Breaking down your data silos with data warehousing and data virtualization

I recently returned from an onsite visit with a client partner who is experiencing challenges managing its EHR data. They have a multitude of data assets but are struggling to turn these assets into insights. Sound familiar?  

The good news is, you’re not alone. Effective data management is one of the most common challenges healthcare organizations face today – with data silos, that is unstructured and isolated data, at the top of their list of concerns. Common examples of these data silos include having multiple physician groups on different EHRs or separate ERP and EHR data. As you know, when data is stored this way it can be difficult to analyze and put to good use.

While the meaningful application of data has improved significantly in recent years, organizations are still hampered by these data silos. We commonly find that organizations struggle to define a clear and unified data strategy – a succinct way to manage their data asset and break down these silos.

If you’re looking to improve your data strategy, you may want to consider implementing data warehousing, data virtualization, or a hybrid of these two models. Below I provide some insights into each of these models and main considerations for evaluating the model that’s best for your organization.Eric-Pennington-web

The benefits of an enterprise data warehouse

To support a unified data strategy, organizations are leveraging an enterprise data warehouse. Vendors like Epic and Health Catalyst offer data warehousing solutions, which help organizations improve cross-functional data integration by having a single landing place for data (and hopefully a single/primary source for analytics).

Data warehousing is currently the most popular method organizations choose to achieve a unified data strategy. And for good reason. It’s a vast improvement over the historical approach many organizations have applied by leveraging a separate platform for each functional area with little intention (let alone possibility) of governance or cross-team collaboration.

For example, we’ve helped client partners effectively implement a data warehousing strategy to integrate claims and clinical data. By taking an IT-driven approach to customizing their existing data warehousing platform, we helped them achieve better data governance between these two areas.

What to consider when evaluating data warehousing

It’s important to note that while it’s effective, the data warehousing implementation process can be quite costly and time intensive. Also, while vendors are making great strides in their ability to support extension and modification of their platforms, there is an inherent rigidity in schema of a data warehouse.

However, as I mentioned above, data warehousing can be, and is for many organizations, a viable and valuable data strategy.

Making the case for data virtualization

As an alternative option to data warehousing, or to augment data warehousing, organizations are increasingly opting for data virtualization. Data virtualization enables organizations to retrieve and manipulate data without requiring technical details about the data, such as how it is formatted or where it is physically located.

Data virtualization can afford the standards and governance of data warehousing while reducing some of the resources and time requirements of data processing, ETL development, and data modeling – enabling a more agile data development environment.

Key considerations of data virtualization

When implementing data virtualization, data governance should be the guiding component of your strategy. Otherwise, you could end up with multiple layers of data that are unrelated, and you’ll be right back where you started – data silos and all.

Also, unlike a date warehouse, which is part of your EHR, you won’t be able to build your data virtualization platform within your EHR. Therefore, it’s important to identify the correct tool for this build. Organizations have had success by using platforms such as Qlik or Informatica to support their data virtualization build-out. There are also additional costs associated with these data management platforms.

A hybrid approach to data warehousing and data virtualization

If you’re unsure of which strategy is best for your organization, there’s good news – you can do both. Organizations are experiencing success by taking a hybrid approach and using data virtualization to supplement their data warehousing strategy.

We recently partnered with an organization that is using this hybrid approach. They have the structure and governance inherent in data warehousing, along with the ability to integrate additional data sources through data virtualization. Data virtualization has also helped them eliminate the overhead costs associated with physical data modeling and ETL development of data warehousing.

If you’re looking to improve your data strategy, please let us know. We’d love to listen to your goals and help determine the best approach for managing your data assets and breaking down your data silos.

CONTACT US TO LEARN MORE LEARN MORE ABOUT DATA & ANALYTICS

Eric Pennington is a manager of Data & Analytics. He has been with Nordic since 2017, helping with the development and execution of Nordic's analytics solutions. Prior to Nordic, he worked for over two years at Qlik, where he led the Healthcare Solutions and Strategy practice and oversaw strategic partnerships in healthcare. He also has four years of experience working for Epic.

Topics: digital health

Module heading text

Get the highest quality chemistry and microbiology testing services aligned closely with current good manufacturing practices (CGMP) for all types of products across all phases of development.

Subscribe to receive blog updates