- Natural language processing (NLP) has proven an effective strategy for translating clinical terms into lay-language, thus improving patient understanding of patient portal notes, according to research published in the Journal of Medical Internet Research.
Patients are increasingly accessing their provider notes via the patient portal as a part of the healthcare industry’s overall efforts to improve patient engagement in care. Although better patient data access has largely been viewed positively, there have been some concerns, the research team from the University of Massachusetts Medical School said.
“Many health organizations also allow patients to access their full electronic health record (EHR) notes through patient portals, with early evidence showing improved medical comprehension and health care outcomes,” the research team wrote. “However, medical terms—abundant in EHR notes—remain a major obstacle for patients to comprehend medical text, including EHRs.”
Concerns about patient health literacy and ability to understand complex medical language have been prevalent amongst naysayers of programs like the OpenNotes movement. OpenNotes is a philosophy that states clinicians should offer fully transparent patient access to clinician notes.
“Limited health literacy has been identified as one major barrier to patient use of EHRs,” the research team stated. “Misinterpretation of EHR content may result in unintended increases in service utilization and change of patient-provider relationships.”
Industry experts have floated NLP as a possible solution to the patient health literacy and complex EHR terminology issue. NLP is able to identify important medical words used within an EHR note and connect patients with a reliable and understandable definition.
“Natural language processing (NLP)-enabled interventions have also been developed to link medical terms in EHRs to lay terms or definitions, showing improved comprehension,” the researchers explained.
NLP is advantageous because it connects patients with a reliable, layperson definition, cutting the risk of patients researching terms themselves. In some cases, patients do not access trustworthy information during medical research.
“The aforementioned NLP-enabled interventions have the advantage of reducing patients’ information-seeking burden by integrating authorized health-related information in a single place, and thereby helping patients easily read through and understand their EHR notes,” the research team explained.
Two different NLP algorithms proved effective for aiding patient understanding of EHR notes, the researchers found. The team deployed two NLP programs, one using the consumer health vocabulary (CHV) resource and the other using transfer learning.
“To achieve this goal, we developed a new NLP system, called adapted distant supervision (ADS), which uses distant supervision from the CHV and uses transfer learning to adapt itself to the target domain to rank terms from EHRs,” the researchers recounted. “We aimed to empirically show that ADS is effective in ranking EHR terms at the corpus level and outperforms supervised learning.”
The team selected a list of important terms to be parsed out by the NLP programs based upon word unithood, termhood, unfamiliarity, and quality of compound term. These criteria essentially called for a word to be both integral for patient understanding of health as well as complicated enough to warrant definition.
Ultimately, the team found that both NLP programs worked better than any baseline NLP programs to which the team compared them.
This opened a significant opportunity to help improve patient understanding of EHR notes and patient health literacy. Currently, only 36 percent of adult patients are categorized as having high health literacy, the researchers said. Industry experts seek to improve this statistic.
Ensuring patients can adequately understand EHR notes is integral to supporting patient portal adoption and use throughout the industry. Many healthcare providers have expressed trepidation about giving patient portal and clinician note access. Those professionals state that patients may become more confused after reading complex medical terminology.
There is some validity to those concerns, the UMass research team conceded. Few patients are actually able to understand every term used in clinician notes, and accessing online resources and explainers is not always useful.
But pairing each complex term with a layperson definition written at the targeted reading level – ideally a seventh-grade reading level, experts say – will help improve patient health literacy and note comprehension. From there, patients will be more likely to use the information in clinician notes to become more engaged in their own healthcare.