Today's electronic health records (EHRs) mostly focus on automating paper data and processes to electronic data and processes. If EHRs are to serve the needs for upcoming quality improvement requirements and ICD-10 conversion mandates, however, they must be re-designed.

In his Education Session 2, "Re-Designing EHRs for Quality Improvement and ICD-10 Conversion," Victor Freeman, MD, MPP, with the Health and Human Services Department's Health Resources and Services Administration, discussed how EHRs need to deliver other capabilities, particularly in the area of data mining, to deliver optimal value for clinicians.

Data-mining capabilities will help healthcare providers comply with requirements from government, accountable care organizations and other entities that rely on reporting for quality assurance and quality improvement initiatives, he said.

To date, most EHRs provide data entry and viewing, with some connectivity. Reporting is minimal, and done mostly through custom design. Freeman pointed out that intensive, multi-level in-house reporting will help deliver improved documentation needed for quality improvement programs. Tagging and flagging are two functionalities that can assist in providing detailed documentation.

"Tagging makes reporting much easier," he said. A data field can be designed to identify a patient for subsequent follow-up, for example, base on a character that is not otherwise discretely noted in the electronic chart, he said. Tagging helps "close the loops" and allows physicians to document what happened to their patients.

Flagging results and result orders for subsequent follow-up minimizes the problem of patients falling through the cracks, Freeman said. "As physicians are held more accountable for patient care, they need more information on patient adherence," he said.

Flagging is also a useful where scanned data, images or narratives are concerned. For scanned patient information, quality improvement staff often has difficulty searching for abnormal because of the lack of discrete fields. Flags can be done when a clinician "signs off" as having reviewed the results and orders; this is the time when clinicians can flag abnormalities or new abnormalities at the same time. "Flags are useful tools because they draw attention to these abnormalities that would be missed in scanned narratives," he said.

With tagging and flagging, designers of EHRs must remember not to create too many alerts and create "alert fatigue." Alerts should be confined to severe details, issues and concerns, and should be adjusted by individuals and groups.

In order for alerts/reminders to be useful, clinician buy-in is critical, Freeman said. These alerts and reminders also need to be designed so they do not block workflow, and clinical decision support needs to be adapted to the clinician. Compliance should also be monitored, and clinical feedback should be delivered with valuable details and respect, he said.

While more detailed patient information will come with ICD-10 conversion, EHRs will be key in helping to capture and document that data. "If clinicians have to change to a new way of (thinking) coding, why not have it occur in the context of an EHR change?" Freeman said.

As forthcoming mandates and programs require reporting capabilities, healthcare providers need to make sure that their EHRs are designed to deliver reporting for quality assurance and quality improvement.