HRIP study materials
HRIP study resources

HRIP key terms

HRIP · Get comfortable with the vocabulary.

Data governance

The system of decision rights and accountabilities that determines who may define, produce, use, and change healthcare data, and how those decisions are enforced across an organization.

Related: Data stewardship, Data ownership, Metadata
Data stewardship

The operational role of named individuals who manage a data domain day to day, resolving quality issues, approving definitions, and enforcing governance policies set at the executive level.

Related: Data governance, Data quality dimensions, Master patient index
Data ownership

The assignment of ultimate accountability for a data domain to a specific leader, including authority to approve access, define standards, and accept or reject quality thresholds.

Related: Data governance, Data stewardship, Minimum necessary standard
Metadata

Structured information that describes other data, such as field definitions, allowable values, source system, refresh timing, and ownership, enabling users to find and interpret data correctly.

Related: Data dictionary, Data lineage, Metadata registry
Data dictionary

A catalog of every data element in a system or study, listing its name, definition, data type, allowed values, units, and responsible steward so users interpret fields consistently.

Related: Metadata, Controlled vocabulary, Case report form
Data lineage

Documentation tracing each data element from its point of origin through every transformation, aggregation, and load, so analysts can verify provenance and diagnose discrepancies.

Related: Metadata, Extract-transform-load, Audit trail
Master patient index

An enterprise registry that assigns one unique identifier to each person across systems, linking records from different sources while flagging potential duplicates for review.

Related: Entity resolution, Health information exchange, Data stewardship
Data governance council

A cross-functional body of clinical, operational, compliance, and analytics leaders that sets data policy, arbitrates disputes over definitions, and prioritizes remediation work.

Related: Data governance, Data stewardship, Data quality dimensions
Controlled vocabulary

A fixed, curated list of permitted terms with defined meanings, used so that values entered by different people remain comparable and machine-processable.

Related: SNOMED CT, LOINC, Data dictionary
SNOMED CT

A comprehensive clinical terminology of concepts with hierarchical relationships, used to encode findings, procedures, and diagnoses in electronic health records with computable meaning.

Related: ICD-10-CM, Controlled vocabulary, Value set
LOINC

A standard code system for identifying laboratory tests and clinical observations, pairing each code with the specimen, property, timing, and scale measured.

Related: Controlled vocabulary, HL7 FHIR, Data dictionary
ICD-10-CM

The diagnosis classification used for morbidity coding and billing in the United States, organized by body system and clinical condition rather than by concept relationships.

Related: SNOMED CT, Value set, Case mix
Value set

A defined subset of codes drawn from one or more terminologies, assembled for a specific purpose such as a quality measure denominator or a reportable condition list.

Related: SNOMED CT, LOINC, Clinical quality measure
Case report form

The structured instrument, paper or electronic, on which study staff record protocol-specified data for each participant, with fields mapped to the analysis dataset.

Related: Data dictionary, Source document, Electronic data capture
Electronic data capture

A validated software system used in research to enter, edit, and store study data directly, replacing paper forms and enforcing edit checks at the point of entry.

Related: Case report form, Audit trail, Query management
HL7 FHIR

A modern interoperability standard that models healthcare information as modular resources exchanged over web APIs, supporting both granular queries and document-style bundling.

Related: HL7 v2 messaging, Application programming interface, US Core
HL7 v2 messaging

A widely deployed pipe-delimited message standard used for admissions, orders, and results, with event types and segments that vary by site through optional fields.

Related: HL7 FHIR, Interface engine, Health information exchange
Health information exchange

The electronic movement of clinical information among separate organizations, governed by agreements on consent, content, and technical transport so treating clinicians can see outside records.

Related: Master patient index, Consent management, HL7 FHIR
Interface engine

Middleware that receives, transforms, validates, and routes messages between systems, applying mappings and filters so that sender and receiver formats need not match.

Related: HL7 v2 messaging, Extract-transform-load, Data lineage
Application programming interface

A defined contract letting one software system request data or actions from another, specifying endpoints, authentication, request parameters, and response formats.

Related: HL7 FHIR, US Core, Consent management
US Core

A set of required FHIR profiles and elements that certified health IT must support, defining which patient, encounter, and result data must be retrievable through standard queries.

Related: HL7 FHIR, Application programming interface, Value set
Randomized controlled trial

A prospective study in which participants are assigned to interventions by chance, so that known and unknown confounders are expected to balance between groups.

Related: Confounding, Intention-to-treat analysis, Selection bias
Confounding

Distortion of an exposure-outcome association by a third factor that influences both, creating a spurious link or masking a real one unless adjusted in design or analysis.

Related: Randomized controlled trial, Propensity score, Effect modification
Selection bias

Systematic error arising when the participants studied differ from the target population in ways related to both exposure and outcome, often through enrollment or attrition patterns.

Related: Confounding, Loss to follow-up, Randomized controlled trial
Intention-to-treat analysis

An analysis that keeps every participant in the group to which they were assigned regardless of adherence, preserving the benefits of randomization and estimating real-world effect.

Related: Randomized controlled trial, Loss to follow-up, Per-protocol analysis
Propensity score

The estimated probability of receiving a treatment given measured covariates, used to match or weight participants so treated and untreated groups resemble a randomized comparison.

Related: Confounding, Observational cohort study, Effect modification
Observational cohort study

A design that follows a defined group over time, measuring exposures and outcomes as they occur naturally, allowing incidence estimation without assigning treatment.

Related: Propensity score, Confounding, Loss to follow-up
Effect modification

A situation in which the magnitude of an exposure's effect on outcome differs across levels of a third variable, so results are best reported stratified rather than pooled.

Related: Confounding, Interaction term, Subgroup analysis
Statistical power

The probability that a study will detect a true effect of a specified size, driven by sample size, effect magnitude, variability, and the chosen significance threshold.

Related: Type II error, Confidence interval, Sample size calculation
Confidence interval

A range of values around an estimate that would contain the true parameter in a stated proportion of repeated samples, conveying both magnitude and precision.

Related: Statistical power, P-value, Standard error
P-value

The probability of observing data at least as extreme as those obtained if the null hypothesis were true; it is not the probability that the null is correct.

Related: Confidence interval, Type I error, Multiple comparisons
Type I error

Rejecting a true null hypothesis, falsely concluding an effect exists; its rate is controlled by the significance threshold chosen before analysis.

Related: P-value, Multiple comparisons, Type II error
Multiple comparisons

The inflation of false-positive risk when many hypotheses are tested on one dataset, addressed by prespecified primary outcomes or adjustment procedures.

Related: Type I error, P-value, Subgroup analysis
Minimum necessary standard

The HIPAA requirement to limit uses, disclosures, and requests of protected health information to the smallest amount reasonably needed to accomplish the intended purpose.

Related: Protected health information, De-identification, Role-based access control
Protected health information

Individually identifiable health information held or transmitted by a covered entity or business associate in any form, subject to HIPAA privacy and security rules.

Related: Minimum necessary standard, De-identification, Business associate agreement
De-identification

The process of removing identifiers so data no longer relate to an individual, achieved under HIPAA by stripping the 18 specified identifiers or by a certified statistical expert.

Related: Protected health information, Limited data set, Re-identification risk
Limited data set

Protected health information stripped of direct identifiers but retaining dates and geographic detail, usable for research, public health, or operations under a data use agreement.

Related: De-identification, Protected health information, Business associate agreement
Business associate agreement

A contract required before a vendor handling protected health information on behalf of a covered entity may do so, specifying permitted uses, safeguards, and breach duties.

Related: Protected health information, Limited data set, Role-based access control
Role-based access control

An access model granting permissions according to job function rather than individual identity, so staff receive only the data rights their role requires.

Related: Minimum necessary standard, Audit trail, Consent management
Data quality dimensions

The measurable attributes used to judge fitness for use, commonly completeness, accuracy, timeliness, consistency, validity, and uniqueness of recorded values.

Related: Data validation, Data profiling, Data stewardship
Data validation

Checks applied during or after entry that test values against rules such as range limits, required fields, and cross-field logic, rejecting or flagging implausible records.

Related: Data quality dimensions, Query management, Audit trail
Audit trail

A secure, time-stamped record of who created, changed, or viewed a data value, capturing the prior and new content so changes can be reconstructed.

Related: Data lineage, Data validation, Role-based access control
Data profiling

Systematic examination of a dataset's structure and content, measuring distributions, missingness, duplicates, and pattern violations to reveal quality problems before use.

Related: Data quality dimensions, Entity resolution, Data validation
Entity resolution

The process of deciding whether records from different sources describe the same real-world person or object, using deterministic rules or probabilistic matching.

Related: Master patient index, Data profiling, Data quality dimensions
Query management

The tracked workflow by which reviewers raise questions about discrepant or missing study data, and sites respond and correct the source, with resolution documented.

Related: Data validation, Electronic data capture, Source document
Data visualization

The graphical encoding of data values as position, length, color, or shape so that patterns, outliers, and comparisons can be perceived quickly and accurately.

Related: Small multiples, Dashboard design, Effect modification
Small multiples

A layout repeating the same chart design across panels, one per category or time period, enabling comparison without the clutter of many lines in one frame.

Related: Data visualization, Dashboard design, Subgroup analysis
Dashboard design

The practice of arranging key indicators, context, and thresholds on a single display so viewers can judge status at a glance and drill into detail when needed.

Related: Data visualization, Small multiples, Clinical quality measure
Clinical quality measure

A quantified statement of healthcare performance with a defined denominator, numerator, exclusions, and data sources, used to compare or reward care delivery.

Related: Value set, Dashboard design, Case mix
Case mix

The profile of patients treated by a provider, described by diagnoses, severity, and resource use, used to adjust comparisons so facilities serving sicker patients are not penalized.

Related: ICD-10-CM, Clinical quality measure, Confounding