Disease associations · Patient foundations
Your patients said yes to research. Not to everything
A foundation's real assets are its patients' trust and the centres that care for them. Once a file leaves, consent language is usually a hope. On CuRE it is a rule the data platform checks before anything moves, and your network becomes something you can ask a question of without becoming a data warehouse
Trust you can enforce
Four things foundations don't think to ask for
Each one usually lives in a contract, an inbox or a spreadsheet. Here each one is something the platform does
01
The fine print holds
Write down the purposes an agreement allows, from clinical research to safety surveillance. When a partner declares any other purpose for a release, market access for instance, the release is refused before a single row moves, even though the agreement is active
Purpose-checked data release
02
Ask forty centres without collecting a record
Enter a sponsor's criteria once. The question runs inside each centre's own records, under that centre's own rules, and only a privacy-protected total comes back. No patient-level row leaves any centre, and nobody spends six weeks chasing coordinators
Federated feasibility counts
03
Out of one deal, still in the registry
A patient who objects to one partnership can leave just that agreement. The withdrawal reaches the data platform, the patient portal, the site's EDC and the randomization system in one step, and her natural-history data stays in the registry she still supports
Per-agreement withdrawal
04
The committee clicks Grant, and that is the access control
An outside investigator applies on the dataset's own page. The committee sees aggregate counts with small cells hidden and an advisory summary against your policy. Access is issued against the signed agreement and the IRB reference, so a revoked agreement actually stops the data
Data access committee workflow
One network
From the agreement to the results letter
The same rules travel with the data at every step, so a foundation can say yes to more partners without promising its patients less
01
Agree
The purposes each partner may use your data for are written once and checked on every release, by every partner, at every centre
02
Ask
A sponsor's "how many patients can you reach?" becomes one question to the whole network instead of forty emails
03
Find
Designed to show which community centres actually see the patients your registry is missing, read from their own records before a trial picks its sites
04
Study
Built toward the day a trial runs inside the population you already follow, with a separate consent for randomization checked on every draw
05
Give back
Trial results return to every participant in plain words, drafted from the locked results and checked by study operations before they appear on the portal
Why foundations sit at the centre
Every centre that joins makes your next answer better
A foundation is the one organization every centre in its disease already trusts. That makes it the natural place for a research network to start, and the place where the cost of a broken promise is highest
On CuRE the network compounds. A centre that connects for any reason, a sponsor's trial or its own registry work, makes your next count more complete. A pharma partner working with three foundations meets the same enforced terms at each. And the centres get something back: their own patients' data, cleaned and harmonized, instead of an annual PDF
That is the exchange: centres supply the data, partners need it, and the foundation keeps the terms
Outcomes, not features
What changes for a foundation
Most research software automates the handoffs between systems. CuRE is built to remove them, so the difference shows up as questions you can finally ask and work nobody has to redo
New questions
Ask the questions you stopped asking
Is this subject enrolled in someone else's trial? How many eligible patients sit across forty centres? Did the cold chain matter to the outcome? Questions that used to start a data project are designed to be a query on one shared record
Fewer handoffs
Map once. Stop re-typing
Data maps to OMOP once and every app reads the same record. Next month's CRO file is recognized and replays in one click, and the same value is never keyed into three systems to be reconciled later
Time to evidence
Start from data that is already there
A new study, a sponsor's feasibility question or a fresh safety signal starts from the record that already exists, not from a new extract and a data-engineering project
Evidence as a byproduct
The audit trail writes itself
Signatures are bound to the exact content they approved, every AI suggestion is sealed with the model that made it, and a CAPA can close on the metric it was meant to fix. Inspection evidence comes out of the work itself
One platform
One record instead of a vendor stack
Randomization, safety, quality, TMF and analytics share one data model, so adding a capability means switching it on rather than buying another integration project and validating it again
Bring us the question your network can't answer yet
A sponsor's feasibility request, a consent promise you can't enforce, a registry that should be running a trial: tell us which one keeps coming back