NSECURED MemberGo / Medical research

Give each approved study the fields its protocol calls for.

The same clinical record can be relevant to several studies. Each protocol defines its own scope. Research teams can bind a study purpose and its approved fields to a specific analysis workload, limiting exposure to identifying details.

Test tubes in a laboratory

Medical research

Which data does this workload need?

Fields to protect

Clinical identifiers, laboratory results and study data. Identifiers and research variables can have separate field types and access conditions.

Fields eligible for this purpose

Subject to the approved scope and basis for processing, a study may use custom fields such as age band, a specified lab result and visit stage.

Synthetic data example

Two access decisions in context

Access allowed

Purpose
study_a
Fields
lab_value_v1 lab_value_v2
Result
200 · access_id: demo_4e1b

Study A has approval to compare a lab value across two visits. Its authorized workload can receive the study fields needed to calculate that change.

Request denied

Purpose
study_b ×
Fields
lab_value_v1 lab_value_v2
Result
403 data_access_denied

A service account for Study B cannot reuse Study A’s authorization simply because it belongs to the same hospital. Without a valid basis for the new purpose, its entire request is denied.

The fields and decisions here are illustrative. Actual access depends on tenant settings, the basis for processing and the approved purpose.

Workflow

How the workflow fits together

  1. 01

    Start with the protocol

    The investigator and data steward list the research question, required variables and identifiers to exclude.

  2. 02

    Define the study workload

    Define the approved purpose, model and study fields in the access policy, then test against synthetic records.

  3. 03

    Check a change of purpose

    Submit requests against the same records under the approved and unapproved study purposes, and compare the decisions.

Proof of concept

Test two study protocols with synthetic data

Invite the investigator and data steward with a field dictionary containing no real patient records. Agree on the variables each protocol may use, then test the two study accounts.