OCL Content Publication Plan

This document outlines OCL’s plans to publish and maintain selected vocabularies in OCL Online. OCL Online is a hosted implementation of all of OCL’s open-source tools in the cloud and pre-loaded with key terminologies, available at openconceptlab.org.

OCL’s Goal for Content Publication

OCL aims to publish and maintain key reference vocabularies and other emerging health data standards, especially those that are important to low and middle income countries. Published content will be available on OCL Online and may also be available for distribution.

Scope of Content Publication

Status

Reference Vocabularies

Emerging Health Data Standards

Available in OCL Production

  • HL7 FHIR Terminologies

  • WHO ICD-10

  • LOINC

  • SNOMED-GPS

  • CIEL Interface Terminology

Planned and In Progress

  • SNOMED-CT

  • RxNorm

  • US ICD-10-CM

  • WHO Smart Guidelines

  • PEPFAR MER and MOH Alignment

  • Partners in Health (PIH) OpenMRS Concept Dictionary

How do we obtain and load content into OCL?

The OCL team works in partnership with Apelon, Inc., which normalizes and publishes terminology content in its DTS system.

  1. Content Inquiry
    • The OCL team requests a set of terminology content from Apelon, based on community needs, availability, and appropriateness to share and eventually publish.

  2. Import File Generation
    • Using a set of field mappings between OCL and Apelon DTS data models, Apelon generates a content file suitable for OCL’s bulk import using their transformation script.

  3. Import File Review and Preparation
    • OCL team reviews the import file for suitability or potential issues with OCL. Example: Align organization identifier with an existing organization in OCL

    • Ensure that permissions are in place that the terminology will not yet be publicly viewable by placing it in a private source accessible only to OCL’s admin account.

  4. Initial Test Import and Teview in OCL QA Environment
    • Attempt to bulk import first ~200 rows of import file into QA environment for OCL. If required, troubleshoot syntactical errors

    • Spot check imported concepts, checking that the correct number of concepts are present in OCL QA, the correct attributes are present in the correct format, and hierarchy is showing correctly, if applicable.

    • Make corrections as needed.

  5. Full Import into OCL Staging Environment
    • Bulk import content into OCL Staging. Note: breaking the import file into chunks may be necessary if there is a memory limit to what OCL’s bulk import can accept.

    • Spot check individual concepts, along with counts of concepts and mappings if desired.

How do we validate content when we load it in OCL?

To ensure that the content loads into OCL as expected, there are several validation steps taken for each set of content to check for accuracy between what should show in OCL versus what is actually appearing in OCL.

  1. Spot Checks: Looking at 5 or more exemplary concepts in depth, we check that the concept appears in OCL and that all the expected attributes are present. If applicable, we check that these concepts are correctly placed in their hierarchy.

  2. Hierarchy Check: If the content is hierarchical, we look down its tree, particularly looking for concepts that should but do not have a parent concept.

  3. Automated Script: A python script uses two files, the bulk import file and the export file created within OCL, for a set of content to check for multiple potential errors. These errors are noted in an output file that includes the following:

    • Missing concepts and mappings that are not present in OCL

    • Extra concepts and mappings that are present in OCL but should not have been loaded in

    • Checks for duplicate concept or mapping IDs in the export file

    • Compares concept attributes to ensure these attributes are similar between the import and the export files

When will content be published in OCL for public use?

Content is eligible for publication in OCL Online’s production environment if they meet the following criteria:

  • This version of the content has been successfully test loaded and validated in a QA or Staging environment

  • The content employs a license that allows distribution by other entities -OR- a distribution license was obtained by OCL to publish the terminology content

Until these criteria are met, content will remain in private repositories in OCL’s test environments. They may undergo updates and model iterations as needed during this time.

How do we keep content up-to-date?

Each terminology has its own process and schedule for releasing updates. OCL and its partners, Apelon and the Regenstrief Institute, work actively to keep published content up-to-date. The table below outlines the release process for curated terminologies published in OCL production.

Note that most content available in OCL is managed directly by other users or organizations and thus will have their own release processes.

Content

Release Process

CIEL

Approximately monthly releases, made available on OCL within 72 hours

HL7 FHIR Terminologies

Updated only as needed

PEPFAR MOH Codelists

New codelists published annually, with maintenance releases as needed