XX Research
ApproachFor health systemsMethodBlog
Sim · Online
Population Lab · All states
Population Lab/Multi-agent simulation of patient populations

Simulate every intervention on every patient — before you deploy it on one.

Run the experiment you can't run in real life — grounded in real-world data, in 24 hours.

See methodology
§ 01 / The problem

Health systems spend millions on interventions that don't move outcomes — because they can't test them on the actual population first.

01

Outreach campaigns miss the people they were designed for.

A new screening program launches. Six months later, uptake is flat in the very ZIP codes it targeted. The post-mortem identifies trust, language, and access barriers — that a synthetic population could have surfaced before a dollar was spent.

02

Operational protocols fail under real load.

The Transfer Center handles more than half of admissions. Black-swan volume surges, staffing changes, and discharge-timing tweaks have non-linear effects on door-to-doctor time. Spreadsheets cannot model them.

03

EHR and workflow changes shift burden instead of removing it.

A documentation-streamlining initiative reduces clicks for one specialty and triples them for another. By the time the data shows it, the rollout is already system-wide.

04

Predictive analytics project the past. Interventions are about the future.

Regression on historical data cannot tell you how patients will respond to a program that has never existed. Agent-based simulation can.

§ 02 / How it works

From census data to a simulated patient response, in three steps.

Every agent maps to verified U.S. Census and ACS attributes — age, income, household composition, language, employment, healthcare access — down to the ZIP code.

STEP 01DEFINE

You ask the question.

Frame any policy or intervention as a research question. We rewrite it, run dual research, and stage 2,000 synthetic agents.

‹ BackHub✓POPULATION2SCENARIO3RESEARCH4SIMULATION5REPORT

Define Scenario

RESEARCH QUESTION
e.g., How would a CMS-mandated equity-stratified screening campaign affect HPV vaccine uptake across rural ZIP codes?0/1500
Will a CHW outreach close the colorectal-screening gap in low-income ZIPs?
How would a $0-copay diabetes program shift A1c outcomes by income tier?
What happens to flu uptake if pharmacy access drops in rural counties?
How does a telehealth-first triage protocol affect ED throughput?
SYNTHETIC AGENTS beta max: 5,000
1,000
2,000
5,000
RESEARCH SOURCES
⊕
Web Search
Always on
⎙
File Search
No repo linked
⌕ Link
STEP 02SIMULATE

We run the population.

Agents post, comment, form communities. 80 activations per round across 10 rounds, in your private graph.

✓POPULATION✓SCENARIO✓RESEARCH4SIMULATION5REPORT
RD 5/10 · 03:02 · 507 POSTS · 6 COMMUNITIES · 528 ACTIVATIONS
SCREENING REMINDER · TRUST DIFFUSION · DROP-OFF
MI_014LA_016MO_011CA_006UT_007KS_002IA_015IA_015NC_010IN_004CT_004MD_016ID_009AL_009KY_002MO_013MA_012CT_005MS_016TX_014UT_007OH_017OK_015KY_007AR_010FL_014IN_005NE_004AL_011UT_006WA_002CO_010AR_012CA_004SC_001TX_014IA_004MD_010GA_011OK_016NJ_009WA_016KY_013LA_002AR_002IA_017KS_018NV_009NC_017IN_010MN_001CO_015KS_017KS_010CO_016AL_015MI_016WV_012MI_016CT_006MI_006WV_014NC_004MD_007NM_001CT_001WA_011NM_009CO_005FL_011GA_017NC_004MD_001AZ_007KY_011MO_009FL_008ID_007MN_003WV_018AL_014AL_011AZ_008HI_001NM_004CO_011NV_005TX_004FL_007NM_015AL_018GA_017UT_018WI_012MA_003AZ_005KS_017VA_017KY_011LA_016MI_008HI_014AR_014HI_001KS_011OH_003UT_001IA_010TN_016NM_017
+
−
STREAM · ROUND 5 OF 10LIVE
80 ACTIVATED · 75 POSTS · 84 COMMENTS
COMMENTAL_002 → NY_010
My mom didn't get screened until 58 and by then it was already stage 3. I'm 42 — guess I'll go ask my PCP next week. The free clinic on 6th will see you without insurance.
POSTCA_008
$0 copay sounds nice but the nearest clinic is a 90-min bus ride. Who's paying for the time off? Half my coworkers can't even take an hour without losing the shift.
POSTTX_005
Got the SMS reminder. Already booked. Took 30 seconds in the app.
COMMENTFL_012 → GA_003
Respectfully — the clinical-pathway argument only holds if the no-show rate stays under 18%. In my ZIP it runs north of 30%. Outreach without transportation funding is theater.
STEP 03STRATIFY

You read the response.

Stance distribution across age, education, income, race, sex — equity-stratified, never aggregate.

DEMOGRAPHIC BREAKDOWN
Predicted response to a CMS-mandated equity-stratified screening campaign, stratified across the simulated population.
AGE
EDUCATION
SEX
INCOME
RACE
STRONGLY OPPOSE
OPPOSED
LEAN OPPOSE
CONFLICTED
LEAN SUPPORT
SUPPORTIVE
STRONGLY SUPPORT
18–29118
38.8%
9.5%
5.2%
8.6%
35.3%
30–4484
33.3%
11.9%
17.9%
32.1%
45–64162
21.0%
8.6%
4.3%
6.2%
22.8%
36.0%
65+96
14.6%
6.3%
9.4%
24.0%
41.6%
← OPPOSECONFLICTEDSUPPORT →
01 / 03
§ 03 / From the Lab

Notes on simulation, healthcare, and the future of intervention design.

How we think about agent-based modeling, synthetic populations, and the gap between predictive analytics and decision support.

All posts
●Issue №03·APR 26
methodology
COVID-19.
X · R
April 25, 2026·Field Note

Hindcasting the LA County COVID Vaccine Rollout: A Pre-Registered Stress Test for Multi-Agent Healthcare Simulation

We're running a pre-registered hindcast on the 2021 LA County COVID-19 vaccine rollout — 500 census-grounded LLM personas reacting to real interventions on a simulated feed, blind to the outcomes we'll later compare against.

Read the issue
●Issue №02·APR 26
product
Sim-Labs.
X · R
April 12, 2026·Field Note

Introducing Simulation Labs: Open-Access Simulation for Everyone

Simulation Labs is our open-access product — a way for anyone to experience the power of population-scale AI simulation without a sales call.

Read the issue
All posts
§ 04 / For health systems

What you can simulate.

Three families of decisions where simulation has the highest leverage. Each delivers a clinical-champion-ready report with predicted outcomes, equity stratification, and intervention sensitivities.

Simulate community health worker programs, mailer campaigns, mobile clinics, navigator pathways across real ZIP-code demographics. Surface equity gaps before deployment.

Example: colorectal screening in Boyle Heights — predicted uptake by age × language × insurance over 6 months, under three campaign designs.

Model the Transfer Center, ED, and inpatient flow as agent populations with capacity constraints. Stress-test discharge timing, staffing changes, and surge response.

Example: simulate Keck Hospital's Transfer Center under a 30% flu-season volume surge — bed-block points and Pareto-optimal staffing.

Simulate provider workflows as agent populations. Test inbox-redesign, top-of-license role alignment, automated waitlists — measure friction, retention proxies, and adoption curves.

Example: model a documentation overhaul across 23,000 providers — net administrative burden by specialty, before rollout.

§ 05 / Built where the work is

Originated inside USC Keck Medicine.

X Research was founded by a USC Keck Medicine staff member working alongside clinicians, operators, and researchers across the system. Our first reference designs — Transfer Center throughput, CBSA equity intervention, provider workflow simulation — are anchored to the operational realities of a high-acuity academic medical center.

X Research is an independent company. References to USC Keck Medicine reflect the founder's individual employment affiliation and do not constitute endorsement by the University of Southern California.

§ 06 / Get started

Request a pilot.

Tell us what intervention you want to test on a synthetic population. We will scope a pilot tailored to your health system.

FORM 06.1 · Pilot requestrequired *
XX Research

Census-grounded synthetic populations and agent-based simulation for healthcare interventions.

Product
ApproachUse casesMethodDeveloper / API
Company
BlogContact
Legal
PrivacyTerms

© 2026 X Research · All rights reserved

X / TwitterLinkedInv0.2