Investigating child abuse: how interview training really matters

A study of the outcomes of child sex abuse cases in the US state of Utah suggests that the introduction of improved techniques for interviewing young victims leads to fewer cases being dropped early in the investigative process and results in a greater percentage of prosecutions.  The findings support the argument for better training of police interviewers who have the highly sensitive task of gathering information about traumatic incidents.

A paper summarising the study (‘Do Case Outcomes Change When Investigative Interviewing Practices Change?’) will appear next month (May 2013) in the journal Psychology, Public Policy, and Law. The research was carried out by an international team including Michael Lamb, Professor of Psychology at Cambridge University, who is an expert on children and forensic interviewing.

The study is the first to focus on the investigative interview as a predictor of outcomes – such as the filing of criminal charges, prosecution, and guilty pleas or convictions.

The research drew on data from Salt Lake County Children’s Justice Centre which in mid-1997 introduced training for all its police interviewers in techniques developed under the auspices of the US National Institute of Child Health and Human Development and known as NICHD Protocol. The study centred on a before-and-after comparison of outcomes of child sex abuse cases across two periods: a pre-Protocol period of 1994-1997 and Protocol period of 1997-2000.

The NICHD approach to interviewing was developed by Professor Lamb and his colleagues in response to widespread evidence that free recall memory prompts, such as open-ended questions, are most likely to elicit accurate information from children. Previous studies led by Professor Lamb, using data gathered in Israel, the USA and the UK, showed that open-ended questions were effective in interviewing children as young as four years’ old about incidents that may have involved abuse.

The NICHD Protocol – or similar approaches – is now favoured by a growing number of countries with training being given to those who carry out the highly sensitive process of gathering evidence from children who have suffered trauma. The open-ended questioning style replaces or contrasts with a more directive, option-posing or suggestive line of enquiry which research has shown to be associated with erroneous responses.

The results from the study of child abuse cases investigated by police in Utah have clear implications for the UK where the NICHD Protocol has been explicitly recommended to forensic investigators since 2011 in the Home Office’s manual, Achieving Best Evidence. 

Professor Lamb and his colleagues believe that similar results would be likely in the other countries where the Protocol has been adopted because such interviews provide investigators with a much better understanding of what actually happened to the child, tend to be very compelling when the interviews are shown to jurors, and also provide interviewers with other investigative leads they can follow up in pursuit of additional evidence. Currently, English, French, Japanese, Hebrew, Finnish, Korean and Portuguese (Iberian and Brazilian) versions of the Protocol are in use around the world.

The researchers initially looked at an overall total of 760 cases of suspected abuse of children aged from three- to 13-years-old.  Some of these cases were dropped and some were transferred to other jurisdictions, reducing the total to 696. Of these, 364 cases resulted in charges being filed.  The comparison of outcomes across the two periods before and after introduction of the Protocol revealed a number of significant differences at two crucial points in the progress of cases through the investigative process: the filing of charges by prosecutors and the final judicial disposition, through either plea negotiation or trial.

Charges were more likely to be filed following the introduction of the Protocol with 42 per cent of cases being filed pre-Protocol and more than 52 per cent being filed once the Protocol was in use. Once charges were filed, both pre-Protocol and Protocol interviews were highly (and similarly) likely to lead to guilty pleas. This filtering-out factor means that although half the cases were prosecuted, Protocol interviews were associated with a significantly higher rate of conviction. When cases were tried, Protocol cases were almost always associated with guilty verdicts.

These findings support the view that improving the quality of pre-trial phases of investigations is extremely important. “The quality of forensic interviewing practices is of utmost importance if the rights of both child victims and innocent suspects are to be protected. When child abuse is suspected, children’s verbal allegations often constitute the only available evidence. Thus our research into best-practice approach to interviewing has important implications for policy and practice,” said Professor Lamb.

Cases involving the youngest children in the study (those aged two to four) were the least likely to yield criminal charges regardless of interviewer training. Previous research has suggested that younger children may be more reluctant than older children to disclose and talk about abuse. It has also been shown that children who were suspected victims of parental abuse provided fewer informative responses than those who were suspected victims of non-parental abusers.

“When young children are interviewed they may provide less complete accounts than older counterparts might have done, meaning that their evidence is not sufficient to convince prosecutors that a conviction could be obtained at trial.  These findings point to the need for development of interviewing techniques sensitive to the needs of young children who’ve been abused.  Children need protection under the law and abusers should face conviction, regardless of their victims’ ages,” said Professor Lamb.

The paper ‘Do Case Outcomes Change When Investigative Interviewing Practices Change?’ by Margaret-Ellen Pope, Yael Orbach, Michael E Lamb, Craig B Abbott and Heather Stewart will appear in the print version of Psychology, Public Policy, and Law in May. http://www.apa.org/pubs/journals/law/index.aspx

For more information on this story contact: Alex Buxton, Office of Communications, University of Cambridge alex.buxton@admin.cam.ac.uk 01223 761673.