Platform capability

A glossary of the words on this site.

Each entry is a working definition for readers of these pages. It is not a legal definition and not a quotation from a standard. Where a standard exists, follow the standard.

Terms

Adverse event

Any untoward medical occurrence in a participant, whether or not it is considered related to the intervention.

SAE

A serious adverse event. Seriousness is not the same thing as severity or relatedness.

SUSAR

A suspected unexpected serious adverse reaction. Reporting clocks depend on the applicable rule.

Blinding

Keeping one or more parties unaware of the assigned intervention.

Allocation concealment

Hiding the next assignment until the participant is enrolled. A predictable list breaks it.

Randomization

Assigning interventions by a chance procedure written in advance.

Stratification

Randomizing within categories so those categories stay balanced.

Estimand

The treatment effect the study intends to estimate, including how intercurrent events are handled.

Endpoint

The variable used to judge an outcome.

Protocol

The document that describes the objectives, design, methods and organization of a study.

Amendment

A written change to the protocol after it has been approved.

eCRF

An electronic case report form. A field value plus the history of changes.

Query

A question about a data value. It does not itself change the value.

Source data

The original record of an observation. Verification compares the eCRF with it.

Database lock

A signed decision that the analysis snapshot will not take further routine edits.

Freeze

A checksummed copy of the data an analysis will use.

SDTM

A CDISC model for tabulating collected data. A shaped file is not a certified submission.

ADaM

A CDISC model for analysis data. Same caution as SDTM.

Define-XML

A CDISC document that describes the datasets. Schema validation is a separate step.

CDASH

A CDISC guide for how data are commonly collected.

ODM

CDISC Operational Data Model, often used to describe forms.

CTMS

The operational record of sites, visits, milestones and issues.

EDC

The system of record for case report form values.

eTMF

The electronic trial master file of essential documents.

RTSM

Randomization and trial supply management.

IRT

Interactive response technology used at the moment of allocation.

eConsent

A consent process that records the version and the signatures. Legality depends on the jurisdiction and the approval.

ePRO

An outcome reported by the participant on an electronic instrument.

eCOA

Electronic clinical outcome assessment. Includes PRO, ClinRO, ObsRO and PerfO.

ClinRO

An outcome reported by a clinician.

ObsRO

An outcome reported by an observer who is not the participant.

PerfO

An outcome based on a task the participant performs.

PRO

A patient-reported outcome.

RBQM

Risk-based quality management. Effort follows prespecified risks.

KRI

A key risk indicator. A measurement, not a verdict.

QTL

A quality tolerance limit the team set for a critical factor.

CtQ

Critical to quality. A factor that, if wrong, threatens participants or the result.

CAPA

Corrective and preventive action. Effectiveness should be checked by someone else.

Monitoring

Oversight of the study, on site or centrally, written down by a person.

Deviation

A departure from the approved process, recorded with a reason.

Audit trail

A record of who changed what, and when. It is not deleted to tidy a screen.

Electronic signature

A signature bound to a person and to the hash of what they signed.

Part 11

21 CFR Part 11, the US rule on certain electronic records and signatures. Controls are not a compliance claim.

ICH E6

The ICH guideline on good clinical practice.

ICH E9

The ICH guideline on statistical principles for clinical trials.

GCP

Good clinical practice. A set of duties. Software does not possess them.

SPIRIT

A checklist for trial protocols.

CONSORT

A checklist and flow diagram for reporting randomized trials.

STROBE

A checklist for observational studies.

STARD

A checklist for diagnostic accuracy studies.

PRISMA

A checklist for systematic reviews.

TRIPOD

A checklist for reporting prediction models.

ISO 14155

The international standard for clinical investigations of medical devices. Certification is separate.

MDR

The EU Medical Device Regulation. A legal text, not a software mode.

IVDR

The EU In Vitro Diagnostic Regulation.

CTIS

The EU Clinical Trials Information System. This product does not submit to it.

IND

Investigational New Drug application in the United States.

IDE

Investigational Device Exemption in the United States.

PHI

Protected health information. Do not type it into a public form.

FHIR

HL7 Fast Healthcare Interoperability Resources. A standard, not a certified interface.

OMOP

A common data model published by OHDSI.

DICOM

The standard file format for medical images.

PS3.15

The DICOM de-identification profile guidance. A local profile is not a certificate of conformance.

MedDRA

A licensed adverse-event dictionary. Not included in this product.

WHODrug

A licensed drug dictionary. Not included.

SNOMED CT

A licensed clinical terminology. Not included.

LOINC

A code system for laboratory and clinical observations.

ICD

The International Classification of Diseases.

CDISC

The organization that publishes CDASH, SDTM, ADaM, ODM and Define-XML.

Core

CDISC CORE, a conformance engine. This product does not run it.

Pinnacle 21

A widely used CDISC validation tool. This product does not run it.

Sensitivity

The proportion of people with the condition who test positive, in a stated population.

Specificity

The proportion of people without the condition who test negative.

ROC

A curve of true-positive against false-positive rates across thresholds.

Kaplan–Meier

A method for estimating the proportion still event-free over time, with censoring.

Cox model

A regression for the hazard, under a proportional-hazards assumption.

Censoring

A participant’s event time is only known to be after a date.

E-value

A sensitivity number for how strong an unmeasured confounder would need to be. Not a proof.

Propensity score

A modeled probability of receiving the exposure, used in observational analysis. It is not randomization.

IPW

Inverse probability weighting. A way to reweight observations. It depends on the model.

Power

The probability of detecting a stated effect if the assumptions hold.

Multiplicity

The problem of looking at many endpoints or many times.

Intercurrent event

An event after treatment starts that affects interpretation, such as discontinuation.

Screen failure

A person who was screened and not enrolled.

Delegation log

The list of who may do which study task.

Essential document

A document that allows the conduct of the study and the quality of the data to be evaluated.

TMF

Trial master file.

CSR

Clinical study report. ICH E3 describes its structure. The software does not emit a finished one.

DSMB

A data and safety monitoring board. The software does not sit on it.

Unblinding

Revealing an assignment. It should be exceptional and signed.

Kit

A packaged investigational product with an identity and a status.

UDI

Unique device identifier.

Excursion

A temperature or handling event outside the labeled range. A person records it.

Provenance

A statement of where a value came from and what was done to it.

Checksum

A hash of bytes, so a later reader can see whether they changed.

Tenant

An organization whose data are not visible to another organization by default.

Sandbox

An organization allowed to hold only synthetic participants.

Scope

The list of actions an API key is allowed to perform.

Webhook

An outbound call to an address you named. Private addresses are refused.

Validation

Documented evidence that a system meets its requirements. A deploy is not a validation.

Audit pack

A download of quality tables. Not an inspection certificate.

Synthetic data

Invented records used for demonstration. They are not participants.

Sponsor

The party that takes responsibility for starting and managing a study, as defined by the applicable rule.

CRO

A contract research organization acting for a sponsor under a written scope.

Investigator

The person responsible for the conduct of the study at a site.

Informed consent

A person’s voluntary decision after being informed. The version and the signature are the record.

Related

Questions

Why is a licensed dictionary not defined in full?

Because the content is licensed. The name and the limitation are enough.

Available with limitations means the described software behavior exists. A professional service is work a person does under a separate scope. Nothing here is a certification.

See the workflow on your protocol.

The conversation is about your study design and what the software can and cannot do today.