This study was part of a larger study of court files for child sexual abuse cases, which reviewed all 800 cases tried in Swedish district courts during the years 1985, 1989, 1992, and 1997. All of the cases examined had been brought to trial. There were 143 cases in which an expert witness had been consulted on the following issues: to give an opinion on credibility or reliability related to the child or a child's statement; to offer comments on matters of credibility or reliability in the written expert report; or to discuss the expert's statement in relation to such issues. Because a qualitative methodology was used, a small sample size was selected (n=20). A cross-professional research team performed a qualitative text analysis of the files of the 20 court cases, which occurred over the span of the 4 years examined. The study found that expert witnesses used a wide variety of assessment procedures for determining a child's credibility and reliability in the sexual abuse cases. Several patterns of communication difficulties between the court and the expert witnesses were identified. They were labeled as "mismatch," "misunderstanding," "misuse," and "missing pieces." The experts and the courts sometimes used the same or similar words (e.g., "credibility"), but apparently applied different meanings to them, indicating the lack of a common conceptual framework for the court and the expert. The study concluded that the combination of brief assignments for expert witnesses and experts' use of diverse assessment procedures constituted a specific risk factor for impaired communication between the court and the expert. This paper concludes with a discussion of various strategies for improving court-expert communication. 2 tables and 54 references