Skip to main content

How CARIS processes data

Data processing

Upon receipt of a report:

  • CARIS checks case details against the Welsh Demographic System (WDS) to validate dates of birth, address details, postcode and NHS numbers. 
  • WBNS (the Welsh Birth notification Service) provides details information such as birth weight, hospital and gestational age at birth.
  • An initial search is made to check if the case has already been recorded. If so, the record is updated as appropriate. If not, a new registration is made.
  • The CARIS database includes built-in postcoding software which revalidates the postcode of mother’s address and from this derives the local authority, ward and approximate grid reference related to the address.
  • Diagnoses, maternal medications and occupations are coded.
  • Further internal checks are made to validate the data to ensure its quality and consistency.

Coding

Good coding of congenital anomalies is essential for CARIS. All anomalies reported to CARIS are coded using the Royal College of Paediatrician and Child Health expansion of ICD10. ‘ICD10’ is the International Statistical Classification of Diseases and Related Health Problems, 10th revision and is published by the World Health organisation (WHO).  All CARIS staff undergo training in using ICD10 and have been taught the principles underlying it.

These include:

  • coding every anomaly identified
  • using the minimum number of codes necessary to accurately describe the clinical picture
  • coding to the greatest level of specificity available in the classification.

In addition CARIS groups anomalies using a modification of the system devised by the Office for National Statistics (ONS). Since 2002, EUROCAT  has developed its own classification of congenital subgroups and this is used where appropriate.

Medications taken by mothers during pregnancy are coded using the Anatomical Therapeutic Chemical Classification (ATC codes) published by WHO. This is used as a standard by EUROCAT registers and allows investigation into possible associations between particular drugs and anomalies.

Occupations are coded using the International Standard Classification of Occupations (ISCO-88).

Status of CARIS cases

When cases are registered, they are assigned a status depending on the source of data and the strength of evidence for the diagnosis. Examples are provided in the table.

Source of Data

Status usually assigned

Cytogenetics & Postmortem Confirmed
Clinical reports on live births Confirmed
Antenatal ultrasound Suspected
Inpatient data and NCCHD (National Centre for Child Health and Development) – specific codes Probable
Inpatient data and NCCHD – non specific codes Suspected

All post-mortems and cytogenetics reports are also classed as confirmed cases. Normally all antenatal ultrasound findings of structural anomalies are treated as suspected and then revised to probable or confirmed at the end of pregnancy. If pregnancy ends in termination or fetal loss and there is no further information available e.g. no post-mortem then definite antenatal findings such as spina bifida or gastroschisis are classed as probable.

Cases with only antenatal soft markers are classed as suspected. All suspected cases are routinely followed-up. Where there is evidence that antenatal findings such as renal dilation has resolved, these cases are signed off as normal.

Inpatient data is used cautiously. Where specific codes are used these are treated as probable cases, and where possible, confirmed via WCP (Welsh Clinical Portal). Where non-specific codes are used they are classed as suspected cases.

All cases are followed up and verified with paediatric case notes. At this point the status is revised to either confirmed or normal i.e. with no congenital anomaly.

All suspected cases over time should resolve into being either a confirmed case or normal.

For a few cases, it is not possible to follow them up; they are marked as having no further information available. In some cases it is impossible to be certain of a diagnosis and these are then marked as possible.

When CARIS calculates the prevalence rate of congenital anomalies, only confirmed and probable cases are used in the calculation. Likewise when reporting cases to EUROCAT and the International Clearing House for surveillance only the confirmed and probable cases are reported for inclusion.

Data security and confidentiality

All data is kept under strict security conditions.

CARIS has received dispensation from the National Information Governance Board for Health and Social Care, under section 251 of the Health and Social Care Act 2006, to process patient identifiable information.

Requests for CARIS information come from many different sources, for example from clinicians, academics, Welsh Government or requests made under the Freedom of Information Act.

In answering requests, strict criteria are applied to ensure patient confidentiality is maintained and that the Data Protection Act is followed at all times. For any situations where there is the slightest doubt, the situation is discussed with senior staff working with CARIS and senior managers of Public Health Wales.

CARIS does not release patient identifiable data to anyone other than the clinicians or clinical teams who supplied the identifiable data in the first place.