WP8 Deliverable D8.3: Report to EU institutions hosting health care databases with guidelines for improving the quality of the congenital anomaly coding
Authors: Dr Ester Garne, Prof Joan Morris, Dr Marian Bakker
Contributing participants and partners (in Beneficiary numerical order): Joan Morris (SGUL), Maria Loane (UU), Ester Garne (RSD), Amanda Neville (UNIFE), Ingeborg Barišic, Ljubica Boban (KDB), Anna Pierini (CNR-IFC), Marian Bakker, Hermien de Walle (UMCG), David Tucker, Sue Jordan (PHW NHS, SU), Clara Cavero Carbonell (FISABIO), Anna Latos-Bieleńska, Anna Jamry Dziurla (PUMS), Anna Heino, Mika Gissler (THL), Anke Rissmann (OVGU), Paula Braz (INSA), Elly Den Hond (PIH).
Executive Summary and Recommendations
Electronic health care databases are potentially valuable data sources for the surveillance of congenital anomalies and for studies evaluating the risk of congenital anomalies following exposures to risk factors such as medications, viruses/infections and environmental factors in the first trimester of pregnancy. For example, regional and national health care databases could be used to monitor primary prevention measures such as the reduction in the risk of a neural tube defect due to the fortification of flour with folic acid. The EUROlinkCAT study evaluated the quality of the coding in hospital databases in eleven European regions in eight countries.
The EUROlinkCAT study concluded that although the hospital databases do contain important data, they currently cannot be used as the only data source for the surveillance of congenital anomalies. The fundamental weakness is that the type of anomaly and organ system involved are not reported for the majority of terminations of pregnancy for fetal anomaly (TOPFAs). This is essential as over 21.0% of major congenital anomalies reported in EUROCAT from 2015-2019 resulted in a TOPFA. In addition, for livebirths some anomalies are overreported in hospital databases whilst others are underreported.
The study also concluded that data routinely collected in electronic health care databases should be improved to enable the data to be used in the surveillance and research of congenital anomalies. Codes for classifying and reporting the congenital anomalies resulting in TOPFAs in electronic health care databases need to be developed. In addition, the accuracy of the coding of congenital anomalies in all births should be improved and algorithms to accurately discriminate between major congenital anomalies and suspected or minor anomalies should continue to be refined. Some countries, mainly in southern and eastern Europe, who do not currently have electronic health care databases, should be supported in establishing these databases in order to enable effective surveillance to occur across Europe.
The following recommendations are made to enable the potential of the data in electronic health care databases to be fully exploited in the primary prevention of congenital anomalies.
For national registration bodies
- Develop registration systems to assign a permanent unique identification (ID) number to each baby as soon as possible after the birth to ensure that ICD codes and procedure codes for the first days after birth can be linked to the baby.
- In the mother’s record, report ICD codes for terminations of pregnancy for fetal anomalies. Create an additional record, possibly in a separate database, that reports each specific anomaly diagnosed in the fetus and which can be linked to the mother’s ID number.
For health care database designers
- Include outpatient contacts in health care databases as less severe congenital anomalies may not be visible in hospital discharge databases if surgery is not required.
- Allow the use of more than one diagnosis code for both hospital discharges and outpatient contacts.
- Include records, possibly in a separate database, that report each specific anomaly diagnosed in the fetus and which can be linked to the mother’s ID number.
- the coding may be amended by more experienced doctors and coders
- the coding may be refined by results of diagnostic examinations/tests arriving after the TOPFA or after the child has left the hospital.
For health care database coders
- Always use the most specific code available for the congenital anomaly and avoid using codes for “other” or “unspecified” congenital anomaly.
- Use extended versions of ICD for the coding of rare congenital anomalies or use other coding systems to make the rare diagnoses visible in health care databases.
- Continually undergo training to ensure optimal coding quality and consistency of the data on congenital anomalies.
For researchers using health care databases
- Link to data from congenital anomaly registries wherever possible.
- Use, if possible, validated algorithms for identifying congenital anomalies in health care databases.
- Collaborate with the people working with the health care databases as they know their data.
- Discuss all results with the people working with the health care databases as an aid to interpretation and quality improvement.
Click here for a PDF of the report