Why discrete data is critical for the future of healthcare | Viewpoint

Healthcare providers store, access, and utilize a massive amount of medical and patient information on a daily basis.

Many hospital systems rely on outdated data archives. Discrete data sets are preferable because they’re detailed, measurable, and reportable. (Image credit: ©ipopba - stock.adobe.com)

As of 2021, the Office of the National Coordinator for Health Information Technology (ONC) reported 78% of hospitals had adopted a certified EHR. A more recent examination of healthcare’s data usage by the World Economic Forum revealed that hospitals produce 50 petabytes of data per year – comprised of clinical notes, lab tests, medical images, sensor readings, genomics, and operational and financial data.

Even with this colossal amount of data in their possession, many hospital systems rely on outdated and unwieldy non-discrete data archives. Discrete data sets are preferable because they’re detailed, measurable, and reportable. It’s all the same data, only more accessible.

The downside of continuing with non-discrete data archives

Before diving into why discrete data is a way forward for the future of healthcare, let’s first explore why non-discrete data isn’t.

Non-discrete data includes file types such as PDFs and images. The data inside these documents can’t be extracted or utilized for reporting with current technologies that are accessible and reliable. Specific data sets and data points can be accessed only by reading the document itself. While new technologies like artificial intelligence are starting to learn how to “read” document data, the templates and learning that these solutions require are still very new and out of reach for most.

Non-discrete data archiving can be less expensive than other options, but its affordability comes with a host of limitations. Because data analytics and reporting aren’t possible with non-discrete data sets, any specific values from documents like archived lab reports will be buried deep in a PDF. As a result, clinicians who need to access information must scan dozens of pages to find the relevant values.

Archived non-discrete data sets don’t support the ability to pull up only the specific data needed to fulfill release of information requests. Instead, entire reports and documents must be released, making this archiving approach suboptimal for legal purposes. Validation and implementation processes are also more labor-intensive and time-consuming, resulting in longer timelines.

Discrete data, on the other hand, is collected and stored in a detailed database table. Because this type of data is both measurable and reportable, its potential to improve data usage methods in the healthcare industry is significant.

Accessing discrete data is critical and efficient

In a healthcare IT context, discrete data is used to access critical patient care and treatment details. For example, a medication data point may consist of dosage instructions for a specific medication, while a full data set reports all instances of when that specific medication was prescribed during a given timeframe. Discrete data archives store the data in individual fields that can be queried for particular pieces of information or compiled into a full report with specified values.

Discrete data sets are also useful for gathering details about medical product recalls. Discrete data reports can help reveal how many patients were prescribed a recently recalled medication or how many patients received a knee replacement and later had complications related to the procedure.

For legal inquiries, discrete data archives allow healthcare providers to satisfy information requests without widening the lens. Suppose there’s an ongoing court case involving a certain physician, patient visit, or medical procedure. With a discrete data archive, the hospital can release only the details about that specific topic instead of entire documents that might contain additional information that might potentially violate the Health Insurance Portability and Accountability Act.

What makes a discrete data set active?

Put simply, discrete data sets are active because they’re capable of researching analytics, compiling data trends, and releasing specific data points instead of entire patient records. When a data set is active, it allows people who access it to conduct data analyses like comparing and contrasting values. Healthcare data analytics naturally benefits from being able to compare real lab results.

If practitioners wanted to track down cases where a specific diabetes medication was prescribed and learn how it correlated with weight loss, discrete data sets would allow them to do that. Active datasets are also advantageous because they can be edited. If patients are erroneously listed as smokers in their intake files, staff can later access and amend the records. The time, source, and reason for edits are also tracked, ensuring that only necessary changes to medical records occur.

In contrast, when a data set is non-discrete, it’s also non-active. Once the information is entered, it can no longer be interacted with in any meaningful way. There is also no reliable method for converting non-discrete data into active data. There are applications that add search functions, but they may not be effective or compatible with all data sets.

A discrete data archiving solution improves provider burnout, reduces ongoing costs

Discrete data sets can be more expensive to create and store than the alternative, but they possess multiple advantages for provider organizations.

• The ability to conduct queries is improved, and medical data is easily extracted. As a result, provider and clinician satisfaction increases, which is critical at a time when the healthcare industry is seeing 53% of physicians feeling burned out, according to a MedScape survey of more than 9,100 U.S. physicians across 29 specialties.
• Unlike non-discrete data archives, discrete data archives do not carry hidden financial and temporal costs. For example, assume that a healthcare provider’s outpatient population includes someone who’s made frequent hospital visits over an extended period of time. In a non-discrete data archive, this person’s entire medical history will appear as a single, massive, unsearchable PDF document. Accessing just one document and pouring over it to locate any relevant data will be incredibly frustrating and time-consuming. Deciding to archive it this way in hopes that a later technology will provide the needed discrete value is risky at best. It’s also costly because providers can’t use the discrete data while the technology is maturing.
  
• Although the upfront cost for discrete data archives can be higher, they’re likely to end up providing more value in the long run. For example, what documents currently exist in a provider’s legacy system? Are they adequate for archiving? Are they fit for fulfilling ROI requests? Do they have all the information collected, or at least all the essential data? Do they provide the data needed to be compliant with current reporting regulations, such as the 21st Century Cures Act? If not, how will the provider generate these documents?

Utilizing a discrete data archival system from the get-go is simply the most efficient and effective strategy for hospitals and health systems looking to save time and money by eliminating maintenance costs and licensing fees, as well as reducing provider burnout by minimizing the time required to work across redundant systems.

Dr. Shelly Disser is vice president, innovation, and collaboration at MediQuant, a healthcare data management company.