RE: research ethics question

["CAWSON, Pat - NC" <PCAWSON@xxxxxxxxxxxx>]

Mark, what you should do depends on what ethical procedure you established to deal with confidentiality and  disclosure when the study was set up in the first place. Every study interviewing children where disclosure of abuse (or for that matter any disclosure of illegal acts at any age) could result from the questions asked, should establish a statement on confidentiality and an  open procedure for dealing with disclosure or any other suggestion of serious risk to a child, and it should be clearly explained and agreed to by everybody, including children, before the interview takes place. Failure to give attention to this at design stage is a serious failing in ethical standards for research. If your procedure allows for breach of confidentiality to protect a child then you must ensure that answers can be checked ASAP  to ensure that nothing needs reporting. If your ethical protocol forbids any breach of confidentiality in any circumstances(as it does in the UK under the Mar!
 ket Research Society ethical procedures for eg)then one way to deal with that is to ensure at the time of interview that everone knows what to do and where to get help with abuse - we would normally give everyone in large scale surveys a leaflet about where to go for help with anything in the interview that distressed them, including 'helpline' numbers in case of child abuse. 



Dr Pat Cawson
Head of Child Protection Research
NSPCC National Centre
42 Curtain Rd
London EC2A 3NH

020 7825 2648
email: pcawson@nspcc.org.uk


-----Original Message-----
From: Chaffin, Mark J. [mailto:Mark-Chaffin@ouhsc.edu]
Sent: 21 June 2003 17:02
To: Child Maltreatment Researchers
Subject: research ethics question


I have a question on research ethics and mandatory child abuse reporting for which I would like to gauge the opinions of child abuse researchers out there.

Suppose data is collected from children on an instrument such as the Conflict Tactics Scale-Parent/Child, using a computer self-report methodology.  In this methodology, the data collector never sees the child's responses, and the responses exist only as an ASCII file of numerical data.  The data collector does not observe signs of abuse or personally receive any disclosure of abuse.  Some weeks or months later, after the data is uploaded, a data analyst notices that some children have endorsed CTS-PC items suggesting severe parent-to-child violence.  The data set includes no identifying information about the child, but the technology does exist within the study to track back to an indentifiable child through a separate subject key table by breaking the code between subject ID's and names.  What is the ethical and legal responsibility of the study?  Would it be:

1) Do nothing--this does not rise to the level of a specific reasonable suspicion which would trigger reporting requirements
2) Track the identity of the child back through the key and make a report to the authorities.  If so, should the study also notify the parent (presumptive perpetrator) that the report was made?
3) Track the identity of the child back through the key and re-contact the child in order to clarify their response to the computer and clarify if there is a reasonable suspicion which rises to the level of a reporting requirement.  If so, should the parent (presumptive perpetrator) be contacted first?
4) Do something different than the options above--please specify
5) You ethically shouldn't do this study in the first place.

Thanks for your input.

Mark Chaffin