Psychological assessment instruments: a coverage analysis using SNOMED CT, LOINC and QS terminology

Psychometric instruments, inventories, surveys, and questionnaires are widely accepted tools in the field of behavioral health. They are used extensively in primary and clinical research, patient care, quality measurement, and payor oversight. To accurately capture and communicate instrument-related activities and results in electronic systems, existing healthcare standards must be capable of representing the full range of psychometric instruments used in research and clinical care. Several terminologies and controlled vocabularies contain representations of psychological instruments. While a handful of studies have assessed the representational adequacy of terminologies in this domain, no study to date has assessed content coverage. The current study was designed to fill this gap. Using a sample of 63 commonly used instruments, we found no concept in any of the three terminologies evaluated for more than half of all instruments. Of the three terminologies studied, SNOMED CT (Standard Nomenclature of Medicine - Clinical Terms) had the greatest breadth, but least granular coverage of all systems. While SNOMED CT contained concepts for over one third (36%) of the instrument classes in this sample, only 11% of the actual instruments were represented in SNOMED CT. LOINC (Logical Observation Identifiers, Names, and Codes), on the other hand, was able to represent instruments with the greatest level of granularity of the three terminologies. However, LOINC had the poorest coverage, covering fewer than 8% of the instruments in our sample. Given that instruments selected for this study were selected on the basis of their status as gold standard measures for conditions most likely to present in clinical settings, we believe these results overestimate the actual coverage provided by these terminologies. The results of this study demonstrate significant gaps in existing healthcare terminologies vis-à-vis psychological instruments and instrument-related procedures. Based on these findings, we recommend that systematic efforts be made to enhance standard healthcare terminologies to provide better coverage of this domain.