SEARCH

SEARCH BY CITATION

Keywords:

  • sexual assault;
  • rape;
  • injury;
  • genital

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. Acknowledgments
  9. References

Objectives:  Forensic investigators remain unsure exactly why some sexual assault victims display acute injury while others do not. This investigation explores potential reasons for these differential findings among female victims.

Methods:  This cross-sectional analysis examined data from consecutive female sexual assault victims, at least 12 years old, who agreed to a forensic exam between November 1, 2002, and November 30, 2006. Exams utilized colposcopy, anoscopy, macrodigital imaging, and toluidine blue dye to delineate anogenital injury (AGI), which was defined as the presence of recorded anogenital abrasions, tears, or ecchymosis. Demographic variables of the victim, including sexual experience and reproductive parity, and assault characteristics were recorded in the database for bivariate and multivariate analysis with AGI.

Results:  Forty-nine percent of the initial 3,356 patients displayed AGI. Of this total, 2,879 cases included complete data for all variables and were included in the multivariate logistic regression model. A statistically significant increased risk for AGI was noted with: educational status (odds ratio [OR] 1.53, 95% CI = 1.25 to 1.87); vaginal or attempted penetration using penis (OR 2.29, 95% CI = 1.74 to 3.01), finger (OR 1.61, 95% CI = 1.88 to 1.94), or object (OR 3.19, 95% CI = 1.52 to 6.68); anal–penile penetration (OR 2.00, 95% CI = 1.57 to 2.54); alcohol involvement (OR 1.25, 95% CI = 1.04 to 1.50); and virgin status of victim (OR 1.38, 95% CI = 1.11 to 1.71). Victims were less likely to display AGI with a longer postcoital interval (OR 0.50, 95% CI = 0.39 to 0.65) and increased parity (OR 0.76, 95% CI = 0.57 to 0.99).

Conclusions:  Approximately half the patients displayed AGI. This rate is higher than earlier studies, but consistent with current investigations utilizing similar injury detection methods. The correlates of injury found reinforce the findings of prior studies, while prompting questions for future study.

Before the mid-1990s, emergency physicians (EPs) provided acute care for sexual assault victims and performed forensic examinations when indicated. In the past 15 years, our country has seen the expansive growth of dedicated sexual assault response teams (SARTs), which now provide much of this acute care. However, despite the formation of SARTs, more than half of EPs still perform at least one sexual assault exam yearly.1 Given the steady yearly incidence of approximately 200,000 reported sexual assaults in the United States, EPs will need to continue to be familiar with the topic and process of caring for such patients in the emergency department (ED).2

A compassionate and complete ED exam is but one element in combating sexual assault, which also involves action at many stages, from early prevention to the successful prosecution of perpetrators. To this end, forensic evidence has become crucial in many instances for successful adjudication. In particular, courts have scrutinized examiner documentation of anogenital injury (AGI), given the wide variability of these injury detection rates following reported sexual assault. Indeed, reported rates in the literature range from 9% to 68%.3–6 Some of this variability may be attributable to differences in the tools used to detect signs of injury, such as use of toluidine blue dye, colposcopy, both, or neither by examiners; further variability may result from differing examiners.7

While early studies failed to show a correlation between criminal prosecution and incidence of genital injury,8 more recent studies have shown higher rates of successful prosecution with documented evidence of genital injury.9,10 Such conflicting results may lead to confusion among clinicians as to the most efficient use of their time when treating victims of suspected assault (e.g., evidence collection vs. time spent counseling).The conflicting results may also affect the type and presentation of evidence prosecutors offer to a jury.

While the characteristics of the assault and the presence or absence of genital injury have been detailed in the literature,11–13 the correlates of genital injury have yet to be conclusively described. The purpose of the present study is to quantitatively analyze correlates of genital injury following suspected sexual assault.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. Acknowledgments
  9. References

Study Design

This retrospective cross-sectional analysis explores correlates of AGI in estrogenized females identified in our earlier pilot study.14 We received an exemption from patient informed consent from the Office for Protection of Research Subjects because our computerized data were devoid of patient identifiers.

Study Setting and Population

The study population consisted of consecutive female patients older than 12 years of age who contacted local law enforcement and agreed to a forensic exam between November 1, 2002, and November 30, 2006. We excluded male victims, victims less than 12 years of age, and victims who did not consent to genital examination.

Study Protocol

Twelve forensic nurses performed all of the forensic examinations using the standardized protocol recommended by the state of California. The nurses undergo extensive preemployment training, including 40 hours of didactic specialist instruction and 40–80 hours of practical clinical mentoring. This training is consistent with the educational standards of the Office of Emergency Services in California and the International Association of Forensic Nurses, for adult and adolescent sexual assault nurse examiners. Continuous case review by the supervisors maintains uniform examination technique and quality assurance. The nurses performed examinations in one of six designated and specially prepared hospital sites separate from the ED, using either colposcopy, or digital macrovisualization and toluidine blue dye application to detect injury. After completion of the exam and patient discharge, nurses entered database variables using a specific software database management program, TACT (Thorough Assault Case Tracking, Infosys Business Solutions, http://www.infosysbiz.com/index.html). TACT is a proprietary for-profit database program that manages casework, produces reports, and generates statistics for sexual assault/domestic violence cases. This software program system was purchased by one of the authors (MW) for her specific case management.

Through arrangement with law enforcement, our forensic nurse team performed all adolescent and adult sexual assault examinations authorized by the local police departments during this time period. The forensic sexual assault examiners who performed the data collection and entry from original examination information were blinded to the study hypothesis.

Study Variables.  The dependent variable in this study was the dichotomous outcome of AGI or no injury, defined as the presence or absence of recorded abrasions, tears, or ecchymosis. Examinations utilized colposcopy (or digital macrovisualization), toluidine blue dye and, when indicated, anoscopy to delineate injury. Data included in our electronic database are separate from data recorded on official state form, include only the presence or absence of injury findings, and lack specific descriptors about injury size and severity.

Our database provided variables on patient demographics and characteristics related to the reported sexual assault. Demographic variables included victim age (less than 21, 21–30, 31–40, 41+ years); race (white, Hispanic, African American, Asian/other); marital status (single, married/cohabitating, divorced/separated/widowed); educational attainment (less than high school, high school graduate, college graduate); virgin status; and number of previous pregnancies (gravida), deliveries, and children (para). Data on characteristics related to the assaults included reported vaginal penetration or attempted penetration with penis, with finger, or with objects; anal penetration or attempted penetration with penis, with finger, or with objects; any alcohol or drugs used by victim; lubrication used; weapon involvement; patient’s loss of consciousness; and number of hours between the assault and the exam or postcoital interval (0–24, 25–48, more than 48 hours).

Penetration variables had an “unknown” response category when a patient’s statement indicated that she was unable to recall if a specific act took place. Many patients chose this response because they were unconsciousness or too incapacitated to remember a particular sexual assault. In other cases the patient was fully conscious but in the process of an extremely frightening and violent assault and might have been unable to tell if penetration was complete or where the attempted penetration was located (anus or vagina).

Data Analysis

Descriptive statistics with relative frequencies were determined for all the variables including the dependent variable that would give an estimate for the prevalence of AGI among this sample. Furthermore, the frequency and proportion of AGI was determined for each of the independent variables to examine bivariate associations. We used the Pearson chi-square test for independence for bivariate analysis between AGI and the selected independent variable. For dichotomous independent variables with a “yes” or “no” response, the “no” response represented the reference category for those variables. For categorical data variables, reference categories were selected based upon categories with a large enough sample that would allow for stable statistical estimates. For example, “high school graduate” was considered the reference category for educational attainment because of the large sample size compared to “college graduate” and “less than high school.”

The multivariate logistic regression model used is a prediction model where all variables are simultaneously entered into the model to determine possible predictors for the dependent variable AGI. Therefore, all variables, regardless of previous bivariate analysis outcomes, were included in the model. One variable, lubricant use, was missing more than one-third of its data observations and was not included in the multivariate logistic model. Two other variables, para and gravida number, were not included in the model due to the strong correlation with delivery number. Interaction terms are typically not included with this type of prediction model and, given the total number of variables in the model, would not be feasible for estimation. With this type of model, the direction of a variable’s effect is more of interest than the magnitude of the effect, that is, whether there is a positive or negative relationship with the variable and AGI, controlling for all other variables in the model. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for this regression model with corresponding p-values.

Microsoft Excel (Microsoft Corp., Redmond, WA) was used for data entry and management. Descriptive statistics and bivariate and multivariate analyses were computed using SPSS 14.0 statistical software (SPSS Inc., Chicago, IL.).

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. Acknowledgments
  9. References

A total of 3,356 cases met inclusion criteria over the 4-year and 1-month period (Table 1). Of these cases, AGI was present in nearly half (49.9%), with 45.0% of victims displaying vaginal injuries and 11.4% displaying anal injuries. Female victims were typically single (77.5%) and high school graduates (66.5%), and ranged in age from 12 to 92 years, with approximately half between 11 and 20 years of age. Whites comprised the largest proportion (39.6%), followed by Hispanics (37.7%). Reported penetration prevalence ranged from 0.7% for anal penetration with an object to 67.9% for vaginal penetration with penis. Alcohol was involved in 45.4% of the cases.

Table 1.   Demographics and Characteristics of Sample: Proportion with Anogenital Injury (AGI), November 2002–November 2006 (n = 3,356)
 Frequency%Frequency (%) with AGIp-Value
Anogenital injury
 Yes1,67349.9  
 No1,68350.1  
Victim age (yr)
 Less than 211,67449.9885 (52.9)0.007
 21–3092927.7435 (46.8)
 31–4041512.4193 (46.5)
 41+33810.1160 (47.3)
Victim race (n = 3,354)
 White1,32939.6703 (52.9)0.017
 Hispanic1,26637.7616 (48.7)
 African American55816.6253 (45.3)
 Asian/other2016.099 (49.3)
Educational status (n = 3,091)
 College graduate67121.7384 (57.2)< 0.001
 High school graduate2,05466.5997 (48.5)
 Less than high school36611.8195 (53.3)
Marital status (n = 3,265)
 Single2,53077.51,297 (51.3)0.012
 Married/cohabitating2888.8121 (42.0)
 Divorced/separated/widowed44713.7224 (50.1)
Weapon involved (n = 3,139)
 Yes48915.6229 (46.8)0.177
 No2,65084.41,329 (50.2)
Vaginal penetration with penis
 Yes/attempted2,28067.91,190 (52.2)< 0.001
 No3209.5116 (36.3)
 Unknown75622.5367 (48.5)
Vaginal penetration with finger
 Yes/attempted1,08532.3603 (55.6)< 0.001
 No1,28438.3564 (43.9)
 Unknown98729.4506 (51.3)
Vaginal penetration with object
 Yes/attempted591.839 (66.1)0.024
 No2,26667.51,109 (48.9)
 Unknown1,03130.7525 (50.9)
Anal penetration with penis
 Yes/attempted47714.2299 (62.7)< 0.001
 No1,93157.5886 (45.9)
 Unknown94828.2488 (51.5)
Anal penetration with finger
 Yes/attempted2126.3135 (63.7)< 0.001
 No2,15164.11,025 (47.7)
 Unknown99329.6513 (51.7)
Anal penetration with object
 Yes/attempted250.713 (52.0)0.505
 No2,34870.01,155 (49.2)
 Unknown98329.3505 (51.4)
Alcohol involvement
 Yes1,52445.4824 (54.1)< 0.001
 No1,83254.6849 (46.3)
Drug involvement
 Yes84425.1425 (50.4)0.735
 No2,51274.91,248 (49.7)
Victim lose consciousness (n = 3,354)
 Yes1,01130.1529 (52.3)0.06
 No2,34369.91,143 (48.8)
Postcoital interval (n = 3,349), hr
 0–242,51575.11,292 (51.4)< 0.001
 25–4847614.2239 (50.2)
 More than 4835810.7137 (38.3)
Victim was virgin
 Yes64119.1349 (54.4)0.01
 No2,71580.91,324 (48.8)
Lubricant used (n = 2,232)
 Yes1556.999 (63.9)< 0.001
 No2,07793.1996 (48.0)
Delivery number (n = 3,350)
 02,54475.91,312 (51.6)0.003
 13289.8154 (47.0)
 2 or more47814.3204 (42.7)
Para number (n = 3,350)
 02,46173.51,277 (51.9)0.001
 134810.4162 (46.6)
 2 or more54116.1232 (42.9)
Gravida number (n = 3,350)
 02,33269.61,224 (52.5)< 0.001
 137911.3171 (45.1)
 2 or more63919.1276 (43.2)

Bivariate analyses found the following statistically significant high-risk variables associated with AGI: educational status at college level or higher (p < 0.001); vaginal penetration or attempted penetration with penis (p < 0.001), finger (p < 0.001), or object (p = 0.024); anal penetration with penis (p < 0.001) or finger (p < 0.001); alcohol involvement (p < 0.001); virgin status of victim (p = 0.01); and use of lubricant (p < 0.001). Variables at low risk for AGI included victims aged 21 to 40 years compared with victims less than 21 years of age (p < 0.001); Hispanic or African American race (vs. white; p = 0.017); married/cohabitating status (vs. single; p = 0.012); postcoital interval greater than 48 hours (p < 0.001); and increased parity number (two or more; p = 0.001), gravida number (one, two, or more; p < 0.001), and delivery number (two or more; p = 0.003).

Of the total number of cases, 2,879 (85.8%) included complete data for all variables under consideration and were included in the multivariate logistic regression model (Table 2). The Hosmer-Lemeshow goodness-of-fit test yielded a chi-square result of 10.9, with 8 degrees of freedom and a p-value of 0.208, indicating support for the null hypothesis that the model fits. Adjusting for variables in the model revealed a statistically significant increased risk for AGI with educational status of college graduate (OR 1.53, 95% CI = 1.25 to 1.87); vaginal penetration or attempted penetration with penis (OR 2.29, 95% CI = 1.74 to 3.01), finger (OR 1.61, 95% CI = 1.88 to 1.94), or object (OR 3.19, 95% CI = 1.52 to 6.68); unknown penetration with penis, per victim recall (OR 1.59, 95% CI = 1.07 to 2.37); anal penetration or attempted penetration with penis (OR 2.00, 95% CI = 1.57 to 2.54); alcohol involvement (OR 1.25, 95% CI = 1.04 to 1.50); and virgin status of victim (OR 1.38, 95% CI = 1.11 to 1.71). Postcoital interval greater than 48 hours (OR 0.50, 95% CI = 0.39 to 0.65) and increased parity of two or more (OR 0.76, 95% CI = 0.57 to 0.99) were possible protective factors for AGI evidence. The remaining variables were not statistically significant in the multivariate model.

Table 2.   Prediction Model Adjusted for Multiple Variables (n = 2,877)
 ORp-Value
  1. OR = odds ratio.

  2. *p < 0.05.

Victim age (yr)
 Less than 211.170.299
 21–300.920.543
 31–40Referent 
 41+1.050.772
Victim race
 WhiteReferent 
 Hispanic0.950.559
 African American0.840.128
 Asian/other0.830.263
Educational status
 College graduate1.53< 0.001*
 High school graduateReferent 
 Less than high school1.190.184
Marital status
 SingleReferent 
 Married/cohabitating0.790.140
 Divorced/separated/widowed1.080.544
Weapon involved
 Yes0.890.299
 NoReferent 
Vaginal penetration with penis
 Yes/attempted2.29< 0.001*
 NoReferent 
 Unknown1.590.022*
Vaginal penetration with finger
 Yes/attempted1.61< 0.001*
 NoReferent 
 Unknown1.550.014*
Vaginal penetration with object
 Yes/attempted3.190.002*
 NoReferent 
 Unknown0.810.330
Anal penetration with penis
 Yes/attempted2.00< 0.001*
 NoReferent 
 Unknown1.430.226
Anal penetration with finger
 Yes/attempted1.340.101
 NoReferent 
 Unknown1.120.715
Anal penetration with object
 Yes/attempted0.470.162
 NoReferent 
 Unknown0.780.433
Alcohol involvement
 Yes1.250.015*
 NoReferent 
Drug involvement
 Yes0.920.402
 NoReferent 
Victim lose consciousness
 Yes0.990.973
 NoReferent 
Postcoital interval (hr)
 0–24Referent 
 25–480.810.069
 More than 480.50< 0.001*
Victim was virgin
 Yes1.380.003*
 NoReferent 
Delivery number
 0Referent 
 10.920.568
 2 or more0.760.049*

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. Acknowledgments
  9. References

In this retrospective analysis of suspected sexual assault victims, 49.9% of victims displayed AGI detectable through the use of toluidine blue dye, colposcopy, and/or gross inspection. The present rate is higher than previously reported rates by Cartwright3 and Bowyer and Dalton,4 but is in line with more recent studies by Riggs et al.11 and particularly Grossin et al.,12 which likely reflect the improvements made in AGI detection over the past two decades. However, it must be noted that more recent studies fail to clearly delineate the methods by which AGI is detected (i.e., use of colposcopy) and may potentially prevent accurate comparison of results.

Seven of the 18 adjusted variables analyzed in the present study are noted to correlate with a statistically significantly higher risk of genital injury. As expected, direct assault on vaginal tissue by the use of penis, finger, or object correlated highly with evidence of injury. The positive correlation between physical penetration of the vagina and the presence of injury makes intuitive sense, given the often violent nature of sexual assault and the force perpetrators often exert in the execution of their crime. Penetration with an object in particular would be expected to correlate with a significantly higher risk of injury when one considers the variety of potential implements and high probability for tears or abrasions with hard, sharp, or structurally unrelenting extracorporeal objects. A recent study by Rossman et al.15 noted vaginal penetration with fingers in 28% of suspected sexual assault victims versus 32.3% in our study; 5.6% of total victims in their analysis had exclusively digital contact. Looking solely at these individuals, they reported detection of genital trauma in 81% of victims, which differs significantly from the approximately 55.6% detection rate we observed examining this variable alone. Possible explanations for this discrepancy may reflect interrater variability or some other unintended detection bias between the examiner groups, despite the fact that both utilized colposcopy to assist in examination. Additionally, one would not intuitively expect the rate of genital injury to be increased with digital penetration alone, given the smaller dimensions of the digits. Both our study and the study by Rossman et al. may lack external validity with regard to digital penetration and injury. Populationwide, digital penetration most likely fails to leave genital injury in the majority of victims; however, it is likely that those injured from this act are more likely to report and be included in our respective analysis, leading to various degrees of incorporation bias.

An interesting finding of our study is the statistically significant increased incidence of injury in victims who were unable to recall direct penetration by either penis or finger. Recall of specific acts that occurred during the assault may be altered or blurred for numerous reasons, including shock, intoxication, or loss of consciousness, but the cause and effect of this association has not been previously described. In our study, the majority of victims who are unsure about penetration lack this awareness due to drug or alcohol intoxication, and this intoxication was most often secondary to covert date-rape drug administration by the perpetrator. Perhaps perpetrators must use increased force to complete the act of rape with an unconscious victim and possibly struggle with optimal positioning for penetration. Alternatively, victims in a semiconscious state may struggle during rape, but lack recall of this. Further studies are needed to explore this complex interplay and its role in likelihood of detecting AGI.

As noted above, anal penetration with the penis was also found to have a positive correlation with evidence of physical AGI. Anal penetration by fingers or objects did not reach statistical significance, which is likely due to the relatively infrequent occurrence of those acts in our study population. Studies examining injury to the rectum and anus have been less prevalent, although some have reported low to moderate levels of anal injuries following sexual assault.6,12 The reasons for the lack of comparable data vary and may be related to the common perception of vaginal penetration as primary act of rape; however, in our present study, 596 victims (17.8%) with available data pertaining to anal involvement reported some attempt at anal penetration by a penis, finger, or object. While standard rape questionnaires are designed to assess for possible anal and rectal injuries, EPs and other clinicians who lack specific forensic training may feel less comfortable properly examining for and documenting visible injuries to these tissues, and indeed may lack the appropriate skills to do so successfully. This may represent an area of medical service for sexually assaulted patients that needs to be addressed further.

Other factors in our study that correlate with a greater probability of AGI include virgin status, higher education level, and alcohol involvement. Victims who had not had intercourse would be presumed to have more injuries following a sexual assault, not only from the often violent nature of assault, but presumably also from having tissues not previously exposed to stretching from consensual intercourse. A study by White and McLean16 specifically examining rates of injury findings in adolescent virgins versus nonvirgins following sexual assault seems to support our present finding in that 53% of the virgin group were found to have evidence of injury versus 32% in the nonvirgin group. However, with the exception of increased hymeneal injuries in the virgin group, the rates of injury to the vulva, posterior forchette, and vagina were almost identical. The estrogenized hymen is redundant and elastic in nature and most often lacks visible signs of injury after first intercourse. Even prepubertal girls without prior sexual experience often fail to display evidence of genital injury following sexual assault.17 Our present study did not specifically break down what types of injury are more common in female virgins; therefore, this result must be interpreted with caution. Victims with college graduate status or above more often presented with evidence of AGI than those achieving a high school–level education or less. We previously postulated that this might be due in part to higher-level education acting as a proxy for fewer previous sexual experiences.14 Other possible factors creating this disparity include earlier reporting within the 48-hour time frame by more educated victims as a result of presumed increased awareness or increased struggle during the assault. Interestingly, alcohol involvement correlated positively with the presence of injury; one might expect victims who had been inebriated to be less resistant to the assault and therefore display less evidence of injury, although this was not found to be the case in our present study.

Two variables with a significant negative association with the presence of AGI were postcoital interval greater than 48 hours and increased parity. Our findings support previous findings by Slaughter et al.,6 who noted a higher incidence of genital injury when the victim presented less than 24 hours after the assault when compared to those who presented up to 48 hours and greater than 72 hours after the assault. Sugar et al.13 also noted a similar trend. Grossin et al.12 used 72 hours as an alternative time interval following suspected sexual assault, and noted a nearly 50% decrease in the number of victims displaying evidence of genital trauma. Additionally, women with a previous history of pregnancy and childbirth would be expected to have lower rates of AGI following assault, for reasons similar to the finding that virgins exhibited higher rates of injury on exam. Although we did not separate the data for those who had experienced vaginal childbirth versus Cesarean delivery, it is clear the female genital tract undergoes significant changes during pregnancy that may confer different properties on the tissue and result in less evidence of trauma following sexual assault.

Knowledge of this decreasing likelihood of visible evidence of AGI is important both for forensic specialists and EPs treating sexually assaulted patients to anticipate the likelihood of finding injury. Understanding of genital injury correlates and patterns may also aid EPs who discover genital injuries in a patient who has not yet disclosed a sexual assault. Additionally, law enforcement agencies and prosecutors armed with this information may construct more appropriate cases to present to juries who may be skeptical when tangible evidence is lacking.

Limitations

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. Acknowledgments
  9. References

Limitations of this analysis include reliance on self-reporting of the initial assault. A small subset of patients may not have been sexually assaulted, and the inclusion of their responses in our analysis might skew the results and underestimate the true prevalence of AGI. Certain variables were sufficiently infrequent to precluded reaching statistical significance, but may prove to be associated in a larger population sample. Selection bias is evident when excluding 14.2% of the observations from multivariate logistic regression due to missing data, which showed significant differences in those data compared to complete case observations in nearly all the independent variables (Table 3). Last, experienced forensic nurses performed all examinations using advanced equipment and techniques, which may not be available to the average emergency practitioner, limiting the external validity of our results.

Table 3.   Comparison of Complete Data Cases (n = 2,879) vs. Missing Data Cases (n = 477)
 Complete CasesMissing Data Casesp-Value
Frequency%Frequency%
Anogenital injury
 Yes1,45750.621645.30.031
 No1,42249.426154.7
Victim age (yr)
 Less than 211,54653.712826.8< 0.001
 21–3076526.616434.4
 31–4032711.48818.4
 41+2418.49720.3
Victim race (n = 3,354)
 White1,11538.721445.10.054
 Hispanic1,10738.515933.5
 African American48016.77816.4
 Asian/other1776.1245.1
Educational status (n = 3,091)
 College graduate60821.16329.70.001
 High school graduate1,91866.613664.2
 Less than high school35312.3136.1
Marital status (n = 3,265)
 Single2,26478.626668.9< 0.001
 Married/cohabitating2277.96115.8
 Divorced/separated/widowed38813.55915.3
Weapon involved (n = 3,139)
 Yes44115.34818.50.181
 No2,43884.721281.5
Vaginal penetration with penis
 Yes/attempted2,03270.624852.0< 0.001
 No2849.9367.5
 Unknown56319.619340.5
Vaginal penetration with finger
 Yes/attempted98634.29920.8< 0.001
 No1,14339.714129.6
 Unknown75026.123749.7
Vaginal penetration with object
 Yes/attempted491.7102.1< 0.001
 No2,04371.022346.8
 Unknown78727.324451.2
Anal penetration with penis
 Yes/Attempted42614.85110.7< 0.001
 No1,73460.219741.3
 Unknown71925.022948.0
Anal penetration with finger
 Yes/attempted1956.8173.6< 0.001
 No1,93167.122046.1
 Unknown75326.224050.3
Anal penetration with object
 Yes/attempted220.830.6< 0.001
 No2,10973.323950.1
 Unknown74826.023549.3
Alcohol involvement
 Yes1,28044.524451.20.007
 No1,59955.523348.8
Drug involvement
 Yes68623.815833.1< 0.001
 No2,19376.231966.9
Victim lose consciousness (n = 3,354)
 Yes78227.222948.2< 0.001
 No2,09772.824651.8
Postcoital interval (n = 3,349), hr
 0–242,17375.534272.80.419
 25–4840113.97516.0
 More than 4830510.65311.3
Victim was virgin
 Yes57620.06513.60.001
 No2,30380.041286.4
Lubricant used (n = 2,232)
 Yes1386.9177.40.793
 No1,86393.121492.6
Delivery number (n = 3,350)
 02,19776.434773.4< 0.001
 12849.9449.3
 2 or more39613.88217.3
Para number (n = 3,350)
 02,12673.933570.80.001
 129910.44910.4
 2 or more45215.78918.8
Gravida number (n = 3,350)
 02,01269.932067.70.001
 133011.54910.4
 2 or more53518.610422.0

Conclusions

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. Acknowledgments
  9. References

Overall, the rates of AGI following suspected sexual assault remain in line with previous descriptions. This retrospective analysis helps to reinforce certain trends that have been noted by other authors, but raises other questions. Specifically, the positive correlation between vaginal and anal penetration remain strong predictors for injury, as does higher educational status and possibly alcohol involvement. The negative association between prolonged times from suspected sexual assault to evaluation is further supported by our present study. Additionally, a higher number of previous pregnancies and deliveries appear to be negatively correlated with evidence of AGI injury.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. Acknowledgments
  9. References

The authors thank the staff of FNS and S. Wignall for their dedication and contribution to this work.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. Acknowledgments
  9. References
  • 1
    Tibbles C, Frost E, Fisher J. Performance of sexual assault evidence collection by emergency physicians: is it still necessary to train emergency medicine residents? [abstract]. Acad Emerg Med. 2007; 14(5 Suppl. 1): S109.
  • 2
    Catalano SM. Criminal Victimization, 2003. National Crime Victimization Survey (NCJ 205455). Washington, DC: Bureau of Justice Statistics, U.S. Department of Justice, 2004.
  • 3
    Cartwright PS. Factors that correlate with injury sustained by survivors of sexual assault. Obstet Gynecol. 1987; 70:446.
  • 4
    Bowyer L, Dalton ME. Female victims of rape and their genital injuries. Br J Obstet Gynaecol. 1997; 104:61720.
  • 5
    Rambow B, Adkinson C, Frost TH, Peterson GF. Female sexual assault: medical and legal implications. Ann Emerg Med. 1992; 21:72731.
  • 6
    Slaughter L, Brown CR, Crowley S, Peck R. Patterns of genital injury in female sexual assault victims. Obstet Gyencol. 1997; 176:60916.
  • 7
    Eckert LO, Sugar N, Fine D. Factors impacting injury documentation after sexual assault: Role of examiner experience and gender. Am J Obstet Gynecol. 2004; 190:173946.
  • 8
    McGregor MJ, Le G, Marion SA, Wiebe E. Examination for sexual assault: is the documentation of physical injury associated with the laying of charges? A retrospective cohort study. CMAJ. 1999; 160:15659.
  • 9
    Gray-Eurom K, Seaberg DC, Wears RL. The prosecution of sexual assault cases: correlation with forensic evidence. Ann Emerg Med. 2002; 39:3946.
  • 10
    McGregor MJ, Du Mont J, Myhr TL. Sexual assault forensic medical examination: is evidence related to successful prosecution? Ann Emerg Med. 2002; 39:63947.
  • 11
    Riggs N, Houry D, Long G, Markovchick V, Feldhaus KM. Analysis of 1,076 cases of sexual assault. Ann Emerg Med. 2000; 35:35862.
  • 12
    Grossin C, Sibille I, Lorin de la Grandmaison G, Banasr A, Brion F, Durigon M. Analysis of 418 cases of sexual assault. Forensic Sci Int. 2003; 131:12530.
  • 13
    Sugar NF, Fine DN, Eckert LO. Physical injury after sexual assault: findings of a large case series. Am J Obstet Gyencol. 2004; 190:716.
  • 14
    Sachs CJ, Chu LD. Predictors of genitorectal injury in female victims of suspected sexual assault. Acad Emerg Med. 2002; 9:14651.
  • 15
    Rossman L, Jones JS, Dunnuck C, Wynn BN, Bermingham M. Genital trauma associated with forced digital penetration. Am J Emerg Med. 2004; 22:1014.
  • 16
    White C, McLean I. Adolescent complainants of sexual assault; injury patterns in virgin and non-virgin groups. J Clin Forensic Med. 2006; 13:17280.
  • 17
    Adams JA, Harper K, Knudson S, Revilla J. Examination findings in legally confirmed child sexual abuse: it’s normal to be normal. Pediatrics. 1994; 94:3107.