STATE of Missouri, Respondent,
v.
Randall FOSTER, Appellant.
No. 28068.
Missouri Court of Appeals, Southern District, Division One.
February 14, 2008.*801 Dan L. Birdsong, Thomas, Birdsong & Mills, P.C., Rolla, for Appellant.
Jeremiah W. (Jay) Nixon, Atty. Gen., and Robert J. (Jeff) Bartholomew, Office of the Attorney General, Jefferson City, for Respondent.
DANIEL E. SCOTT, Judge.
Defendant appeals his conviction of child molestation, statutory sodomy, and child endangerment charges.
Mother[1] and Defendant married in May 2001. Mother and her five children moved into Defendant's home. Around Christmas 2001, after Defendant hit Mother on several occasions and made accusations against her, she took her children, moved to another town, and got a restraining order. After the move, Mother felt something was wrong with nine-year-old Child, whose grades were suffering and handwriting had become "darker and sloppy." Child told Mother that Defendant had molested her several times when she was home sick from school.
Mother called the "hotline," and an investigator videotaped an interview of Child the next day. Samuel Brayfield, D.O., a member of Missouri's SAFE-CARE[2] Network, examined Child and found no supporting physical evidence. He still felt Child was sexually abused, based mostly on what she told him, and he said so in his report.
The case was tried 4½ years later. Although the jury saw Child's videotaped statement and heard several witnesses including the investigator, Mother, Child, Dr. Brayfield, and Defendant the two-day guilt phase basically remained a "he said-she said" credibility dispute. The jury deliberated from 5:25 to 8:15 p.m. without reaching a verdict, and from 8:30 *802 to 9:20 a.m. the next morning before finding Defendant guilty. He was sentenced, after the trial's second phase, to an effective prison term of 28 years.
Defendant claims Dr. Brayfield's testimony improperly bolstered Child's credibility and truthfulness, and that of child sex witnesses generally, by investing "scientific cachet" on the key issue of credibility. State v. Williams, 858 S.W.2d 796, 800 (Mo.App.1993). Missouri strictly prohibits expert evidence on witness credibility. Id. at 800. Such testimony usurps the province of the jury. Id. at 799. It presents the danger that jurors may be over-awed by, or may defer too quickly to, the expert's opinion. Id. at 800. The Williams court noted that other states have overwhelmingly rejected expert opinion, not only on a particular witness's credibility, but also as to the credibility, accuracy, reliability, or truthfulness of sexually abused children as a class. Id. Accord State v. Tyra, 153 S.W.3d 341, 348 (Mo. App.2005); State v. Matthews, 37 S.W.3d 847, 850 (Mo.App.2001); State v. Bowler, 892 S.W.2d 717, 720 (Mo.App.1994).
Despite and seemingly in violation of these well-known principles, the prosecutor's opening statement assured the jurors that Dr. Brayfield, while finding no physical evidence, found Child "very credible and felt like she was abused."
When Dr. Brayfield later testified as an expert, he described his specialized training in examining abused children, and his active involvement in the SAFE-CARE Network since 1989, performing 20-50 SAFE exams annually. He testified that 90-95% of his exams involve sexual abuse allegations, but he finds physical evidence less than 5% of the time. Over objection, he opined to a reasonable medical certainty that most of the other 95% were sexually abused as well. When asked how he, "as a professional SAFE-[CARE] examiner," could make abuse findings without physical evidence, Dr. Brayfield focused on "the history that the child gives." He explained how he takes a child complainant's history, and that it is "how the child has told the story, at what point in their life the story comes out and how the story comes out. Certainly myself, I've been practicing since 1984. I feel I have a sense in whether a child's telling me a lie or not." Defendant's mistrial request was denied, the court stating: "He's an expert to give opinions, and I'm going to allow it."
Thereafter, Dr. Brayfield talked about taking Child's history for this case. They had a "fairly long talk;" discussed the importance of truthfulness; and "very interestingly," Dr. Brayfield testified, Child expressed "a very good insight into lying" and its consequences. He opined on redirect that Child had been sexually abused.[3]
The State characterizes as "profile" evidence Dr. Brayfield's opinions that most of his SAFE exam patients were sexually abused without physical evidence thereof, and that he can sense when children lie. We recognize that:
"[I]t may be appropriate for an expert to testify that a child demonstrates age-inappropriate sexual knowledge or awareness, and that a child's behaviors are consistent with a stressful sexual experience." The latter type of testimony was characterized as "general testimony describing behaviors and other characteristics commonly observed in sexually abused victims (often called general `profile' testimony)," which the *803 trial court has great discretion in admitting.
State v. Matthews, 37 S.W.3d 847, 850 (Mo.App.2001) (citations omitted). However, the doctor's challenged testimony does not describe behavior or characteristics commonly observed in abused children or otherwise fit the description of "profile" evidence. Rather, he essentially testified that there is generally no physical evidence, but most of his sexual abuse complainants are telling the truth. He believes this to a reasonable medical certainty, as a professional SAFE-CARE examiner. He believes this in part, if not primarily, because he feels his years of training and experience give him a sense about children's truthfulness.
Dr. Brayfield may be correct, and no doubt is sincere, but "expert testimony that expresses an opinion with respect to the credibility or truthfulness of witnesses of the same type under consideration invests `scientific cachet' on the central issue of credibility." Williams, 858 S.W.2d at 800. "A doctor's testimony is probably given great weight, especially the testimony of an expert who told the jury he had performed over one hundred SAFE exams." State v. Miller, 924 S.W.2d 513, 516 (Mo.App.1996). By his own testimony, Dr. Brayfield performed far more than 100 such exams, and the prosecutor already told the jurors they would "hear the doctor state that he found [Child] very credible and felt that she was abused." Moreover, part of the challenged testimony was in response to the judge's own question.[4] When the court itself questions a witness, jurors likely assume the information is not just proper, but important. Considering all the circumstances here, the testimony in question was erroneously admitted.
We review for prejudice, not mere error. State v. Garrett, 139 S.W.3d 577, 581 (Mo.App.2004). Error may be harmless if other evidence of guilt is overwhelming, or if the improper evidence was cumulative and not highlighted. State v. Rose, 86 S.W.3d 90, 103 (Mo.App.2002). Neither is the case here.
As to the former, the question is not whether the State made a submissible case, but whether there was overwhelming evidence of guilt. State v. Dexter, 954 S.W.2d 332, 342 (Mo. banc 1997); State v. Steger, 209 S.W.3d 11, 18 (Mo.App.2006). As noted, this basically was a "he said-she said" credibility case, with no confession, no eyewitnesses, and no physical evidence. Although "overwhelming evidence" is difficult to define, there must be no reasonable doubt that the defendant committed the crime, and the degree of prejudice from the inadmissible evidence must be insubstantial. Dexter, 954 S.W.2d at 342.[5] The other evidence of Defendant's guilt is not "overwhelming" as our courts have used that term.
Nor can we say the improper evidence was not highlighted. Some of it came from the court's own questioning, implying judicial imprimatur. The prosecutor also assured jurors they would hear that Dr. *804 Brayfield found Child "very credible." Finally, the improper evidence was not cumulative.
Although in reviewing the sufficiency of the evidence this Court considers the evidence in the light most favorable to the judgment, we do not do so in evaluating the potential prejudice of trial error. State v. Banks, 215 S.W.3d 118, 122 (Mo. banc 2007). Having carefully considered the whole record, and the jury's inability to reach a verdict on the first evening of deliberations, we cannot say there was no reasonable probability of a different verdict without the erroneous evidence. State v. Morgenroth, 227 S.W.3d 517, 523 (Mo. App.2007). We are compelled, therefore, to reverse the judgments and remand the case for a new trial. We do not address Defendant's remaining point, which may not arise on retrial.
PARRISH, P.J., and BATES, J., concur.
NOTES
[1] We will refer to the alleged victim as "Child," and her mother as "Mother," per RSMo § 566.226 (2007 Supp).
[2] Sexual Assault Forensic Examination-Child Abuse Resource and Education.
[3] Defendant did not object to the doctor's opinions about Child's insight into lying or that she was sexually abused.
[4] "THE COURT: Do you have an opinion with a reasonable degree of medical certainty of the patients that you examine that do not show physical evidence of sexual abuse, in your opinion, have been sexually abused?"
[5] The Dexter court continued, at page 342:
[A] test to determine whether there is overwhelming evidence of guilt in a particular case is not easily articulated. Perhaps the most vivid articulation in Missouri's jurisprudence is that expressed in State v. Martin, 797 S.W.2d at 765 (Mo.App. 1990), "if [defendant] were tried one hundred times on this evidence, with or without [the detective's] testimony, she would be convicted one hundred times." (quoting State v. Smart, 756 S.W.2d 578, 582 (Mo.App. 1988) (NUGENT, J. concurring)).