OBJECTIVE: To describe forensic evidence findings and reevaluate previous recommendations with respect to timing of evidence collection in acute child sexual assault and to identify factors associated with yield of DNA.
METHODS: This was a retrospective review of medical and legal records of patients aged 0 to 20 years who required forensic evidence collection.
RESULTS: Ninety-seven of 388 (25%) processed evidence-collection kits were positive and 63 (65%) of them produced identifiable DNA. There were 20 positive samples obtained from children younger than 10 years; 17 of these samples were obtained from children seen within 24 hours of the assault. Three children had positive body samples beyond 24 hours after the assault, including 1 child positive for salivary amylase in the underwear and on the thighs 54 hours after the assault. DNA was found in 11 children aged younger than 10 years, including the child seen 54 hours after the assault. Collection of evidence within 24 hours of the assault was identified as an independent predictor of DNA detection.
CONCLUSIONS: Identifiable DNA was collected from a child's body despite cases in which: evidence collection was performed >24 hours beyond the assault; the child had a normal/nonacute anogenital examination; there was no reported history of ejaculation; and the victim had bathed and/or changed clothes before evidence collection. Failure to conduct evidence collection on prepubertal children beyond 24 hours after the assault will result in rare missed opportunities to identify forensic evidence, including identification of DNA.
WHAT'S KNOWN ON THIS SUBJECT:
The American Academy of Pediatrics in 2005 recommended that body swabs collected in prepubertal children >24 hours after a sexual assault are unlikely to yield forensic evidence. Since that time, additional studies have supported this recommendation.
WHAT THIS STUDY ADDS:
This study addresses the identification of DNA in acute child sexual assault. Failing to conduct evidence collection on prepubertal children beyond 24 hours will result in missed opportunities to identify additional forensic evidence, including identification of DNA.
Forensic evidence collection is often an important component of the evaluation of acute child sexual assault (ACSA).1,–,3 In an effort to best utilize financial and personnel resources, to limit potential anxiety of the child and caregiver, and to maximize the yield of forensic evidence collection, it is important for the provider to understand the clinical indications for forensic evidence collection.
Original guidelines from the American Academy of Pediatrics (AAP) endorsed conducting forensic evidence collection on all children within 72 hours of a reported sexual assault.4 Several studies, however, have since challenged the benefit of applying these guidelines uniformly to young children. Christian et al5 first described the epidemiology of forensic evidence findings in prepubertal victims of sexual assault. In this retrospective review of 273 children aged younger than 10 years, some type of forensic evidence (semen, sperm, blood, saliva, hair, or foreign debris) was found in 24.9% of children, with >90% of findings retrieved within 24 hours of the assault. The majority of forensic evidence (64%) was found on clothing or linens. After 24 hours, all forensic evidence—with the exception of evidence involving 1 child on whom a pubic hair was found 44 hours after the assault—was recovered from clothing or linens. The authors suggested that swabbing a prepubertal child's body for evidence is unnecessary >24 hours after the assault. To reflect these data, the AAP updated their recommendations in 2005 to read “Body swabs collected in prepubertal children >24 hours after a sexual assault are unlikely to yield forensic evidence.”2 Since revision of the AAP recommendations, additional studies have supported limiting forensic evidence collection in prepubertal children to the first 24 hours after the assault. Young et al6 reported the utility of forensic evidence collection in children younger than 12 years. Forty-nine children in this study were evaluated within 72 hours of a reported sexual assault. Only 3 children had semen recovered, and all 3 samples were from clothing or linens. Palusci et al7 also investigated the yield of forensic evidence collection in a population of children younger than 13 years. Of the cohort aged younger than 10 years, no child had positive forensic test results collected from a body site if examined ≥24 hours after the assault.
None of these studies specifically addressed isolation and identification of DNA as part of the forensic evaluation of ACSA, regardless of the child's age. DNA technology continues to advance rapidly, including the use of Y-chromosome markers to detect trace amounts of male DNA on vaginal and anal samples8 and ongoing expansion of the Combined DNA Index System.9 As these technologies advance, there may be an increasing role for DNA analysis in the context of ACSA. The objectives of the present study were to reevaluate previous recommendations with respect to timing of evidence collection in ACSA, to describe the epidemiology of forensic evidence findings in ACSA, and to identify factors associated with yield of DNA. We hypothesized that the yield of forensic evidence with current DNA technology may be greater than previously reported and may be retrieved beyond 24 hours after the assault.
Patient Population/Record Abstraction
A retrospective review of medical and legal records of children 0 to 20 years of age who had forensic evidence collection done at Nationwide Children's Hospital (Columbus, OH) between January 1, 2004, and December 31, 2007, was undertaken. Forensic evidence analysis was conducted either by the Columbus Police Department Crime Laboratory (Columbus, OH) or by the Ohio Bureau of Criminal Identification and Investigation (London, OH) Data were abstracted for demographic characteristics, disclosure of abuse, physical examination findings, forensic evidence collected, and results of forensic evidence analysis. In cases in which an exact number of hours between assault and evidence collection could not be determined, the time since assault was based on review of medical and legal documentation and determination of dates (ie, a difference of 2 days between assault and evidence collection was coded as a time since assault of <48 hours). This study was approved by the institutional review board of Nationwide Children's Hospital, the Columbus Police Department, and the Ohio Bureau of Criminal Identification and Investigation.
Medical examinations were completed in the emergency department or the Center for Child and Family Advocacy at Nationwide Children's Hospital. Examinations were performed by pediatric emergency physicians (including residents working with attending physicians), child abuse pediatricians, and pediatric sexual assault nurse examiners. Determination of testing and treatment for sexually transmitted infections was at the provider's discretion.
Forensic evidence was collected according to protocol as described in the Ohio Child and Adolescent Sexual Abuse Protocol available from the Ohio Department of Health.10 It is standard practice at our institution to conduct forensic evidence collection on any child who discloses an episode of sexual abuse/assault that has occurred within the past 72 hours and the history indicates (1) contact with the alleged perpetrator's genitalia, semen, blood, or saliva, (2) a struggle that may have left skin or blood of the alleged perpetrator on the victim's body, or (3) that the victim's clothing or body may be covered by trace evidence. Forensic evidence may also be collected when the history is unclear and there is reason to suspect that any of these conditions may be possible.
Chain of custody was maintained, and evidence-collection kits were signed out to the appropriate law enforcement jurisdiction.
Depending on the jurisdiction, evidence was analyzed at either 1 of 2 crime laboratories. Procedures for evidence analysis in both laboratories relied on similar chemical analysis techniques.
Presumptive testing for semen was conducted using the acid phosphatase spot test.11 One laboratory performed confirmatory testing regardless of presumptive testing results and used microscopic evaluation for spermatozoa of smears or extracts after staining.12 In the absence of observable spermatozoa, testing for prostate-specific antigen was conducted using monoclonal anti-human prostate-specific antigen antibody.13 The other laboratory performed confirmatory testing using only the prostate-specific antigen method. Presumptive testing for saliva was accomplished using amylase as a marker.14 Presumptive testing for blood was determined using the tetramethylbenzidine test.15 In 1 laboratory, these test reagents were calibrated against standardized minimum sensitivity controls.16
In addition to standard DNA analysis of body fluid stains, random sampling from blind swabbings and clothing/linen cuttings was conducted on the basis of details learned during the investigation. These samples were subjected to DNA testing on a case-by-case basis.
Descriptive statistics of the study population are reported. Univariate and multivariate analyses were conducted to determine associations between various clinical and demographic factors and evidence-collection findings. Intercooled Stata 9.2 (Stata Corp, College Station, TX) was used for all statistical analyses. A P value of <.05 was considered statistically significant.
Characteristics of the sexual assaults are listed in Table 1. A total of 516 children had evidence collection performed between January 1, 2004, and December 31, 2007. Of these 516 collections, 388 (75%) were submitted to a crime laboratory by the law enforcement agencies. Stated reasons for failure to submit evidence-collection kits for processing include an unfounded allegation (40%), perpetrator confession (23%), law enforcement refusing to investigate (10%), family not cooperating with legal investigation (7%), recantation of the allegation (7%), and missing perpetrator (2%). Eighty-eight percent of kits processed were collected from females, 32% were children younger than 10 years, and 21% of children had acute anogenital examination findings.
The exact number of hours elapsed between the sexual assault and the evidence collection was determined in 205 (52.8%) of the children in whom evidence collection was processed; 154 (39.7%) were seen within 24 hours of the assault. An additional 63 children (16.2%) were known to be seen within 24 hours of the assault, although the exact number of hours was unknown. Forty-eight children were seen in a time frame known to be within 48 hours of the assault, and 20 children were seen in a time frame known to be within 72 hours of the assault.
There was 1 death in our study: a 1-year-old boy who presented to the emergency department with multisystem trauma that included acute sexual assault.
History of the Assault
Seventy-two percent of cases involved a “high-risk” disclosure of either genital-genital and/or anal-genital contact. In 10% of the cases, the history of assault was unknown because the child was either preverbal or had altered mental status. The alleged perpetrator of the sexual assault was known to the child in 83% of cases. In 22% of the cases, the child reported ejaculation by the perpetrator.
Physical Examination Findings
Twenty-one percent of children had acute anogenital injury at the time of evidence collection. Of the 81 cases with acute injury, 70 of these were genital injuries (99% in females) and 11 were anal injuries (55% in females). Fifty-eight genital injuries were identified within 24 hours of the assault. Acute anogenital injury was associate with a positive evidence-collection kit (odds ratio: 1.82 [95% confidence interval: 1.07–3.10]). Five children had other injuries, including 1 with bruising of the thigh, 1 with bruising of the chest, and 3 with suction-induced petechiae of the neck.
Forensic Evidence Findings
Of the 388 evidence kits processed, 97 (25%) kits were positive for ≥1 sample of semen, amylase, and/or blood (total: 106 samples). Eighty (82%) of the kits were positive for semen, 24 (25%) for amylase, and 2 (2%) for blood. Of the 97 positive kits, 63 (65%) were collected within 24 hours of the assault.
There were 20 positive samples, representing 18 positive kits, obtained from children younger than 10 years. Nine positive samples were obtained from body sites, 4 positive samples were obtained from clothing, and 7 positive samples were obtained from both body and clothing sites. Seventeen of the 20 positive samples were obtained from children seen within 24 hours of the sexual assault. Two additional children had 4 positive samples (both from body and clothing sites) within 48 hours of the assault, although the exact time since assault was unknown. One child was positive for salivary amylase in the underwear and on the thighs at 54 hours after the assault.
In children aged 10 years and older, there were 86 positive samples, representing 79 positive kits. Thirty-four positive samples were obtained from clothing sites, 32 from body sites, and 20 from both body and clothing. Fifty-three of the 86 positive samples were obtained from children seen within 24 hours of the sexual assault. An additional 20 positive samples were obtained from children seen within 48 hours of the assault, although the exact time since assault was unknown. Two children's underwear tested positive for semen at 50 and 60 hours, respectively, and 1 child's outer clothing tested positive for semen at 63 hours.
Data describing the positive sample location by age and time are shown in Table 2. Sixty-three (65%) of the 97 positive evidence-collection kits yielded identifiable DNA. DNA was found in 11 children younger than 10 years. Nine of these children were seen within 24 hours of the assault, 1 was seen at an unknown time since the assault, and 1 was seen at 54 hours after the assault. In children aged 10 years and older, there were 52 kits that yielded DNA.
Table 3 summarizes the clinical and demographic factors associated with positive evidence collection and DNA detection. Factors associated with both a positive evidence-collection kit and DNA detection on univariate analysis include a reported history of perpetrator ejaculation, a disclosure of genital-genital and/or anal-genital contact, and not having bathed and/or changed clothing before the collection. Perpetrator age older than 18 years and acute anogenital examination findings were associated with positive evidence collection but not detection of DNA. Collection of evidence performed within 24 hours of the assault was not associated with a positive evidence-collection kit but was associated with detection of DNA. Using multivariate logistic regression analyses, only a reported history of perpetrator ejaculation was independently associated with having a positive evidence-collection kit. Collection of evidence within 24 hours of the assault was identified as an independent predictor of DNA detection. These data are summarized in Table 4.
Previous studies have not addressed the isolation and identification of DNA as part of the forensic evaluation of ACSA. In this study, DNA was identified in 63 of the 97 (65%) positive evidence-collection kits. Although only collection of evidence within 24 hours of the assault was found to be an independent predictor of DNA identification in this study, we observed that identifiable DNA was collected from a child's body despite cases in which evidence collection was performed >24 hours beyond the sexual assault; the child had a normal/nonacute anogenital examination at the time of collection; there was no reported history of ejaculation; and the victim had bathed and/or changed clothes before evidence collection. We find these results challenge some of the current practices and indications for conducting evidence collection in the pediatric victim population.
Children who have been sexually assaulted will often provide an incomplete description of the sexual assault. Consistent with previous work by Christian et al,5 we found a poor correlation between the child's description of the assault and the forensic evidence findings. Nine children yielded forensic evidence findings that were not expected on the basis of the nature of the disclosure. For example, a 3-year-old girl who reported only digital-genital contact and kissing was subsequently found to have semen detected on both a vaginal swab and in her underwear. In addition to these 9 children, an additional 8 children who specifically denied perpetrator ejaculation tested positive for semen. Furthermore, 9 children tested positive for amylase with digital-anal and/or digital-genital contact and no reported oral contact. Whether these cases represent an incomplete disclosure of sexual abuse or detection of a perpetrator's saliva used to lubricate a digit is unclear.
Our data suggest that the decision to collect forensic evidence, or conducting a “targeted” evidence collection, based solely on discrete indicators such as the child's disclosure of abuse, the victim's age, or the presence or absence of anogenital findings will result in missed opportunities to identify both trace evidence and perpetrator DNA. Similarly, closing an investigation without submitting the kit to a crime laboratory risks the loss of physical evidence that may or may not support other information obtained (ie, the presence of semen where no genital contact is alleged, the presence of body fluids from other than the identified perpetrator).
We concur with the most recent recommendations of the AAP that state “Body swabs collected in prepubertal children >24 hours after a sexual assault are unlikely to yield forensic evidence.”2 It should be recognized that in our study, failure to conduct evidence collection on children younger than 10 years seen ≥24 hours after the assault would have missed identification of 3 samples of forensic evidence, including 1 sample that yielded identifiable DNA. Each of these children, however, also had positive samples obtained from clothing, which supports the need to include collection of pertinent clothing and/or linens as part of the forensic assessment. In the era of advancing DNA technology that yields greater potential for identifying forensic evidence in acute sexual assault, a reevaluation of clinical and forensic practice may be warranted.
Previous studies have failed to consider characteristics of the reporting forensic laboratories. Testing methods used, sensitivity of reagents, skill of the forensic analyst, and extent of the testing are not standardized by any best practices. In comparing the 2 forensic laboratories in this study, there was a significant difference in positive evidence-collection kits (37 of 76 in laboratory A [49%] vs 60 of 215 in laboratory B [28%]; P = .02) but not detection of DNA. Although our study was not a comparison of forensic laboratories, we do propose that technical aspects of the forensic examination may be as important as the clinical examination.
Our observation that 55% of positive samples were obtained from clothing in children younger than 10 years and 26% in older children reaffirms that the procurement and analysis of clothing remain vital in the investigation of ACSA.
There are several limitations that should be recognized when interpreting our data. First, these data were collected retrospectively, and variations in the quality and completeness of data collection exist. Second, there is a possible selection bias in the processing of evidence-collection kits, which can be difficult to measure because there are various reasons why kits may have been collected but never processed. If a given kit was collected but never processed because the perpetrator confessed, for example, that kit may be much more likely to be positive and bias the results toward a falsely lower prevalence of positive kits. In addition, the difference in laboratory prevalence of positive evidence-collection kits may have underestimated the true prevalence of positive kits. Third, Tanner staging and sexual maturity rating were not recorded, and therefore the true pubertal status of each child cannot be stated. To be consistent with previous work (eg, that of Christian et al,5 Palusci et al7), we selected a conservative age cutoff of younger than 10 years as a proxy to determine prepubertal status, recognizing that a small number of children in this group may, in fact, have started puberty. Conversely, however, it is possible that some children 10 years of age and older in this study were not yet pubertal. Finally, evidence collection was completed by a variety of providers, including pediatric emergency physicians, child abuse pediatricians, and pediatric sexual assault nurse examiners, each with varying levels of training and experience.
Forensic evidence collection is often an important component of the evaluation of ACSA and is ideally conducted under the supervision of trained physicians with appropriate medical history-gathering and physical examination techniques in a supportive environment. Although our data are consistent with the most recent recommendations of the AAP that state prepubertal children are unlikely to have body swabs positive for evidence >24 hours after a sexual assault, it is important to recognize that failing to conduct evidence collection on prepubertal children will result in missed opportunities to identify additional forensic evidence, including identification of DNA. As DNA technology advances, continued scrutiny of current clinical and forensic practice will be necessary.
We thank Sherry Curran for her assistance in data analysis.
- Accepted April 12, 2011.
- Address correspondence to Jonathan D. Thackeray, MD, Center for Child and Family Advocacy, Nationwide Children's Hospital, 655 E Livingston Ave, Columbus, OH 43205. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
COMPANION PAPERS: Companions to this article can be found on pages 221, 233, and 374 and online at www.pediatrics.org/cgi/doi/10.1542/peds.2010-3037, www.pediatrics.org/cgi/doi/10.1542/peds.2010-3288, and www.pediatrics.org/cgi/doi/10.1542/peds.2011-1455.
- ACSA —
- acute child sexual assault
- AAP —
- American Academy of Pediatrics
- Enos WF,
- Conrath TB,
- Byer JC
- Kellogg N
- De Jong AR,
- Rose M
American Academy of Pediatrics, Committee on Child Abuse and Neglect. Guidelines for the evaluation of sexual abuse in children: subject review [published correction appears in Pediatrics. 1999;103(5 pt 1):1049]. Pediatrics. 1999;103(1):186–191
- Christian CW,
- Lavelle JM,
- De Jong AR,
- Loiselle J,
- Brenner L,
- Joffe M
Federal Bureau of Investigation. CODIS: NDIS statistics. Available at: www.fbi.gov/about-us/lab/codis/ndis-statistics. Accessed October 29, 2010
Ohio Department of Health. Ohio protocol for sexual assault forensic and medical exams. Available at: www.odh.ohio.gov/odhPrograms/hprr/sadv/sadvprot.aspx. Accessed October 28, 2010
- Gaensslen RE
- Wurster JW,
- Laux DL
- Laux DL
- Copyright © 2011 by the American Academy of Pediatrics