack
- R · Python · Stan · Next.js · Postgres
Who I work with
Medical research teamsClinical trial sponsorsAcademic labsHealth tech startupsPharma & biotechPublic health orgs
Services
What I actually deliver
Custom Research Software
- Data pipelines & automation
- Interactive dashboards & apps
- Statistical tooling & APIs
- Reproducible analysis frameworks
Advisory & Education
- Fractional statistical leadership
- Team training & workshops
- Grant support & methodology review
- Curriculum & course development
About
Hi, I'm Micah Thornton.
Statistician, software builder, and research advisor. I move between R/Python notebooks, registrational study plans, and production codebases without handing anything off to chance.
The job is to uncover the signal, package it so people can act on it, and leave teams with systems they can evolve. I partner directly with founders, PIs, and R&D leaders - no bloated teams, just thoughtful work and fast loops.
Principles
- Evidence over theatrics
- Software is part of the analysis
- Reproducibility by default
- Clarity for stakeholders
Tooling
RPythonStanNext.jsSupabasePostgresDuckDBAirflowAWS
Work
Recent outcomes
Cut recruitment time by 6 months
Rebuilt the statistical analysis plan, automated interim checks, and gave sponsors realtime dashboards.
Launched HIPAA-ready tooling in 8 weeks
Designed the modeling approach, built the orchestration layer, and handed off runbooks to the product team.
Reporting cycle: 30 days → 48 hours
Tuned the models, codified QA, and deployed a reproducible pipeline the team now owns.
Ways to work
Pick the cadence that fits
Sprint build
3-6 weeksComplete study plan, modeling approach, or bespoke tool delivered fast.
- Statistical plan or product spec
- Interactive dashboards/notebooks
- Runbooks + live walkthrough
Fractional lead
QuarterlyHands-on leadership embedded with your team to guide studies and cross-functional decisions.
- Weekly working sessions
- Review + sign-off on analyses
- Mentorship for analysts/engineers
Advisory retainer
MonthlyAs-needed access for audits, second opinions, or roadmap support without spinning up a full engagement.
- Same-day feedback
- Lightweight artifacts
- Private channel + office hours
Portals
Choose your lane
Each routes to a dedicated property
🏥
Launching soonConsulting Portal
Engagements, proposals, secure data rooms, and project dashboards for active clients.
Enter portal⚙️
In developmentSoftware & R&D
Custom tools, model sandboxes, and private documentation for ongoing build-outs.
Enter portal🔬
Beta accessResearch Lab
Collaboration space for partner labs, grant workstreams, and reproducible pipelines.
Enter portal📚
PreviewEducation
Courses, workshops, and cohort support for students and research teams.
Enter portalProcess
Less ceremony. More leverage.
Every engagement follows the same four-step rhythm — fast to start, rigorous in execution, clean at hand-off.
01
Scope call
30 minutes. You describe your constraints, data, and goals. I ask the hard questions. No slides, no pitch.
02
Frame
Audit what you have, surface assumptions, and choose the statistical or software approach that fits reality — not the textbook.
03
Build
Prototype quickly, lock down governance, and deliver analyses or tooling that stakeholders can trust and verify.
04
Operationalize
Hand off documentation, dashboards, and runbooks so your team extends the work without ongoing dependency.
Testimonials
What partners say
“Micah translated our messy study into a statistical plan the IRB approved immediately.”
CP
Clinical Program Lead
Neuro Device Sponsor
“He moves seamlessly between code and strategy. Our analytics platform shipped weeks ahead of schedule.”
HP
Head of Product
Health Tech Startup
“The lab finally has reproducible pipelines and documentation. We can iterate without firefighting.”
PI
Principal Investigator
Academic Research Lab
Resources
Methods briefing
Working notes on trial design, model choices, and tooling decisions.
Access resourceCapabilities deck
One-pager covering scope, timelines, and past projects you can share internally.
Access resourceWriting
Thinking out loud.
Notes on statistics, software, and research
When to use Bayesian vs. frequentist - a practical guide for clinical researchers
The choice isn't philosophical. It depends on your prior information, regulatory context, and what you're trying to communicate.
Coming soonBuilding reproducible pipelines: the stack that doesn't break in six months
Most analysis code is write-once. Here's how to build workflows your team can actually maintain and extend.
Coming soonWhat to put in your statistical analysis plan (and what to leave out)
An SAP that's too vague gets you in trouble. One that's too rigid ties your hands. Threading that needle takes practice.
Coming soonQuick brief
Prefer a lightweight intro?
Share a couple of lines about your project and I'll follow up within two business days with next steps, timelines, and what I'd need to get started.
Or email micah@thorntonstatistical.com
FAQ
Common questions
Do you take on fractional or retainer work?+
Yes. Most partnerships blend fixed-scope builds with an ongoing cadence for questions, audits, or roadmap support.
Will you work with internal engineers or analysts?+
Absolutely. The best outcomes happen when we share the same repo, dashboards, and delivery rhythm.
What industries do you focus on?+
Primarily medical research, clinical trials, pharma, public health, and health tech — but rigorous statistics applies broadly.
What does delivery look like for software builds?+
Thin slices ship fast: spec → prototype → hardened build with docs, tests, and runbooks your team owns.
Do you teach workshops or offer student support?+
Yes. I run bespoke workshops for research teams and maintain a students portal with office hours and resources.
How do engagements typically start?+
A 30-min working session to understand your constraints, data, and goals. From there I'll sketch a path within 48 hours.
Get in touch
Ready to work together?
Send a brief with your hypotheses, timelines, and constraints. I'll respond with a concrete path - statistical design, a custom tool, or a hybrid.