What We Do
Services
Data Management
Data extraction is the critical first-step when replacing one EHR with another.
Data extraction is the critical first-step in the process of replacing one EHR with another. An accurate, discrete EMR data extraction sets the stage for successful subsequent processes (like transformation and load) that will result in a medical data conversion or patient data migration that IT and data governance teams can be proud of.
Harmony Healthcare IT, an industry leader in data management solutions, has been ranked as the top data extraction and migration healthcare IT company according to Black Book™ Rankings, a division of Black Book™ Market Research. For over a decade, we have specialized in EMR extraction from both acute and ambulatory source systems of all kinds – spanning different software brands, database types and areas of healthcare – across the US and Canada.
We have extracted or migrated patient, employee or business records from over 550 EHR and ERP brand names. That includes the major leading EHR brands you would expect like EPIC, Allscripts, Cerner, CPSI/Evident, eClinicalWorks, GE, Greenway, Healthland, McKesson, Medhost, MEDITECH, NextGen and Practice Partner.
It also includes the “rest” of the EMR brands with less market share as well as what most would consider the one-off or even home-grown EMRs with even lower install rates.
If it’s software in a healthcare setting, chances are we’re familiar with it or can quickly become familiar.
Over the years, clinical systems have been developed in a vast array of database types. So, it follows that our healthcare and patient data extraction experience spans a variety of database schemas – some of which are more complex and drive the level of difficulty of the EMR data extraction (i.e., Cache versus MSSQL). Here are some databases we have worked with – whether through front-end or back-end access (note: we’ve even been able to strip data out of reports when database access wasn’t available):
* with data driver overlay
So, we work with a lot of different operating systems that are out there. We worked with pretty much anything that’s UNIX based. So, Linux, SCO UNIX, HPUX AIX, all those flavors and things, different platforms of Windows all the way back to 2003, all the way through current. So, we’ve worked with a pretty broad range and many different platforms and operating systems. I would say they’re all difficult to work with. So, no system is, well, one size fits all kind of thing. You start working on a system and you figure out how to work with them. They all have their nuances and challenges and those kinds of things after you do it the first time you kind of get in a rhythm of how they might have designed or, you know, architected that product where you kind of get your brain wrapped around. Oh, I’ll look over here, look over here kind of thing. But I would say that no one is more complex than the other. They’re just different. Usually, what we’ll do is try and break them down into smaller pieces. So, if we’ve got a particular element in a project that’s really difficult, we’ll try and break that down to figure out if in the extraction process is a challenge with how that database was formed, how it was originally designed the elements of it, those kinds of things that we can basically say, hey, you know, we probably should start over here rather than over here or work our way through a particular object, set those kinds of things as we’re moving through that process.
We came across a problem statement in the extraction process. NexGen, the vendor and their product have a hard coded dental module which is the dentagrams or images of the jaw line, the teeth and then the markings that the providers could be marking if they find anomalies or issues with the teeth or cuplines. Harmony was able to engage their Engineering Team and replicate the images that were hard coded in. Next, the vendors themselves could not extract the data for us. So, that was a huge win for our providers to have a seamless transition and those dentagrams or images available within Harmony due to the success of their engineers that redesigned successfully and had the data available.
While many EMR extraction vendors might specialize in pulling data only from acute or ambulatory systems (not both), we extract data from systems spanning the continuum of healthcare. Large enterprise health systems are our mainstay. We excel at inventorying, prioritizing, extracting, migrating and archiving data from complex legacy system portfolios. That includes clinical, financial, business and administrative data from inpatient and outpatient systems. We perform ERP and EMR data extraction for health systems, hospitals, and medical practices of any size. We also manage data for labs, blood banks, perinatal units, community health centers, long-term care centers, retail health clinics, pharmacies, dental practices, correctional facilities, student health centers, payer organizations and more.
Data extraction simply means retrieving data from various sources. It is the critical first step when replacing one EMR with another. An accurate, discrete data extraction sets the stage for successful subsequent processes (like transformation and load) that will result in a data migration that IT and data governance teams can be proud of. That said, often there is a large amount of unstructured data housed in an EMR. There needs to be a plan for this data as well as the normalized, structured data elements.
The benefits of data extraction include being able to access the data and move it wherever it needs to be, whether that be a new EHR, an active archive, cold storage, etc.
Healthcare data provides the clinical, employee and business records that are needed to track and support longitudinal care and administrative operations for healthcare providers. As such, sometimes the data needs to be moved to a new location. Data extraction is simply the process of pulling data from an EMR or EHR, usually so it can be migrated and retained.
As the extraction vendor will have access to the organization’s critical IT systems, it is vital to have a partner with demonstrated extraction experience with the associated source systems as well as the depth of experience and to work closely with the customer’s technical or source system matter experts. Since the vendor will be handling sensitive data, such as PHI or employee records, it is imperative the vendor is vetted for adhering to cybersecurity best practices.
Extraction can be done directly by an IT vendor, or if a third-party or application vendor must be used, there should be a discovery call to discuss the scope and delivery capabilities of the legacy vendor. There are different methods of healthcare data extraction, depending on the type of data and source system. One common method is to use an application programming interface (API) to extract data from EHRs. APIs allow developers to access EHR data in a structured format that is suitable for analysis. Another method is natural language processing (NLP) which is used to extract unstructured medical records. NLP algorithms can identify and extract relevant information from the records, such as diagnoses, medications and procedures.
Watch a short, simple video explanation about extraction and archiving.
EMR data extraction from CPSI/Evident to Cerner
This episode focuses on data procurement and extraction, focusing on the challenges of handling legacy data in healthcare, particularly for large organizations like Integrated Delivery Networks (IDNs).