As I mention in my Q&A post “What Are the Three V’s of Big Data”, Big Data is a significant topic of conversation in Health Information Technology (HIT) because Healthcare Organizations (HCOs) need to determine how Big Data figures into their operations. The previous post cites a definition of Big Data that states, essentially, that Big Data results from a collection of datasets that can no longer be managed or processed using traditional tools and techniques.

Many healthcare organizations are approaching this seeming Big Data “barrier”. According to HIMSS, three key drivers that are triggering HCOs to seriously look at Big Data architecture and analytics are:

  • Healthcare is entering into a “post-EMR deployment” phase where HCOs are now interested in leveraging the vast amount of EMR data being collected to obtain operational and clinical insights that drive institutional changes.
  • HCOs are under immense pressure to reduce costs and improve quality of care by “applying advanced analytics to both internally and externally generated data”.
  • Larger volumes of structured and unstructured data can now be managed and analyzed through “faster, more efficient and cheaper computing (processors, storage, and advanced software) and through pervasive computing (telecomputing, mobile devices and sensors)”.

There are many kinds of data that are now available and accessible to healthcare organizations for advanced analytics. Some of these sources were outlined in a document I helped author for the Institute for Healthcare Technology Transformation:

  • Human-generated data: Although most database systems and analytical tools can handle structured data from source-systems, the challenge for these systems (and for analysis) stems from both the unstructured and semi-structured data increasingly common in electronic medical records (EMRs), physicians’ notes, email, and paper documents. Text mining techniques are one way that data scientists are extracting information from unstructured and semi-structured data so it can be aligned to standard terminologies and coded in ways that can be analyzed with traditional data-mining techniques.
  • Web and social media data: Social media platforms such as Facebook, Twitter, Linkedin, blogs, and other emerging media is generating clickstream and interaction data that can be tremendously useful for a variety of purposes ranging from disease surveillance to gauging patient satisfaction and sentiment.
  • Machine-to-machine data: The increasing number of connected devices in healthcare facilities is generating terrific volumes of data that includes readings from sensors, meters, and other devices. For example, Radiofrequency Identification (RFID) technology makes it possible to track with high accuracy patient, staff, and equipment locations within a facility but generates a high volume of data that accumulates rapidly. When analyzed appropriately, however, this information is of immense value for examining processes and movements throughout a facility for the purpose of streamlining operations.
  • Biometric data: Biometric data includes fingerprints, genetics (i.e., genomics), blood pressure (and other vital signs), medical images, retinal scans, and similar types of data. For example, an individual’s genome can be analyzed to predict how certain medications will be metabolized by a patient to aid physicians in prescribing medications that will have the greatest chance of achieving desired treatment goals with minimum side effects.

As evidenced by the above, the growth of Big Data in healthcare isn’t necessarily driven by one single type or source of data, but rather a combination of sources each made possible through its own series of innovations in information or biomedical technology. It is clear, also, that no one single approach to manage, analyze, and utilize this information will be sufficient to obtain maximum value from these data sources and achieve desired healthcare quality and performance improvements.

Join the conversation! Am I missing any key drivers? Are there any other important sources of data that I’ve missed? Please feel free to leave your comments below. 



The following article is a brief summary of an article that I wrote for SearchHealthIT.

The National Institute for Standards and Technology defines cloud computing as “a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction.”

Because it is still an emerging technology, healthcare organizations (HCOs) are still in the early stages of determining how the cloud fits into the health information management and technology ecosystem. Providers are still calculating how to balance the possible benefits of cloud computing in healthcare with the obvious security, technical and legal risks.

According to a Cloud Computing in Health by Canada Health Infoway, cloud computing models support three ways of provisioning computing resources as services:

  • Software
  • Platforms
  • Infrastructure

The potential of cloud computing in healthcare is to enable providers to better meet changing technology, regulatory, and market demands. But healthcare organizations understand that the use of the cloud is not without risk, and this is perhaps one of the most significant barriers to cloud adoption by healthcare organizations. Some identified risks include:

  • Data breaches
  • Data loss
  • Account hijacking
  • Denial of service
  • Malicious insiders
  • Insufficient due diligence

There are numerous opportunities for significant financial, technological and service-related benefits associated with cloud computing. Yet, as with most emerging technologies, there are risks (both known and unknown) that must be mitigated to realize the potential benefits and, most importantly, to ensure the security and privacy of any data stored in the cloud.

Healthcare executives must balance the risks, benefits, and business and IT needs of the organization to best determine if, how and where cloud computing should be featured in their health IT provisioning strategy.

Please click here to read the entire article on


What You Need to Know About Healthcare Analytics on the Cloud

by Trevor Strome January 22, 2015

This is a summary of a recent article I wrote for  The promise of cloud-based analytics Because of the computing power, data storage, and network bandwidth that some modern analytics and simulations require, the healthcare industry requires cost-effective and scalable solutions. Cloud-based services offer a promising option. Many of the reporting and analytics tools used [...]

Read the full article →

Healthcare Analytics Q&A: What Are the “Three V’s” of Big Data?

by Trevor Strome January 20, 2015

Healthcare Analytics “Q&A” Series Every day, I receive emails from readers of asking questions about healthcare analytics topics, trends, and issues. These questions arise out of posts they’ve read on this website or encountered elsewhere. Although I endeavour to answer each question individually, I’ve created this “Question & Answer” series on to provide [...]

Read the full article →

Upcoming NIHI Seminar: Advanced Healthcare Performance Measurement

by Trevor Strome January 15, 2015

 Join me on January 29, 2015 for my National Institutes of Health Informatics (NIHI) web seminar entitled “Advanced Healthcare Performance Measurement“. Click here to register. About the Session The complexity of health care demands that a robust approach to measuring quality be followed. Healthcare quality and performance must be defined in terms that are quantifiable—meaning they [...]

Read the full article →

Six Ways to Improve the Usefulness of Data Visualizations in Healthcare

by Trevor Strome January 13, 2015

  Improve decision-making with data visualization Data visualization is an important way to communicate information quickly and effectively. In fact, data visualization is one of the key principles that contribute to effective decision-making in healthcare organizations. In recognition of this, many business intelligence (BI) and data analysis software packages provide sophisticated data visualization capabilities. These [...]

Read the full article →

Strategic Analytics Framework for Leveraging Data and Health Information Systems

by Trevor Strome January 5, 2015

Healthcare analytics and IT are progressing rapidly Healthcare analytics is a rapidly advancing field. Healthcare organizations (HCOs) along with the analysts, data scientists, and Information Technology (IT) professionals who work in them are faced with myriad choices regarding how best to meet the information and decision-making needs of administrators and, most importantly, clinical staff working [...]

Read the full article →

Keeping Security and Privacy Top-Of-Mind when Using Healthcare Data

by Trevor Strome November 17, 2014

Healthcare organizations are always working to improve the quality of their care and the efficiency of their business operations. Data analytics for these clinical quality improvement efforts require access to data for determining process baseline performance, detecting trends and patterns in quality based on key indicators, and simulating potential outcomes of new processes and workflows. [...]

Read the full article →

Healthcare Data Analytics Summit – Toronto, December 2014

by Trevor Strome September 29, 2014

The Third National Summit on Healthcare Data Analytics is coming to Toronto, Ontario December 2-3, 2014. I’m honored to be chairing this outstanding event this year, as there is an outstanding lineup of speakers. There are expert speakers attending from across North America and internationally. Several of the timely and important topics include: population health [...]

Read the full article →

The HIPAA Basics that Healthcare Analytics Professionals and Data Scientists Must Know

by Trevor Strome September 26, 2014

Healthcare analytics professionals and data scientists access and use healthcare data on a near-constant basis. Whether it is for designing dashboards or building predictive models, the use of sensitive information is a necessity in efforts to leverage analytics to improve healthcare. Also essential, however, is keeping the private health information that we access and use [...]

Read the full article →