Did you know that 30% of the world’s data volume comes from the healthcare industry?
The scale of the data that healthcare businesses work with, and the complex data formats used, are just two of the many reasons why you need modern data tools to effectively solve those complex data challenges.
Thankfully, because CloverDX is designed to solve 100% of your data issues (not just the easy 90%), it’s a good match for healthcare companies and their tricky data needs.
In this blog, we’ll unwrap four use cases where CloverDX met the needs of a company in this sector.
‘Be lenient in the data formats you accept. Be strict in the data formats you produce.’ - Postel’s robustness principle.
In this first example, our client works with large-scale healthcare data to make it actionable.
Their overall process to work with data is straightforward enough. They take HL7 messages, then transform it into a standard JSON format, and then load it into their analytic warehouse. However, variability in formats and versions, and the fact that mapping requires input from domain experts (who don’t necessarily have developer skills) meant they needed support.
To help them resolve this, we made some key changes to make it simpler for the domain specialists to provide input into the mapping.
Watch a demo of the process in CloverDX:
First, we separated the mapping and transformation rules from the execution data pipeline itself. We then provided straightforward, easy-to-understand representations for expressing these rules using a format the domain experts were comfortable with (Excel in this example).
Here's a diagram to show what the mapping looked like:
Once we have the mapping documents, we use these in the data pipeline. So, the HL7 data comes in, and then it’s translated to a ‘raw JSON’ format. Then, the mapping runs - grabbing the appropriate Excel doc, performing the required mapping, and creating a new JSON file that’s ready for processing and pushing downstream.
This method not only enables domain experts to provide inputs using tools they’re comfortable with but is also extensible. The pipeline can take any message, convert from any flavor of HL7, apply a custom mapping, and create an output file. What‘s more, it can run on autopilot, detecting new HL7 messages as they arrive and running the mapping automatically, so the clean JSON can go into the analytic warehouse.
In the US, health insurance providers offer many different plans. Our client helps simplify the process of employers choosing which plans to offer their staff by collecting and analyzing insurance plan descriptions for them. This helps employers get more control over both their costs and compliance.
The challenge our client faced was trying to automatically process these plans. The plans came as spreadsheets not well-optimized for a computer to read. And the plans would frequently change, for example, new columns, new rows, etc.
You can see how complex these spreadsheets can get:
Here's what our five-step automated solution looked like:
Now, our client can automate the process of extracting plan descriptions, so they can make accurate, up-to-date information available to their clients.
Watch a more in-depth explanation of this use case:
In this instance, our client manages the benefits enrollment activity for employers. Specifically, this means managing the process of people filling out forms to change their coverage. They act as an intermediary between health care consumers and the insurance provider.
Automating the data pipeline for our client here meant establishing a series of data enrollment events:
At the same time, there were automated audit reports of everything that happened. This means the customer can now see which reports are accepted or rejected, as well as what changes were made.
Thanks to the power of automation, our client can now handle more business without having to scale their headcount.
Watch the video for a walkthrough of this solution for automating the management of benefits change requests:
In our fourth and final use case, we're looking at a client that provides integrated healthcare technology solutions to medical practices. This covers everything from practice management to managing EHR (Electronic Health Records) software. The challenge they brought on CloverDX for was migrating medical practices to a new EHR system.
Their business depends on the fast, economical and accurate onboarding of clients, moving them quickly and accurately from legacy EHR systems to our clients' platform.
To optimize these migrations, we helped template workflows for source EHR systems. This allowed existing staff to handle more data migrations whilst at the same time reducing errors. Of course, this also makes re-migrations faster if the initial migration exposes data issues that require a re-run.
In addition to these templated workflows, we created automatic audit reports. This improved customer confidence that the migration went well.
The final video clip describes how the architecture built with CloverDX worked to create faster, more efficient data migrations, and to enable the same process to be used for migrations from different systems.
As we’ve seen, healthcare tech and healthcare insurance companies face a myriad of data challenges. Most healthcare data processing is complex, having to go beyond standard data formats.
But good data pipeline design - helped by flexible, customizable data tools like CloverDX - can help simplify the base solution, the common parts of moving data from A to B, allowing you to focus on the healthcare-specific tasks.
If you’d like to hear more about any of the solutions above, or see if CloverDX could help build a solution for your healthcare data process, just get in touch here and we'll get right back to you.