How to avoid dumpster diving for data in today’s legacy EHRs

If you listen to healthcare industry chatter, you’ll hear that we are in the “post-EHR” world in which everyone has electronic health records that can easily be shared between providers, thanks to the requirements of the 21st Century Cures Act.

We may be getting ahead of ourselves, however, because users still cannot find what they are looking for even in their own systems––and this situation will only get worse when data begins flowing in from other providers.

This is not a new revelation. In fact, I mentioned the challenge in a 2017 interview about the coming “data tsunami.” With increased data sharing between providers, however, the data tsunami is getting even worse.

Users have long struggled to find relevant clinical information within their legacy EHRs. Sure, clinicians have access to problem lists, medication lists, lab orders and results, physician and nursing notes and other documentation, but the details are stored in a variety of formats. Some data is recorded using standard terminologies and code sets such as ICD, CPT, LOINC, RxNorm, and SNOMED, ​​but much of the information is also captured in free-text notes.

Because we keep dumping so much data across EHRs, clinicians are forced to search section after section of a patient’s chart to find diagnostically relevant information for specific problems. Rather than making clinicians dumpster-dive for data, it’s time to give clinicians a better way to find the clinically relevant information they need to deliver quality, cost-effective care.

The challenge of finding data in legacy EHRs

One reason clinicians struggle to find what they need in EHRs is because most legacy systems were not designed to provide a diagnostically holistic view for specific clinical conditions. Let’s use renal failure as an example, which may be listed on a patient’s problem list. To assess the patient’s renal status and determine if the condition is being managed effectively, a clinician must navigate to the medication list, then to the laboratory results, and then to the encounter notes. With newer value-based care models that reward clinicians for cost effective care and quality outcomes, users need tools that make it easier to find the specific information they need for informed decision making.

EHRs provide a wealth of clinically rich data, but clinicians need a way to access and use that data at the point of care. If clinicians want to review details from free-text encounter notes, they don’t have time––nor likely the patience––to open the EHR dumpster, dive in, and hope to come across what they want.

Rather than dumpster diving, clinicians need tools that allow them to select any item from the problem list and instantly see the clinically relevant information from all the different places and various formats in the EHR. One way to do this is with the integration of a clinical relevancy engine that can identify the relationships between diagnoses, medications, lab orders and results, and history and physical examination findings, including mappings to the relevant terminologies and code sets in each domain, and then use diagnostic filtering to present relevant information at the point of care.

By adding this type of technology into legacy EHRs, users can quickly find clinically relevant information that supports patient care and addresses value-based care requirements.

Dumpster diving for data in the 21st century

The advent of the 21st Century Cures Act will exponentially increase the challenge of finding relevant data within an EHR. When the interoperability floodgates are opened and providers are required to send more information back-and-forth, the contents of one provider’s dumpster will be added to another’s. A clinician might be willing to reach in her own trash can to find something, but she will likely be a lot more hesitant to go digging around in someone else’s garbage.

If interoperability is going to improve outcomes and contribute to the success of value-based care, new tools are needed to support integration of relevant clinical data with specific documentation, workflow, and reporting requirements. Some of the more widely discussed approaches, however, may fail to address the challenge.

Consider, for example, emerging technologies designed to work in the background to capture the encounter and lighten documentation burdens for clinicians. These solutions include ambient artificial intelligence that use microphones to capture sound, translate the conversation into text notes, and apply natural language processing to identify relevant clinical data. While such tools sound promising, the error rate with these solutions is 10-20% at best, which means someone must manually review the captured data to ensure accuracy. Imagine if banks captured sound, turned it into text, then analyzed the findings to determine––with a high degree of inaccuracy – how much money was in your account!

Another alternative is the use of scribes to record encounter information. While this approach capture visit information at the point of care, it does not necessarily help create a record with clean, structured data––which is crucial for addressing newer data and outcomes reporting requirements, such as electronic clinical quality measures (eCQMs), and for managing Medicare Advantage patients using hierarchical condition codes (HCCs).

While ambient AI technologies and scribes may help capture encounter information and even help facilitate after-the-fact analytics, neither option solves the dumpster-diving-for-data problem: that is, clinicians need solutions that help users find relevant clinical data at the point of care to drive better patient outcomes and cost-effective care.

One provider’s approach

At Phoenix Children’s Hospital, administrators wanted to give clinicians an easier way to access critical patient information. Clinicians desired instant access to key clinical indicators for each specific patient and their specific conditions to support action at the point of care. They wanted information organized in alignment with the way clinicians think and work, including summaries for each area of ​​concern.

The organization also needed the solution to integrate into their legacy EHR. They adopted a clinical relevance engine that included workflow customization tools to drive patient-specific dashboards, actions, and documentation.

The technology has helped clinicians quickly find the information they need to manage and track patient care, and more importantly, to improve patient outcomes. In some departments, physicians enter as much as 97% of the notes in a structured format, allowing them to push clinical documentation data to a data warehouse that populates clinical dashboards and provides visual representations of individual patients, as well as the patient population. Instead of wading through patients’ charts to understand their status, clinicians can see at a glance how patients are progressing, identify any gaps in care, and quickly make treatment decisions.

With the enforcement of the 21st Century Cures Act, clinicians will be flooded with even more data to manage and interpret. To save clinicians from dumpster diving to find relevant data, organizations must embrace new technologies that work with their legacy systems and give users the patient and-problem-specific information they need at the point of care.

Photo: Filograph, Getty Images


Leave a Comment

Your email address will not be published.