The untapped potential of patient-generated data

Since the beginning of the COVID-19 pandemic, the shift in the point of care from the physician’s office to the patient’s home – via telehealth and patient health portals – has demonstrated the necessity and viability of this newer care option. This shift has also reinforced the importance of using patient-friendly terms to ensure that individuals clearly understand what their healthcare providers are communicating.

Now that patients have been empowered to access, share, and store their health data electronically using services like portals and personal health records (PHRs), a significant information source — the patient’s own account — can be conveyed or saved for future reference. Health diaries presenting subjective serial accounts of treatment; addenda by the patient to clarify subjective statements; and additional or new health-related information communicated by the patient are becoming more readily available to clinicians.

The opportunity and challenge of patient-generated data

Capturing and standardizing this trove of information in order to document, communicate, and mine it for secondary uses provides both an opportunity and a challenge. The opportunity is the ability to convert patient prose into true, codable data. This lets the patient forward information as data that could become intelligible and actionable for rules-based logic. The value to the patient is the ability to record information and to receive guidance based on subjective data. The challenge is developing a terminology not rooted in a recognized standard.

Patient terms – such as pain, nausea, and fever – are often used in the subjective section of a medical document. These terms are clinically relevant even though, as subjective descriptions, they are not coded. Subjective data, signs, physical exams, and lab results are all gathered and used by the clinician to arrive at the most explicit and accurate interpretation of these symptoms – the diagnosis. When a diagnosis eludes the clinician, he or she might present the symptoms as the highest resolution definition of the patient’s problem. Some of these, like fever and headache, have diagnostic codes assigned to them, but many, like cold and scratchy throat do not.

Addressing the terminology needs of patients

Electronic health record (EHR) vendors use proprietary dictionaries to capture subjective patient data. Typically, this data appears in the nursing notes section or the chief complaint; history of present illness; and subjective sections of a physician’s note. These dictionaries could potentially be used by vendors as rudimentary patient terminologies for PHRs and patient portals.

Unlike physician-recorded diagnoses and procedures that can be mapped to standardized codes, the very nature of patient-generated input is free-form. That said, patient-authored health comments are invaluable to the clinician and relatively simple to express. Certain terms currently exist in standard coding vocabularies such as SNOMED CT® and ICD-10-CM. In those vocabularies, the significance of the term (“fever,” for example) means that the physician has validated the finding.

The terminology needs of patients – the most numerous and largest stakeholders in healthcare – remain unaddressed. Building a terminology that could be used to express subjective concepts in portals, PHRs and EHRs could fill this void. Limiting the number of primary concepts, vetting synonyms, and ensuring that ambiguous terms are excluded would likely check the size of the terminology and ensure its manageability.

The patient’s subjective descriptions serve an important role and provide an additional opportunity for understanding disease progression and healing processes. With the spread of patient portals and PHRs, building a standard for capturing those concepts opens a new opportunity to better appreciate the dynamic evolution in patients’ health and more effectively address their needs.

For more on the pivotal role of terminology in healthcare, click here.