Secondary Logo

Journal Logo

A Short Screening Tool to Identify Victims of Child Sex Trafficking in the Health Care Setting

Greenbaum, V., Jordan, MD*; Dodd, Martha, DNP, FNP-BC*; McCracken, Courtney, PhD

doi: 10.1097/PEC.0000000000000602
Original Articles

Objectives The aim of this study was to describe characteristics of commercial sexual exploitation of children/child sex trafficking (CSEC/CST) victims and to develop a screening tool to identify victims among a high-risk adolescent population.

Methods In this cross-sectional study, patients aged 12 to 18 years who presented to 1 of 3 metropolitan pediatric emergency departments or 1 child protection clinic and who were identified as victims of CSEC/CST were compared with similar-aged patients with allegations of acute sexual assault/sexual abuse (ASA) without evidence of CSEC/CST. The 2 groups were compared on variables related to medical and reproductive history, high-risk behavior, mental health symptoms, and injury history. After univariate analysis, a subset of candidate variables was subjected to multivariable logistic regression to identify an optimum set of 5 to 7 screening items.

Results Of 108 study participants, 25 comprised the CSEC/CST group, and 83 comprised the ASA group. Average (SD) age was 15.4 (1.8) years for CSEC/CST patients and 14.8 (1.6) years for ASA patients; 100% of the CSEC/CST and 95% of the ASA patients were female. The 2 groups differed significantly on 16 variables involving reproductive history, high-risk behavior, sexually transmitted infections, and previous experience with violence. A 6-item screen was constructed, and a cutoff score of 2 positive answers had a sensitivity of 92%, specificity of 73%, positive predictive value of 51%, and negative predictive value of 97%.

Conclusions Adolescent CSEC/CST victims differ from ASA victims without evidence of CSEC/CST across several domains. A 6-item screen effectively identifies CSEC/CST victims in a high-risk adolescent population.

From the *Children's Healthcare of Atlanta and †Emory University School of Medicine, Atlanta, GA.

Disclosure: The authors declare no conflict of interest.

Reprints: Virginia Jordan Greenbaum, MD, Stephanie V. Blank Center for Safe and Healthy Children, Children's Healthcare of Atlanta, 975 Johnson Ferry Rd, NE, Suite 350, Atlanta, GA 30342 (e-mail:

Supported by the Blank Family Foundation through funding portions of the study (portions of salary for Drs Greenbaum and Dodd), but it had no role in study design; collection, analysis, and interpretation of data; and the writing of the report or of the decision to submit the paper for publication.

Jordan Greenbaum, MD, wrote the first draft of this manuscript. Her general work in research on child maltreatment is supported by the Blank Family Foundation, and this includes the work of Marni Dodd, DNP, FNP-BC, on this project.

Human trafficking involves severe human rights violations of adults and children and may include forced labor, forced sexual activity, organ removal, or slavery.1 Although exact numbers are not known, victims have been identified from every region of the world.2 In a recent study by the United Nations, nearly 1 of every 3 reported victims was a child, with girls outnumbering boys by a ratio of 2:1.2 The commercial sexual exploitation of children (CSEC) is closely related to sex trafficking and involves “crimes of a sexual nature committed against juvenile victims for financial or other economic reasons.… These crimes include trafficking for sexual purposes, prostitution, sex tourism, mail-order-bride trade and early marriage, pornography, cyber enticement, stripping, and performing in sexual venues such as peep shows or clubs.”3 When victims are younger than 18 years, force, fraud, deception, or coercion is not required for either human trafficking or CSEC. The prevalence of child sex trafficking (CST) and CSEC in the United States is unknown. Edwards and colleagues4 found that, among a nationally representative sample of more than 13,000 US 7th to 12th graders, 3.5% admitted to exchanging sex for money or drugs. In a study of youth at a homeless shelter, nearly 15% experienced some form of human trafficking, and an additional 8% reported engaging in survival sex.5

Existing studies indicate that victims of sex trafficking experience significant physical and sexual violence, sexually transmitted infections (STIs) including human immunodeficiency virus/acquired immunodeficiency syndrome, pregnancy and unsafe abortions, pelvic inflammatory disease, malnutrition, untreated chronic medical conditions, posttraumatic stress disorder, major depression and suicidality, anxiety disorders, alcohol and drug addiction, and somatic symptoms.6–15 Research also indicated that victims do seek medical care. In one study of US women and adolescent victims of trafficking, 88% reported seeking medical care during the period of exploitation.16 A major challenge to health care provider intervention is the extreme difficulty in identifying victims. Fear of the trafficker, distrust of authorities, shame, hopelessness, trauma bonds, and other factors make victim self-identification rare. Moreover, there is very little empirical research on reliable indicators of CST or appropriate screening questions to ask adolescent patients in the medical setting. Our objective was to describe characteristics of CSEC/CST patients presenting to a large metropolitan pediatric emergency department (ED) or child protection clinic. Using these characteristics, we sought to develop a quick screening tool to accurately identify CSEC/CST victims among a high-risk adolescent population.

Back to Top | Article Outline


This was a cross-sectional study approved by the Children's Healthcare of Atlanta Institutional Review Board. A waiver of consent was granted as the information obtained from the study was collected as part of an enhanced, standardized medical history or related to standard medical observations and practices and obtained from existing medical records. Eligible patients identified as suspected victims of CSEC/CST between the ages of 12 and 18 years who presented to 1 of 3 Children's Healthcare of Atlanta's EDs or one of its child protection clinics between June 1, 2013, and April 10, 2014, were compared with youth who presented during the same period with allegations of sexual abuse or sexual assault and for whom there was no evidence to suggest CSEC/CST. All patients were evaluated by nurse practitioners or physicians trained in trauma-informed care, CSEC/CST, and sexual assault. Inclusion criteria included English-speaking male and female adolescents, aged 12 to 18 years, with concerns of CSEC/CST or sexual assault/sexual abuse. Those with extreme developmental delays, who appeared intoxicated, or who were otherwise deemed unable to answer questions accurately were excluded from the study at the discretion of the medical provider.

Eligible participants were interviewed outside the presence of the caregiver or other person accompanying the youth to the visit. Medical providers informed the patients that answering questions was entirely voluntary and advised them of the limits of confidentiality before beginning the interview. Using a trauma-informed approach, the clinician asked the patients a series of questions related to medical history, reproductive history, substance use and other high-risk behavior, mental health issues, and abuse/injury history. The clinician also recorded observations about the initial presentation of the youth, including chief complaint, patient demeanor, signs of intoxication, and the identity and behavior of the person accompanying the youth. Finally, the medical provider documented results of the physical examination and laboratory evaluation. Depending on the child's reported experiences, perceived stress level, and willingness to participate, the clinician may or may not have asked all of the approximately 50 designated questions on the history form. For example, if the youth denied previous drug use, further questions about drugs were not asked. If a patient seemed distracted and eager to end the interview, the examiner may not have asked all questions.

After the child's visit, study personnel reviewed the medical record and extracted additional data from the index visit, as well as from previous medical visits to Children's Healthcare of Atlanta facilities.

Patients were classified as “suspected CSEC/CST” (CSEC/CST) if the information obtained by the medical provider indicated a high likelihood that the youth had been the victim of such, based on the definitions of the United Nations1 and the Institute of Medicine.3 This information may have come from authorities (eg, law enforcement reported finding the youth during a raid of a site of known prostitution) or from information gleaned from the medical record, from parents or others, or from the youth. Most of the patients who presented to the EDs were brought in by law enforcement or child protective services with specific concerns of CSEC/CST. Patients were classified as “alleged acute sexual assault/sexual abuse” (ASA) based on chief complaint and/or information gleaned by the medical provider during the visit. Patients in the ASA group had no identified indication of previous or concurrent CSEC/CST. Legal or child protective service confirmation of abuse/assault or commercial sexual exploitation was not required for study eligibility.

Statistical analyses were conducted using SAS 9.3 (Cary, NC). Statistical significance was assessed at the 0.05 level unless otherwise noted. Descriptive statistics were calculated for all variables of interest and included means and standard deviations, medians and interquartile ranges, or counts and percentages, when appropriate. Furthermore, CSEC/CST patients were compared with ASA victims using independent 2-sample t tests and χ2 tests for continuous and categorical variables, respectively. When expected cell counts were small (<5), exact tests were performed.

Variables found to be significant at the 0.1 level in the univariate analyses were selected as candidate screening items. Multivariable logistic regression models were used to identify the best subset of screening items that correctly identified a CSEC/CST patient. Multiple models containing between 5 and 7 screening items were constructed, and the area under the receiver operating curve (AUROC) was calculated for each. An AUROC of 1.0 represents a perfect discriminator, and an AUROC of 0.5 represents discrimination no better than chance. A final set of screening items was selected based on (1) high AUROC value, (2) the percentage of patients who were able to answer the question, and (3) the level of difficulty in obtaining the information from the patient or medical chart. Comparison of the distribution of risk scores among groups was made using Wilcoxon rank sum tests. Validation of the final risk factor model was performed using bootstrapping techniques. One thousand bootstrap samples of size 108 were sampled from the original data set with replacement. An AUROC was calculated for each bootstrap sample.

Once a set of screening items was selected, an appropriate cutoff was established to determine the number of positive responses that would identify a CSEC/CST victim or an ASA victim with high certainty. Using various cutoffs (2 positive, 3 positive, 4 positive, and 5 positive screening questions), we calculated sensitivity, specificity, and positive and negative predictive values (NPVs) to determine the number of positive questions that should be used.

Back to Top | Article Outline


There were 108 participants in the study, including 25 patients in the CSEC/CST group and 83 patients in the ASA group. Results are summarized in Tables 1 and 2. There were no statistically significant differences in age or gender between the 2 groups. However, racial/ethnic differences between groups were noted (P = 0.005), with an increased proportion of African American patients and those grouped as “Other” in the CSEC/CST group. The following variables were significantly more common in the CSEC/CST group than in the ASA group: presence of any type of tattoo; lifetime history of fractures, significant wounds, or traumatic loss of consciousness, either accidental or inflicted; history of sexual activity; history of violence at the hands of parents/caregivers and at the hands of others; history of drug use and of multiple drug use; history of running away from home; and history of involvement with law enforcement or with child protective services. As a group, CSEC/CST patients had been sexually active for a longer period, had a greater number of partners, and were more likely to have had an STI and a history of pregnancy. They were also more likely to have an STI detected at the time of the index visit.





Results of the multivariable logistic regression analysis yielded a 6-question screening tool (Table 3). The model AUROC was 0.97, indicating near-perfect discrimination of CSEC/CST victims based on these screening tool items. Bootstrap validation of these risk factors in 1000 samples showed high discriminability for all samples with an average AUROC of 0.97 (range, 0.89-1.0). We assigned a 1-point score for each positive response and summed the values to obtain a total score (ranging from 0 to 6). The median total score (25th-75th percentile) was 4 (3-4) for CSEC/CST patients and 1 (0-1) for ASA patients (P < 0.001). To further classify patients as CSEC/CST, various cutoff scores were evaluated for their sensitivity, specificity, and positive and NPVs (see Table 4). On the basis of this analysis, a cutoff score of 2 was chosen to classify a patient as CSEC/CST. For patients with positive responses to at least 2 of the 6 screening questions, the odds of being a CSEC/CST victim were 22 times higher than a patient who had fewer than 2 positive responses (odds ratio, 21.6; 95% confidence interval, 5.9-79.7; P < 0.001).





Back to Top | Article Outline


In this study of youth aged 12 to 18 years, we identified 16 demographic, behavioral, physical, and historical factors that differentiate CSEC/CST victims from those with allegations of sexual assault/abuse but no evidence of CSEC/CST. Our results provide critical quantitative evidence to support risk factors and experiences described in qualitative reports of sex trafficking,17 studies involving mixed populations of adults and adolescent sex trafficking victims,16,18 and reports published outside peer-reviewed journals,17 thus adding to a small but growing evidence base.

Empiric research on sex trafficking and commercial sexual exploitation remains relatively limited.3 Oram et al6 conducted a systematic review of those peer-reviewed articles published before August 2011 that documented the prevalence or risk of violence and/or measures of physical, mental, or sexual health among victims of human trafficking. Their search identified only 19 articles, 18 of which combined adult and juvenile victims. None of these studies reported on child victims from the United States. When studies combine groups of participants (adult and juvenile, domestic and international, labor and sex trafficking victims), it is difficult to identify the unique characteristics of minors exposed to CSEC/CST, especially those who are US citizens or legal residents (US domestic victims).

Lederer and Wetzel16 described physical and emotional effects of sex trafficking in the United States using a mixed-methods approach involving qualitative and quantitative data. Participants included 106 women and adolescent survivors between 14 and 60 years. The data are extensive, with information on myriad psychological and physical symptoms and experiences of violence and drug use. However, the study does not indicate how many of the participants were minors during their period of trafficking and does not separate adults from minors in the description of results. Cusick19 published a literature review on the minors exploited through prostitution, but this did not include extensive discussions about the health effects on victims. Smith et al17 published a general report on domestic minor sex trafficking in the United States and briefly described data on a variety of physical and emotional consequences reported by victims, but no details of the study producing the results are provided, and these data form a minor portion of the report. Curtis et al15 studied CSEC youth in New York City and gathered limited data on victim health. They found that 20.7% reported a history of an STI and that drug and alcohol use was common: 53.8% reported using marijuana, 26.1% used cocaine, and 25.3% used alcohol. Only 20.5% denied using any drugs or alcohol.

A few of our results were surprising. Although anecdotal experience suggests that many teens have tattoos, our study showed that the presence of a tattoo of any kind—not necessarily one that is sexual in nature or containing a man's name as a form of “branding”—was significantly more frequent among CSEC/CST youth than youth with alleged ASA (48% vs 5%). This finding, and the finding that CSEC/CST youth were more likely to have a lifetime experience of a fracture, significant laceration, or traumatic loss of consciousness from any etiology (32% vs 11%), will need to be verified in a larger multisite study.

Victims of human trafficking typically do not self-identify, and the possible reasons for this are legion.7 Many feel shame, humiliation, fear of the trafficker or of being arrested, guilt, or loyalty to the trafficker related to trauma bonds. Many youth do not view themselves as victims, or they assume their current life is the best available to them. Others may have been told to lie to authority figures or may be closely watched by the trafficker.10,14,18,20,21 However, evidence exists that victims do seek medical attention. In their study of CSEC youth in New York City, Curtis et al15 found that more than 75% reported seeing a medical provider within the last 6 months. Lederer and Wetzel16 found that 88% of women and adolescent victims of trafficking reported seeking medical care at some point during their period of exploitation. In our study, 38% of CSEC/CST victims had seen a health care provider within the past 2 months, although these visits did not necessarily occur during the period of exploitation.

The ability of health care providers to recognize youth at high risk of CSEC/CST is critical to offering victims medical, mental health, and social services. Our study is the first to create a short, data-driven screening tool specifically designed for youth in a health care setting. We found that at least 2 positive answers to a 6-item questionnaire identified CSEC/CST patients with a sensitivity of 92%, specificity of 73%, positive predictive value (PPV) of 51%, and NPV of 97%. Although the PPV is low, the “intervention” prompted by the positive screen is relatively benign and would consist of the provider asking additional questions about high-risk sexual activity and commercial sexual exploitation using a trauma-informed approach that minimizes stress related to traumatic memories. As with any personal questions asked of patients, it is critical for the health care provider to be sensitive to signs of stress, to be nonjudgmental and open-minded during the interview, and to have appropriate resources available for the child who exhibits distress and/or who discloses CSEC/CST.22

Our study has several limitations. Although we included male and female youth, we were only able to identify 3 male adolescents, all in the ASA group. Furthermore, all of our CSEC/CST participants were victims of domestic exploitation; there were no international trafficking victims. This limits the generalizability of our results. Nonetheless, as evidence suggests that female victims are more likely to be involved in sex trafficking than male victims2 and the majority of identified victims of sex trafficking in the United States are US citizens or legal residents,23 these results are not surprising. The true number of male CSEC/CST victims is unknown and even harder to estimate than the number of female victims, as men may be less likely to be viewed as victims and may not be identified at all.24,25

The sample size of CSEC/CST victims was relatively small in our study, and not all patients provided information for all of the questions in the general assessment. The study was completed at a single pediatric health care facility in a single large city in the southern United States. These factors limit the generalizability of the study, and additional research is needed to determine whether similar results are obtained among youth living in other geographic regions, with varying cultural norms, presenting in different medical settings. In addition, the vast majority of our CSEC/CST patients had already been identified by authorities so our results may not generalize to the entire population of victims, including those who have not come to the attention of others. The ultimate goal of a screening tool is to help identify patients who present with varied clinical complaints and no known history of victimization and who do not spontaneously self-identify. Finally, we have no absolute knowledge of victimization, and it is possible that some of the ASA patients were actual CSEC/CST victims. This limitation would lead to an underestimation of the differences between the 2 populations and imply that true results would be even stronger than what we reported. It is also possible that, in some cases, law enforcement or provider identification of a youth as a CSEC/CST victim was incorrect, although, at the very least, these youth would be at very high risk of victimization.

Our study results need to be validated with adolescent populations outside a southern metropolitan area, as regional differences may exist. Future research will include a multisite study of youth presenting to a variety of medical settings, with a range of chief complaints, with or without previous documentation of CSEC/CST activity. Inclusion of youth with and without risk factors for CSEC/CST in the comparison group may well show that characteristics differentiating victims are even stronger than what was demonstrated in this study, which used a comparison group with relatively high rates of risk factors.

Our study demonstrates that female youth aged 12 to 18 years who are alleged victims of CSEC/CST significantly vary from those who are victims of alleged sexual assault/abuse on a number of demographic, behavioral, physical, and historical factors. Although many of these factors have been proposed as risk factors for CSEC/CST in the past, this has been largely based on qualitative and anecdotal data. This study provides quantitative support for the existence of multiple identifiable risk factors for victimization. Furthermore, we created a short 6-item questionnaire that effectively separates victims of alleged CSEC/CST from those with alleged sexual assault/abuse and no evidence of CSEC/CST.

Back to Top | Article Outline


1. United Nations Human Rights. Protocol to Prevent, Suppress and Punish Trafficking in Persons Especially Women and Children, Supplementing the United Nations Convention Against Transnational Organized Crime. 2000. Available at: http://wwwohchrorg/EN/ProfessionalInterest/Pages/ProtocolTraffickingInPersonsaspx. Accessed July 7, 2013.
2. United Nations Office on Drugs and Crime. Global Report on Trafficking in Persons. 2014. Available at: Accessed November 30, 2014.
3. Institute of Medicine and National Research Council. Confronting Commercial Sexual Exploitation and Sex Trafficking of Minors in the United States. Washington, DC: The National Academies Press; 2013.
4. Edwards JM, Iritani BJ, Hallfors DD. Prevalence and correlates of exchanging sex for drugs or money among adolescents in the United States. Sex Transm Infect. 2006;82:354–358.
5. Bigelsen J, Vuotto S. Homelessness, Survival Sex and Human Trafficking: As Experienced by the Youth of Covenant House New York. 2013. Available at: http://wwwcovenanthouseorg/sites/default/files/attachments/Covenant-House-trafficking-studypdf. Accessed November 30, 2014.
6. Oram S, Stockl H, Busza J, et al. Prevalence and risk of violence and the physical, mental, and sexual health problems associated with human trafficking: systematic review. PLoS Med. 2012;9:e1001224.
7. Konstantopoulos WM, Ahn R, Alpert EJ, et al. An international comparative public health analysis of sex trafficking of women and girls in eight cities: achieving a more effective health sector response. J Urban Health. 2013;90:1194–1204.
8. Abas M, Ostrovschi NV, Prince M, et al. Risk factors for mental disorders in women survivors of human trafficking: a historical cohort study. BMC Psychiatry. 2013;13:204.
9. Muftic L, Finn M. Health outcomes among women trafficked for sex in the United States: a closer look. J Interpers Violence. 2013;28:1859–1885.
10. Estes RJ, Weiner NA. The Commercial Sexual Exploitation of Children in the U.S., Canada and Mexico. 2002. Available at: http:// Accessed July 7, 2012.
11. Farley M, Cotton A, Lynne J, et al. Prostitution and trafficking in nine countries: an update on violence and posttraumatic stress disorder. J Trauma Practice. 2004;2:33–74.
12. Raymond J, Hughes D. Sex Trafficking of Women in the United States: International and Domestic Trends. 2001. Available at: http:// Accessed August 4, 2012.
13. Nixon K, Tutty L, Downe P, et al. Everyday occurrence: violence in the lives of girls exploited through prostitution. Violence Against Women. 2002;8:1016–1043.
14. Gragg F, Petta I, Bernstein H, et al. New York Prevalence Study of Commercially Sexually Exploited Children: Final Report. New York State Office of Children and Family Services; 2007.
15. Curtis R, Terry K, Dank M, et al. The Commercial Sexual Exploitation of Children in New York City: Volume 1: The CSEC Population in New York City: Size, Characteristics and Needs. National Institute of Justice, US Department of Justice; 2008.
16. Lederer L, Wetzel C. The health consequences of sex trafficking and their implications for identifying victims in healthcare facilities. Ann Health Law. 2014;23:61–91.
17. Smith L, Vardaman S, Snow M. The national report on domestic minor sex trafficking: America's prostituted children. In: International SH, ed. 2009. Available at: Accessed July 7, 2013.
18. Zimmerman C. Stolen Smiles: A Summary Report on the Physical And Psychological Consequences of Women and Adolescents Trafficked in Europe: London School of Hygiene and Tropical Medicine. 2006.
19. Cusick L. Youth prostitution: a literature review. Child Abuse Review. 2002;11:230–251.
20. Clawson HJ, Dutch N. Identifying Victims of Human Trafficking: Inherent Challenges and Promising Strategies From the Field. 2008. Available at: Accessed January 15, 2015.
21. Crane P, Moreno M. Human Trafficking: What is the Role of the Healthcare Provider? 2011. Available at: Accessed June 20, 2014.
22. Zimmerman C, Watts C. World Health Organization Ethical and Safety Recommendations for Interviewing Trafficked Women. Health Policy Unit, London School of Hygiene and Tropical Medicine; 2003.
23. Bureau of Justice Statistics of Department of Justice. Human Trafficking/Trafficking in Persons. Available at: Accessed August 3, 2012.
24. Dennis J. Women are victims, men make choices: the invisibility of men and boys in the global sex trade. Gend Issues. 2008;25.
25. ECPAT USA. And Boys Too: An ECPAT-USA Discussion Paper About the Lack of Recognition of the Commercial Sexual Exploitation of Boys in the United States. 2013. Available at: http://ecpatusaorg/wp/wp-content/uploads/2013/08/AndBoysToo_FINAL_single-pagespdf. Accessed November 10, 2013.

child sex trafficking; commercial sexual exploitation of children; human trafficking; child prostitution

Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.