Best Practice Recommendations Still Fail to Result in Action: A National 10-Year Follow-up Study of Investigative Interviews in CSA Cases MIRIAM JOHNSON 1,2 *, SVEIN MAGNUSSEN 1 , CHRISTIAN THORESEN 3 , KYRRE LØNNUM 1 , LISA VICTORIA BURRELL 1 and ANNIKA MELINDER 1 1 The Cognitive Developmental Research unit (EKUP), Department of Psychology, University of Oslo, Oslo, Norway 2 Central Norway Regional Health Authority, Trondheim, Norway 3 Oslo University Hospital, Oslo, Norway Summary: The present study examined a national sample of Norwegian investigative interviews in alleged child sexual abuse cases (N = 224) across a 10-year period (2002–2012), in order to decide whether practice had improved over the decade in terms of the types of questions asked. The results indicate that the frequency of open-ended, directive, option-posing, and suggestive questions asked was unchanged during the 10-year period, but that the frequency of repeated questions had increased signifi- cantly. When analyzed together with the results from a previous study of interview practice covering the period of 1990–2002, long-term trends were found to be similarly stable, with no change in the frequency of open-ended and suggestive questions asked over a 22-year perspective. However, a decreasing frequency of option-posing questions observed accompanied by an increasing frequency of directive questions may be considered a positive trend. Copyright © 2015 John Wiley & Sons, Ltd. Over the last 30 years, the investment of considerable re- sources to avoid negative outcomes in alleged child sexual abuse cases (CSA) has served as a motivator for profes- sionals’ development of best practice recommendations de- signed to maximize the child’s ability to elicit accurate recollections of events in a child-sensitive and legally defen- sible manner (e.g., Fisher & Geiselman, 1992; Lamb, Hershkowitz, Orbach, & Esplin, 2008; Milne & Bull, 1999). The professional consensus concerning both the ap- propriate ways to conduct investigative interviews and interview-related factors hindering accuracy has, however, not ensured that interviews are typically conducted according to the recommendations. Despite the availability of practice guidelines that recommend interviewers to rely on open- ended prompts and avoid option-posing and suggestive tech- niques that at times negatively affect children’s memory reports (see Goodman & Melinder, 2007, for a review), field research has identified an overall failure to adopt these guide- lines into practice. A growing number of field studies that have examined interview practice in Israel (Hershkowitz, Horowitz, & Lamb, 2005; Lamb et al., 1996; Orbach et al., 2000), Sweden (Cederborg, Orbach, Sternberg, & Lamb, 2000), Australia (Agnew, Powell, & Snow, 2006; Guadagno & Powell, 2009), Finland (Korkman, Santtila, & Sandnabba, 2006; Santtila, Korkman, & Sandnabba, 2004), the United States (Sternberg et al., 1996; Warren, Woodall, Hunt, & Perry, 1996), Canada (Cyr & Lamb, 2009), and the United Kingdom (Davies, Westcott, & Horan, 2000; La Rooy, Lamb, & Memon, 2011; Westcott & Kynan, 2006) indicate that there is a significant gap between what is known and what is consistently done. The widespread tendency in inves- tigative interview practice across countries shows that inter- viewers rarely use open-ended questions and invite free narratives but rely heavily on option-posing (i.e., forced choice questions with limited response alternatives) and sug- gestive prompts (e.g., Cederborg et al., 2000; Santtila et al., 2004; Sternberg et al., 1996; Warren et al., 1996). On the basis of 45 investigative interviews, Sternberg et al. (1996) reported that interviewers asked 4% invitations, 40% leading, 10% suggestive, and 27% directive questions. A similar study analyzing 22 interviews of Israeli children reported 2% invitations, 47% directive, 25% leading, and 8% suggestive questions (Lamb et al., 1996). A Swedish study by Cederborg et al. (2000) analyzing 72 police inter- views showed that the interviewers asked 6% open-ended, 41% directive, 39% option-posing, and 14% suggestive questions. Although the tendency of inappropriate questioning strategies was assumed to be associated with untrained inter- viewers, research has shown that inappropriate questioning strategies are still dominant even when investigators have received specialized training (e.g., Aldridge & Cameron, 1999; Cederborg & Lamb, 2008; Sternberg, Lamb, Davies, & Westcott, 2001). However, several recent studies have shown that interviewers trained by structured and systematic training programs employ more open-ended invitations and fewer inadequate interview practices (e.g., the use of option-posing and suggestive prompts; Cederborg, Alm, Lima da Silva Nises, & Lamb, 2013; Cyr & Lamb, 2009). A study of 195 Norwegian forensic interviews in alleged CSA cases during the period of 1990–2002 revealed a small decrease in the frequency of suggestive and yes/no option- posing questions, but the frequency of open-ended questions was low and had not changed much over the period (Thoresen, Lønnum, Melinder, & Magnussen, 2008; Thoresen, Lønnum, Melinder, Stridbeck, & Magnussen, 2006). During the time segment studied (1990–2002), the Norwegian Police University College introduced manuals and training programs in forensic interviewing of children in 1999, and regulations on interviewing child witnesses were introduced in Norway in 1998 (Norwegian Ministry of Justice and the Police, Section G-70/98). In recent years, the training programs have been further extended by incor- porating additional training where the police officers receive *Correspondence to: Miriam Johnson, The Cognitive Developmental Re- search unit (EKUP), Department of Psychology, University of Oslo, Oslo, Norway. E-mail: miriam.sinkerud@psykologi.uio.no Copyright © 2015 John Wiley & Sons, Ltd. Applied Cognitive Psychology, Appl. Cognit. Psychol. 29: 661–668 (2015) Published online 10 August 2015 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/acp.3147