Introduction

Eurocat is a network of population based registers of congenital anomalies in Europe covering about a quarter of the birth population in 19 countries (http://www.eurocat.ulster.ac.uk/). We surveyed registries with regard to the requirement for informed consent and its implementation.1

Participants, methods, and results

We sent a questionnaire on ethics and confidentiality developed by the Eurocat Working Group to 35 registries in 2003 and updated June 2004; 29 registries from 15 countries replied (table). Eight registries reported experience of opt-in informed consent.

Five registries depend on medical records and notification from clinicians. One experienced a fall in registration (less than 10 written consents in the entire year in which opt-in consent was instituted, compared with 249 cases in the year before opt-in) such that an exemption was negotiated enabling a switch to opt-out consent. Currently 0.1% of parents opt out. A second registry, in which notifying clinicians ask for consent by post, is permitted to keep a reduced, anonymous set of documentation on cases without consent (about 18%). A third registry gives administrative help for clinicians obtaining consent by post (amounting to 1-3 hours a case) but still estimates 15-20% loss of cases through non-response, although only 0.5% of parents actively refuse to participate. A fourth registry is not fully operational because of low notification levels related to the consent requirement. All these registries reported difficulties persuading busy clinicians to undertake the additional work of obtaining consent for the registry, or convincing clinicians of the value of collecting registry data. Healthcare professionals have also to coordinate consent procedures to avoid parents being approached multiple times. A fifth registry does not yet know how ascertainment is affected but reports less than 1% parental refusal.

Of the other three registries operating opt-in consent, one registry covering a small population has research paediatric staff who examine all babies (malformed or not) born in participating hospitals, for which consent is obtained at booking. This registry reports only two parental refusals since 1990. One registry is based on interviews of cases and controls shortly after birth by clinicians who then notify the case to the registry; this registry is not aware of problems, although it has little information from clinicians on parental refusals. One registry is a voluntary association of clinicians who obtain verbal consent from their patients when registering the case and is not aware of serious problems, although this has not been formally evaluated.

Comment

Eurocat experience shows that informed consent is a serious threat to the operation of registries relying on clinician notification or access to medical records. Despite extremely low parental refusal, opt-in informed consent poses logistical problems, as other types of registry have found.2^–4 Although much has been written about the right of the individual to be adequately informed and to give consent (the parents in the cases of newborns), further research should evaluate parents' desire to participate in activities that may lead to the protection of the health of children in the community and the subsequent ethical duty on the part of the clinician to inform and to request consent. However, this places a further burden on clinical workload.5

Discussion about opt-in informed consent seems to have eclipsed discussion about effective forms of opt-out consent and also about data confidentiality and research ethics procedures that would be acceptable to the public. The primary concern of most patients is not the use of their data for research but inappropriate access to medical data, and there is insufficient debate about what safe-guards to ensuring confidentiality and the appropriate use of personal data would be sufficient to replace the requirement for individual consent.