22 health data science terms to know

The term "data analytics" is a major industry buzzword, but what does it really mean?

Analytics solutions can enhance a health system's data protection, reveal trends that inform patient care and lead to more effective patient outcomes.

Here are 22 terms related to health data science that healthcare executives should know, compiled with the help of Technopedia.

Health informatics refers to the study and adoption of IT and other methods that helps manage and deliver health information.

Big data describes a large amount of data that organizations can analyze to identify trends,  patterns and other insights. Big data can be collected from a number of players — payers, providers, supply chain or patients — depending on what knowledge an organization is trying to gain. There are three key dimensions of big data are volume (how much data there is), velocity (the data processing speed) and variety (the number of different types of data).  

To make sense of big data, organizations partake in what's called data mining, which refers to the process of sorting data sets to find patterns or relationships. There are six key data mining queries that help identify the most important relationships in data, including:

  • Support: Notes how frequently the items appear in the database
  • Confidence: Notes the number of times "if/then" relationships occur
  • Sequence or path analysis: Identifies patterns where one event leads to another
  • Classification: Identifies new patterns, which may ultimately change the way the data is organized
  • Clustering: Groups data sets and aggregates data points based on how similar each element is to another to find and visually document previously unknown groupings
  • Forecasting: Discovers patterns in data that can lead to reasonable predictions about the future

To put big data to work, a variety of approaches can be taken. These include the following.

  • The idea of behavioral analytics has recently grown in popularity. In healthcare, behavioral analytics often answers the question "Will patients be engaged?" It can also be turned back on hospitals for insight into how providers can improve their workflow and patient relationship.
  • Comparative analytics helps organizations better understand how their volumes, use rates and payments shift and compare to their peers. It also helps them deliver on the pay-for-performance equation with utilization and outcome measurements, as well as peer benchmarking.
  • Descriptive analytics uses data aggregation and mining to answer "What happened?" These tools are often used to gain insight into the past.
  • Diagnostic analytics uses data discovery, or mining, and correlations to answer "Why did it happen?" These approaches are often used to examine data or content.
  • Predictive analytics uses statistical models and forecasting techniques to answer "What could happen?" These tools are often used to understand future events or predict possible patient outcomes.
  • Prescriptive analytics uses optimization and simulation algorithms to answer "What should we do?" These approaches are often used to advise action based on potential outcomes.

An algorithm is a set of simple or complex step-by-step instructions a computer or software system uses to complete a specific task, such as to solve a problem, answer a question or deliver an alert.

Data cleansing refers to the process whereby data is reviewed for inaccuracies or inconsistencies. This involves detecting corrupt records that are incomplete, incorrect, inaccurate or irrelevant from a database and then replacing, modifying or deleting the faulty data.

Scalability describes a computer application, process or network's ability to function effectively when its size or volume changes. While the rescaling could refer to either growing or shrinking, it is more common to scale upward. A scalable business is able to handle an influx of demand, increased productivity, trends and changing needs.

A data center houses physical computing equipment — including servers, routers, switches and firewalls — in addition to supporting tools — such as backup equipment, fire suppression facilities and air conditioning.

While the terms cloud and data center are often used interchangeably, there are distinct differences between the two. Cloud refers to an off-site method of data storage. By using multiple data centers and the internet, the cloud simulates virtual data storage and can be accessed and shared across a network. Cloud services are typically outsourced to third-party providers who manage updates and other maintenance of the virtual data environments. Cloud vendors usually own multiple data centers in a couple different locations for cloud-hosting to ensure availability during outages or other failures.

An enterprise data warehouse stores all of a business' information and makes it accessible throughout the organization. It often includes the following: a unified approach for organizing and representing data; the ability to classify data according to subject and give access according to those divisions; a normalized design; a robust infrastructure with contingency plans to allow for business continuance, accessibility and a high level of security; scalability.

A clinical data registry documents information about the health of patients and the care they receive over a period of time. These registries typically focus on patients who share the same care needs, and they also offer healthcare providers insights on effective treatments.

More articles on data analytics & precision medicine:

Penn State launches big data project to drive population health, personalized medicine research
Researchers use predictive analytics, EHRs to forecast hypertension onset
Cincinnati Children's Hospital Medical Center launches Maternal and Infant Data Hub

Copyright © 2024 Becker's Healthcare. All Rights Reserved. Privacy Policy. Cookie Policy. Linking and Reprinting Policy.

 

Featured Whitepapers

Featured Webinars