Article Text

Download PDFPDF

223 Achieving SGD 3 goals with genetics, internet technology and data-driven improvement
  1. Liesbeth Siderius,
  2. Elmas Citak,
  3. Rob Braamburg,
  4. Sanath Lamabadururiya,
  5. Anton Heys,
  6. Marc de Graauw
  1. Netherlands


Background Some 93 million children under 15 years of age live with a moderate or severe disability (WHO, 2015). Disabilities, such as hearing deficit, cerebral palsy, developmental disabilities, turn out to have genetic causes. In 2019 the WHO launched a classification of digital health interventions v1.01: Shared language to describe the uses of digital technology for health. In lower and middle-income countries the use of electronic data in medical care is still on the verge of development. Awareness of available genetic tests could be improved.

Objectives International codes and classification are designed to support data harmonization and data exchange. We questioned how new techniques can promote healthy lives and well-being.

Methods Most states provide maternal and child health handbooks (WHO) containing essential information about pregnancy, birth, growth, development, physical examination and vaccination, to promote and maintain health. We identified, among other codes, the Logical Observation Identifiers Names and Codes, a universal standard and electronic database for clinical care and management, applicable to child health care (such as codes 8339-4Birth weight Measured and 8287-5Head Occipital-frontal circumference by Tape measure) for registering the first feature of a (rare) disabling condition. Medical guidelines on Shwachman Diamond Syndrome (SDS) and Thalassemia Major were reviewed on recommended measurements. LOINC codes specific for SDS and thalassemia were applied in text mining for processing PubMed document sets.

Results We identified a subset of international interoperable codes to help to identify (rare) disabling conditions presenting in the first years of life. Child health handbooks can be enriched with a simple laboratory test to help diagnose diseases as a possible part of a (rare) condition.

Abstract 223 Table 1

Text mining is a powerful tool for processing PubMed document sets to identify diagnostic test in literature. Using unsupervised techniques such as clustering and spatial placement, with one can quickly gain insight into the contents of the documents, discover hidden properties and determine how to further explore and label the data.2 Terms that belong to LOINC codes in filtering the whole (ranked) document set one can identify also the important rare disease papers that most likely are relevant for a medical test.

Conclusions The application of specific code-sets ensures the harmonization of data and the possibility of data exchange. As demonstrated with the LOINC the establishment of an interoperable child health record, including children with chronic illness and disabilities, is feasible. Collaboration between paediatricians, families, health system managers and data services is necessary to provide digital solutions to support the SDG3.




Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.