Standards in action: historical and current perspectives

Since its first issue, the Journal of the American Medical Informatics Association has published hundreds of papers about standards to support interoperability among various types of electronic health data and information systems. In this editorial, I highlight 2 perspectives from 2 long-time thought leaders in biomedical and health informatics.^1,2 This is followed by 2 Research and Application papers that focus on terminology standards, the Unified Medical Language System (UMLS) and SNOMED-CT.^3,4 The last highlighted paper addresses information blocking practices related to specific classes in the United States Core Data for Interoperability Version 1 (USCDI v1).⁵

Donald W. Simborg, MD, founding member of Health Level 7 (HL7), reflects on its origins and impact on interoperability in hospitals.¹ He particularly notes that although HL7 is the most widely used data-interchange protocol worldwide in health care, the promise of open architecture envisioned by its founders did not prevail in healthcare as it did in other industries due to some unique characteristics of the healthcare environment. Enrico Coiera proposes a framework to support the scientific research of standards so that they can be better measured, evaluated, and designed.² He argues that fitting work using conformance services is needed to repair these gaps between a standard and what is required for real-world use and identifies 3 strategies: (1) universal conformance (all agents access the same standard); (2) mediated conformance (an interoperability layer supports heterogeneous agents); and (3) localized conformance (autonomous adaptive agents manage their own needs). Coiera’s approach conceptually decouples interoperability and standardization, and he concludes that although “standards facilitate interoperability, interoperability is achievable without standardization.”

Newbury, Liu, Idnay, and Weng evaluated the suitability of the UMLS and SNOMED-CT in encoding outcome concepts in 500 randomized controlled trial abstracts.³ Forty-four percent of 2617 outcome concepts were fully covered in the UMLS and SNOMED-CT was present as a source in 61% of the fully mapped outcomes. Given the importance of computational representation of clinical outcomes for clinical evidence extraction and appraisal, the study findings suggest that these resources require extension in this area.

To address the time-consuming task of assigning new atomic terms to UMLS concepts to meet user needs such as those identified in the study above, Mao et al⁴ applied heuristic, deep learning, and hybrid AI methods to predict semantic group assignments for new UMLS Metathesaurus atoms. Based on a sequence of 7 different prediction methods, they developed 2 hybrid semantic group prediction methods combining the strengths of heuristic and deep learning methods. The semantic group prediction accuracy of the hybrid approaches on 1 563 692 new unseen atoms was 96.5% as compared to 94.3% for heuristic or deep learning approaches alone.

The USCDI v1 was adopted as a standard in the Office of the National Coordinator for Health Information Technology 21st Cures Act Final Rule as a foundation for broader sharing of electronic health information (EHI) to support patient care. The Final Rule also establishes statutory penalties for entities that fail to share EHI with patients and/or their proxies (ie, information blocking). However, the relationship between children and adolescents and their parents/guardians raises challenges regarding disclosure of sensitive EHI about the patient or their proxies. To understand current implementation practices, Sinha et al⁵ conducted an online survey and structured interviews with 10 organizations in 6 states that focused on child and adolescent patient and proxy portal access for selected USCDI v1 data classes. They also collected information about the data segmentation capabilities of patient portals. All organizations implemented different portal account types for different user types and were capable of sharing labs, medications, problem lists, imaging, and notes problem lists, imaging, and notes with proxies. However, organizations varied in how sensitive data within USCDI v1 data classes were protected and shared thus raising the concern of privacy breaches and potential confusion about completeness of data. The authors conclude that increased clarity is required from the Office of the National Coordinator for Health Information Technology about information blocking practices and that EHR vendors need to improve data segmentation capabilities.

The highlighted papers provide examples of standards in action and illustrate the importance of the context in which standards implementation occurs. They also identify several areas in which research is required to advance standards and interoperability.