Intimate Partner Violence and clinical coding: issues with the use of the International Classification of Disease (ICD-10) in England

Abstract

Objectives
To investigate the availability of intimate partner violence (IPV) related population health information in England and the possibility of identifying IPV exposed population sample frames from administrative health data systems in England employing the International Classification of Disease (ICD).

Methods
Research design was an exploratory mixed method approach that involved trend analysis of numbers of applications of ICD IPV classifications for admissions to NHS hospitals in England over a five-year period and semi-structured focus group interviews with clinical coders at an NHS Hospital.

Results
Use of ICD IPV classifications were generally low across NHS Trusts in England. There was notable variation in the numbers of applications across NHS providers which demographic differences or rates of violence perpetration would not account for. The interview findings revealed conceptual ambiguity regarding IPV classifications which presented challenges for clinical coding and raised questions about the reliability and validity of ICD’s IPV classifications.

Conclusion
It would not be possible to extract robust data about populations exposed to IPV for the purposes of audit, governance or research from health information systems using current ICD-10 classifications. Development of these ICD codes is essential for violence and abuse to be captured more accurately in health information systems and afforded greater prioritization and funding proportionate to the health burden and service demands that IPV is responsible for.