Why Metadata Is the Real Secret to Organizing Healthcare Records
When you scan and archive a document, the ideal result is not just to have a photograph of the original instead of the original itself, but to have a photograph that you can search and organize just like any other type of electronic file. To do this, your scanning project needs metadata – and that metadata must be applied at the time of the scan.
Scanning isn’t the same as organizing
Many healthcare organizations have had to learn this lesson the hard way. They perform a backfile conversion project, have thousands of paper charts turned into image files and assume the most difficult part of the process is over. Then, someone asks for a patient’s lab results from 2016, and they realize that the information is buried in a file labeled “Scans\_Batch\_14” along with four hundred other unlabeled PDFs.
OCR helps with this problem in that it makes the text inside a scanned page searchable, so you can at least search for a keyword. But it doesn’t “know” whether a page is a discharge summary or a referral letter. It doesn’t know which patient it is from unless that text is extracted and made into a separate field. And if that step hasn’t been accomplished, you have a pile of searchable photos, not a set of usable records.
Why folders stop working
Organizations that are moving from traditional paper-based or legacy electronic document storage solutions to document management systems understand this challenge. They’re modernizing their records and asking how to migrate their existing mountains of documents without losing the ability to find anything.
The trick is that a human will have to formalize that metadata. If you’re converting a paper chart, someone will have to decide what the patient ID is and plug it in, because the computer doesn’t know. But as long as you’ve got someone doing that – for legacy documents, as they come in the door, or for documents automatically generated by your information system – a metadata-rich environment solves the problem of retrieval forever.
Old records were never built to move
Old paper charts are challenging because they came before any kind of digital order. There’s no standardized naming, no standardized ordering, and probably every clinic or provider that had contact with the file over the years has stored it slightly differently. One hospital intake form isn’t necessarily laid out how yours is. A chart from 2003 isn’t in the same order as one from 2019, even at the same place.
That’s why paper medical records from the past are harder to shift than you might think. It’s not the number of them that blows out a migration timeline – it’s the variance. Every chart needs a human or a computer to take a look, work out what it actually is, and assign it fields that a current system can understand. Skimp on this step at the scanning phase, and you’ll just have to return to it later, with more urgency and more expense.
What a properly indexed conversion looks like
The solution isn’t difficult to understand, although it may be challenging to implement. Metadata must be incorporated into the conversion process, rather than be added as an extra element. This involves the proper categorization of each document using specific fields such as patient identifiers, document type, provider, and date. This should be done during the scanning phase of the workflow.
The best way to achieve this is by using an online document management system that recognizes metadata as the primary means of organization. Instead of dumping files into an electronic drawer, this type of system allows you to retrieve records based on their attributes. This will help you immediately keep track of any changes since the software will recognize and monitor everything.
The compliance stakes are higher in healthcare
While sloppy metadata might fly in other industries for a while, it’s not the same for healthcare. With healthcare data, metadata has a very immediate and direct purpose. Storing and sharing data in healthcare have certain aspects that are very unique to the industry.
First, there’s security and privacy. Legally speaking, healthcare data is one of the most tightly regulated types of data across the world. When healthcare data is shared between organizations they need to share it in a manner that is compliant with laws that regulate patient privacy and data security. Metadata can be used to provide specifications about whether a patient’s data can be shared or not.
Then there are legal requirements. In many countries, there are laws that mandate that certain types of files or records related to a patient need to be maintained for a certain period of time for legal review if needed. Metadata can be used to define these records and indicate the period of retention.
And finally, easy and rapid discovery. When a file is created the organization generally knows the context or the category to which the file belongs. When another person is looking for that file they generally want to find files similar to what they are looking for. For instance, a doctor looking for a patient’s file would generally like to see all the information recorded about that patient. Metadata provides a context for the file and this helps the organization remember what other relevant data was created in relation to the file being searched.
The cost shows up later, not sooner
Organizations pay a heavy price for skipping metadata during original digitization efforts. Improper metadata does more than simply push a filing job off to the future; it can effectively cancel out the benefits of digitizing in the first place. When files can’t all be easily retrieved, a ream of paper and a terabyte can become pretty similar problems.
Not everything is created equal. Navigating boxes of paper is slow, and its footprint in office or surgical suites quickly becomes significant. Live views in an EHR or CIS are lightning-fast compared to a paper chart, but powerful database queries bring a computer to a crawl when the database isn’t designed with a clean structure. Especially when your digital document store sprouted not from any structured plan, but in great clumps over the years.
