Unleashing the Potential of Clinical Coding in Paediatric Research: A Leap Towards Better Child Health Outcomes

Clinical coding plays a fundamental role in healthcare by standardising the recording of diagnoses and procedures, thereby ensuring that vital health data can be easily interpreted and analysed. But how can it improve our understanding of chronic paediatric conditions (CPCs) and their impact on child and young people’s health across the UK? A new study by Swann et al aims to address this question by making paediatric research simpler across primary and secondary care datasets. This article explores the key aspects of this study and its potential to enhance the efficacy of clinical coding in paediatric health research.

Bridging the Gap Between Primary and Secondary Care

Historically, population health analyses, particularly those related to CPCs, have largely relied on secondary care data. These data are categorised using standard ICD-10 codes, with comorbidities being captured using the same. However, this approach has a significant limitation – it doesn’t account for a large proportion of healthcare use by children and young people (CYP) in primary care, which isn’t standardised in the same way.

Given that primary care also uses different coding systems (SNOMED in England and Read v2 in Scotland and Wales), the lack of a harmonised codelist for CPCs makes it challenging to link primary and secondary healthcare use across the UK, limiting our understanding of the true prevalence and impact of CPCs.

A Promising Methodology to Harmonise Codelists

To bridge this gap, the researchers developed harmonised primary and secondary care codelists for CPCs, covering the three most common UK coding systems – ICD-10, SNOMED, and Read v2. The process started by using Hardelid et al.’s well-established ICD-10 codelists for CPCs as a foundation. The conditions were then screened by two paediatricians, and acute conditions were removed. Existing primary care published codelists for asthma, diabetes, cancer, severe mental illness and autism were integrated. New codelists were then created for the remaining conditions using the CALIBER R package to map ICD-10 codes to SNOMED and Read v2 codes.

This innovative approach to clinical coding provides a comprehensive, harmonised perspective that effectively captures the patient journey across both primary and secondary care environments.

Unveiling Results and Implications for Paediatric Healthcare

The researchers successfully mapped twenty-two CPC subgroups. Conditions were considered active if they had been coded for in the past five years. Some conditions were deemed permanent and hence considered active if coded for at any time. The codelists created by the study are now publicly available and will significantly assist researchers to undertake harmonised primary and secondary care studies, considering the impact of CPCs on CYP across the UK.

This breakthrough has far-reaching implications. By understanding the burden of chronic conditions in CYP, we can develop effective strategies to improve clinical care, paving the way for better health outcomes. It also highlights the value of clinical coding as an integral tool in healthcare research, strengthening its role as an indispensable pillar in shaping public health strategies.

The novel approach put forward by Swann etal shows the potential of effective clinical coding practices, and the transformative impact they can have on paediatric healthcare. As we strive to create a healthier future for our children, embracing the advances in clinical coding will be instrumental in bringing this vision to life.


  1. Swann O, Williams TC, Fraser LK, Farrell J, Harrison E, Docherty A, Pollock L. Making paediatric research simpler across primary and secondary care datasets; harmonised codelists for chronic paediatric conditions across the UK. Not our study, 2023.
  2. CALIBER, Technology Reference Update Distribution NHS Data Migration tables. 2023.

Related Articles


This site uses Akismet to reduce spam. Learn how your comment data is processed.