Clinical Guides: Data Literacy: Data Concepts

Attribute

Definition: A characteristic or property of an entity.
Purpose: Describes an entity's specific data points.
Example: "DateOfBirth" of a patient.

Big Data

Definition: Extremely large and complex datasets.
Purpose: Allows analysis of massive amounts of varied data.
Example: A national health system’s data on millions of patients.

Data

Definition: Raw facts or figures without context.
Purpose: Basis for analysis or decision-making.
Example: A list of patient heart rates.

Data Analytics

Definition: The analysis of data to draw conclusions.
Purpose: Helps inform decisions or find trends.
Example: Analysing patient admission data to predict hospital staffing needs.

Data Dictionary

Definition: A centralised reference of all data elements in a system, with detailed descriptions of their attributes.
Purpose: Ensures consistent data definitions and proper usage.
Example: A hospital's data dictionary describing each field in the patient table, such as "PatientID: Integer, 10 digits."

Data Governance

Definition* A set of rules for managing data.
Purpose: Ensures data quality, security, and compliance.
Example: Policies for handling patient data to comply with privacy laws.

Data Integrity

Definition: Accuracy and consistency of data over its lifecycle.
Purpose: Ensures reliable and valid data.
Example: Validating that "PatientID" is unique and not duplicated.

Data Mart

Definition: A focused subset of a data warehouse.
Purpose: Provides a specific team or department with relevant data.
Example: A cardiology data mart containing heart-related patient data.

Data Mining

Definition: The process of identifying patterns in large datasets.
Purpose: Extracts useful insights for decision-making.
Example: Analysing patient data to identify risk factors for diseases.

Data Model

Definition: A conceptual representation of how data is structured, stored, and related.
Purpose: Provides a blueprint for designing databases or systems.
Example: A healthcare data model showing relationships between patients, doctors, and appointments.

Data Warehousing

Definition: Centralised storage of data from various sources.
Purpose: Enables complex analysis and reporting.
Example: A hospital’s data warehouse storing data from labs, radiology, and billing systems.

Database

Definition: An organised collection of structured data.
Purpose: Stores data for efficient retrieval and management.
Example: Electronic health record (EHR) system.

Entity

Definition: A real-world object or concept represented in data.
Purpose: Organises data by grouping related attributes.
Example: "Patient" entity in a hospital database.

ETL (Extract, Transform, Load)

Definition: The process of extracting data, transforming it, and loading it into a data warehouse.
Purpose: Prepares and integrates data for analysis.
Example: Extracting patient data from multiple systems, cleaning it, and loading it into a hospital’s data warehouse.

Index

Definition: A structure that improves data retrieval speed.
Purpose: Makes searching data faster.
Example: An index on "AppointmentDate" for quick access.

Metadata

Definition: Data that describes other data.
Purpose: Provides context or structure for data interpretation.
Example: Metadata for a "Patient" table includes field descriptions like "Name: String, Length: 50 characters."

Query

Definition: A request to retrieve or manipulate data in a database.
Purpose: Allows users to extract specific information.
Example: Query to find all patients with upcoming appointments.

Structured Data

Definition: Data organised in a defined format, such as tables.
Purpose: Easy to search and analyse.
Example: Patient names, ages, and diagnoses in a database.

Unstructured Data

Definition: Data without a predefined format.
Purpose: Contains rich information, but harder to process.
Example: Doctors' notes, emails, or medical images.

Healthcare systems generate vast amounts of data from various sources, which can be categorised as:

Clinical Data

This includes medical information captured during patient care, such as diagnoses, treatments, medications, laboratory results, and imaging reports.
Clinical data is typically stored in Electronic Medical Records (EMRs) and is essential for direct patient care, clinical decision-making, and research.
Examples: Blood pressure readings, prescription histories, clinical notes, and lab test results.

Administrative Data

Administrative data refers to non-clinical information collected for operational purposes, such as billing, insurance claims, appointment scheduling, and resource management.
This data is crucial for health service administration, cost management, and healthcare policy planning.
Examples: Insurance information, billing codes, staff rosters, hospital admission and discharge data.

Patient-Generated Data

This type of data is collected directly from patients outside of traditional healthcare settings, often via mobile apps, wearables, or online health platforms.
It can include health diaries, fitness tracker information, symptom reports, or self-monitoring data (e.g., glucose levels in diabetics).
Patient-generated data supports preventive care, personalized medicine, and patient engagement.

Electronic Medical Records (EMRs)

EMRs are digital versions of patient records maintained by healthcare providers.
EMR systems facilitate the secure exchange of health information and enable data-driven decision-making, improved patient outcomes, and enhanced coordination of care.
They contain a broad range of data, including clinical notes, imaging, lab results, medication lists, and past medical history.

Other Data Sources

Genomic Data: Collected from DNA sequencing, this data can provide insights into genetic predispositions to diseases.
Social Determinants of Health: Factors like socioeconomic status, education level, and access to healthcare services also influence health outcomes and are increasingly considered in health data analyses. These could be captured from the ABS census.

Data standards ensure interoperability, consistency, and accuracy in healthcare information exchange, allowing for seamless communication between different systems and healthcare providers. Key healthcare data standards include:

SNOMED CT (Systematized Nomenclature of Medicine - Clinical Terms)

SNOMED CT is a comprehensive, multilingual healthcare terminology system designed to capture detailed clinical data.
It facilitates consistent recording and sharing of clinical information and supports analytics, research, and decision support systems.
Used worldwide, including in Australia, SNOMED CT ensures uniformity in the description of medical terms, symptoms, procedures, and conditions.

HL7 (Health Level Seven)

HL7 is a set of international standards for the transfer of clinical and administrative data between healthcare information systems.
HL7 standards define protocols for electronic messaging, allowing systems from different vendors to communicate with each other.
HL7's FHIR (Fast Healthcare Interoperability Resources) standard is particularly important for modern healthcare apps and mobile health solutions as it provides lightweight, web-based data exchange.

ICD-10 (International Classification of Diseases)

ICD-10 is an international coding standard developed by the World Health Organization (WHO) for classifying diseases, conditions, and external causes of injury.
It is widely used for reporting morbidity and mortality statistics and in billing systems for healthcare services.
Australia uses a modified version of ICD-10 called ICD-10-AM (Australian Modification) for coding diseases and procedures.

FHIR (Fast Healthcare Interoperability Resources)

FHIR, developed by HL7, is a newer standard designed to simplify data exchange between healthcare systems.
It uses modern web technologies (e.g., RESTful APIs, JSON, and XML) to allow easy and flexible data sharing.
FHIR is particularly suited to mobile health applications, cloud-based systems, and EHRs, enabling secure, real-time data exchange.