Abstract
BACKGROUND: Routinely collected electronic health records (EHR) offer a valuable opportunity to carry out research on immunization uptake, effectiveness, and safety, using large and representative samples of the population. In contrast to other drugs, vaccines do not require electronic prescription in many settings, which may lead to ambiguous coding of vaccination status and timing.
METHODOLOGY: We propose a comprehensive algorithm to identifying childhood immunizations in routinely collected EHR. In order to deal with ambiguous coding, over-recording, and backdating in EHR, we suggest an approach combining a wide range of medical codes in combination to identify vaccination events and using appropriate wash-out periods and quality checks. We illustrate this approach on a cohort of children born between 2006 and 2014 followed up to the age of five in the Clinical Practice Research Datalink (CPRD) Aurum, a UK primary care dataset of EHR, and validate the results against national estimates of vaccine coverage by NHS Digital and Public Health England.
RESULTS: Our algorithm reproduced estimates of vaccination coverage, which are comparable to official national estimates and allows to approximate the age at vaccination. Electronic prescription data only do not cover vaccination events sufficiently.
CONCLUSION: Our new proposed method could be used to provide a more accurate estimation of vaccination coverage and timing of vaccination for researchers and policymakers using EHR. As with all observational research using real-world data, it is important that researchers understand the context of the used dataset used and the clinical practice of recording.
| Original language | English |
|---|---|
| Article number | e5848 |
| Journal | Pharmacoepidemiology and Drug Safety |
| Volume | 33 |
| Issue number | 8 |
| Early online date | 2 Aug 2024 |
| DOIs | |
| Publication status | Published - 31 Aug 2024 |
Data Availability Statement
The study uses data from the Clinical Practice Research Datalink (CPRD). CPRD does not allow the sharing of patient-level data. The data specification for the CPRD dataset is available at: https://cprd.com/cprd-aurum-may-2022-dataset. The code lists can be found at: https://github.com/Eyedeet/vaccine_methods_ehr_public/tree/main/codelists.Acknowledgements
This work uses data provided by patients and collected by the NHS as part of their care and support. Using patient data are vital to improve health and care for everyone. There is huge potential to make better use of information from people's patient records, to understand more about disease develop new treatments, monitor safety, and plan NHS services. Patient data should be kept safe and secure, to protect everyone's privacy, and it is important that there are safeguards to make sure that it is stored and used responsibly. Everyone should be able to find out about how patient data are used. We would like to thank Jennie Johnson and the PRIMIS team for their insights on clinical coding practices.Keywords
- Humans
- Electronic Health Records/statistics & numerical data
- United Kingdom
- Child, Preschool
- Algorithms
- Infant
- Vaccination/statistics & numerical data
- Vaccination Coverage/statistics & numerical data
- Male
- Immunization/statistics & numerical data
- Female
- Infant, Newborn
- Vaccines/administration & dosage
- Cohort Studies
