Key PointsQuestion Is exposure to police associated with adverse health outcomes for Black youth in the US?
Findings In this systematic review of 29 studies that included 19 954 participants, police exposure was associated with multiple health outcomes for Black youth, including adverse mental health, risk behaviors, and impaired safety.
Meaning Police exposure should be considered a critical determinant of health.
Importance Black youth in the US experience disproportionate contact with police even when accounting for criminal or delinquent behavior, which some experts say is fueled by racism and discrimination. While the literature supports the link between racism and adverse health outcomes, less is known about the impact of policing on the well-being of Black youth.
Objective To systematically review the literature describing the association between police exposure and health outcomes for Black youth 26 years and younger.
Evidence Review A search of PubMed, Embase, Criminal Justice Abstracts, PsycInfo, and Web of Science was conducted. Eligible studies included original peer-reviewed research published from 1980 to December 2020, with a participant population of Black youth, a focus on police exposure, and health as the outcome. Additional articles were identified by hand-searching reference lists of included studies. Data extraction was performed, followed by critical appraisal of all included studies using a convergent segregated approach in which quantitative and qualitative studies were synthesized separately followed by an overarching synthesis across methods.
Findings A total of 16 quantitative studies including 19 493 participants were included in the review and demonstrated an association between police exposure and adverse mental health, sexual risk behaviors, and substance use. A total of 13 qualitative studies including 461 participants were included in the review, which corroborated and contextualized the quantitative evidence and provided additional health outcomes, such as fear for life or hopelessness.
Conclusions and Relevance Evidence shows that police exposures are associated with adverse health outcomes for Black youth. Clinicians, scientists, public health practitioners, and policy makers can partner with local governments to enact reforms that mitigate the health impact of policing on youth.
Black youth in the US experience contact with the police as early as 8 years old,1 and by age 24 years, they have nearly 9-fold the number of encounters with police as their White counterparts.2 Furthermore, Black youth are more likely to experience use of force3 and 5-fold as likely to experience injury during these interactions.4 Such racial disparities in police contact are poorly explained by individual-level differences in delinquency. In fact, among Black individuals, arrest odds strengthen after accounting for differential crime involvement.5-9 Research also shows that police tend to evaluate Black boys as being 4 years older than their actual age and as less innocent compared with White children.10 Such disproportionate contact and differential treatment has been identified as a form of structural racism.11,12 While the literature supports the link between racism and adverse health outcomes,13 less is known about the impact of policing on the well-being of Black children.
The American Academy of Pediatrics recently examined the impact of racism on child health and called for evidence-based strategies to dismantle structures that perpetuate racism.14 Additionally, the American Public Health Association highlighted police violence as a public health issue given its disproportionate impact on marginalized populations.15 Despite a national push to address racism and policing, researchers have yet to critically appraise the state of the science regarding policing, race, and child health. This mixed-methods systematic review fills a critical gap by examining the existing literature documenting the association between police exposures and health for Black youth, an important step in informing future changes to practice and policy. We hypothesize that police exposures will be associated with adverse health for Black youth.
This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline.16 We examined studies that explored the association between police exposures and health outcomes for Black youth 26 years and younger.
Literature Search Strategy
We searched the following databases on December 19, 2020: PubMed, Embase, Criminal Justice Abstracts, PsycInfo, and Web of Science (eFigure 1 in the Supplement). Search terms were developed by one of us (M.J.) with the assistance of a medical librarian and encompassed the following concepts: (1) police, (2) Black, and (3) health outcomes (eFigure 1 in the Supplement).
Study Selection and Data Extraction
Following removal of duplicates, 2 independent reviewers (M.J. and A.M.) performed title, abstract, and full-text screening as well as data extraction using Covidence software (Veritas Health Innovation). Included studies were peer-reviewed original research (quantitative or qualitative) conducted in the US and published between 1980, when community policing became prevalent, and December 2020. The study population was Black youth. Studies were included if participants 26 years and younger and 50% or more of the study population included Black participants or if there was a subgroup analysis of Black participants (ie, race stratification, with race as a modifier variable). For studies with less than 50% Black participants and race as a covariate, authors were contacted to request subgroup analyses. If subgroup analyses were not available, the study was excluded. Additionally, Hispanic/Latinx ethnic subgroups were excluded if Black race was not indicated.
The main exposure of interest was exposure to police, which was operationalized to include police contact (presence of police in schools, personal experiences ranging from benign stops to use of force and arrest, and vicarious exposure) and perceptions of police discrimination. Presence of police in schools was included given research indicating that more than 1 million children in the US attend schools with police.17 Vicarious exposures and perceptions of police were included given evidence supporting their impact on child health.18
Health outcomes included physical health, mental health, risk behaviors (eg, substance use), and safety. Three studies assessed physical health outcomes and were excluded given no comparable studies.19-21 Mediation analyses were included if direct associations between police exposure and health outcomes were reported. We identified additional articles by hand-searching reference lists of included studies. Disagreements were resolved through discussion with coauthors (K.B.M., M.T., and R.L.J.T.).
Quantitative and Qualitative Quality Assessment
To assess study quality for quantitative studies, we used the Quality Assessment Tool from the Effective Public Health Practice Project, which assesses studies based on selection bias, study design, confounders, data collection, and analysis.22 For qualitative studies, we used the National Institute for Health and Care Excellence methodology checklist,23 which assesses studies based on theoretical approach, study design, data collection, validity, analysis, and ethics. Each study was assigned an overall rating of strong if they had no weak ratings, moderate if they had 1 weak rating, and weak if they had 2 or more weak ratings.
We used a convergent segregated approach, which involves synthesizing all studies separately according to their design (quantitative vs qualitative), followed by an overarching synthesis across methodologies (Figure 1).24 We grouped health outcomes from quantitative studies into categories and characterized the direction of their associations (positive, negative, or null) with police exposures. A P value less than .05 was considered statistically significant, and null associations indicated no statistical significance. A meta-analysis was not performed owing to study heterogeneity. For the qualitative studies, we performed thematic synthesis.25 Using MAXQDA software version 18.2.5 (VERBI Software), we conducted open-ended coding on each text-unit (ie, sentences or paragraphs) within the Results and Discussion sections. Similar concepts were grouped into descriptive themes, which were grouped into analytic themes and organized by health domain. If the same data were used in more than 1 study,26-33 the additional studies derived from the same data were only included if they elicited new themes. A combined synthesis of the results was performed by juxtaposing quantitative and qualitative findings within a matrix to compare findings.
We screened 2997 titles and abstracts and 259 full-text manuscripts. A total of 23 manuscripts met inclusion criteria, and 6 studies were added from reference lists. Characteristics of included studies are summarized in Table 1. Detailed summaries of all included studies can be found in the eTable in the Supplement, and the PRISMA flow diagram can be found in eFigure 2 in the Supplement.
The 29 studies in this review (16 quantitative and 13 qualitative) include data from 19 954 participants aged 9 to 26 years old. A total of 27 studies (93%) included adolescents or young adults (aged 12 to 26 years). A total of 16 studies (55%) included male and female participants. A total of 11 studies (38%) assessed male participants only, 3 of which (27%) focused on men who have sex with men. A total of 2 studies (7%) focused on transgender adolescent girls and young women only. A total of 22 studies (76%) involved a sample composed of 50% or more Black persons. The remaining 7 studies (24%) included subgroup analyses by race.
Quantitative and qualitative studies operationalized police exposure broadly. A total of 13 of 16 quantitative studies (81%) assessed police contacts, while 3 of 16 (19%) assessed perceptions of police discrimination. All used self-reported measures. Qualitative studies addressed a variety of police exposures, including police contact and perceptions of police discrimination, many of which addressed multiple types of police exposures within one study.
Quantitative and qualitative studies operationalized health outcomes broadly to include mental health, risk behaviors, and safety.
Figure 2 summarizes the quality ratings by category as well as global ratings for all studies.34,35 Most included studies (21 of 29 [72%]) were of moderate or strong quality. Among the quantitative studies, 3 (20%) were rated strong, 7 (40%) were rated moderate, and 6 (40%) were rated weak. Among the qualitative studies, 5 (39%) were rated strong, 6 (46%) were rated moderate, and 2 (15%) were rated weak.
Table 2 summarizes the associations between police exposures and health outcomes for the included quantitative studies.36-51 Most quantitative studies (13 [81%]) indicated a positive association between police exposures and adverse health for Black youth.
Nine studies (56%) examined 20 distinct associations between police exposures and adverse mental health.36,37,41,42,44,46,48,50 Police exposure was associated with poor public regard (β = −0.305; SE = 0.135; P = .02),41 sadness (odds ratio [OR], 1.47; 95% CI, 1.18-1.82; P < .001),44 anger (β = 0.12; 95% CI, 0.03-0.21; P < .0148; OR, 1.40; 95% CI, 1.14-1.72; P < .00144), externalizing behaviors (β = 0.081; SE = 0.048; P < .05),50 fear (OR, 1.56; 95% CI, 1.20-2.02; P < .001),44 psychological distress (β = 0.05; SE = 0.02; P < .01),37 and feelings of safety (OR, 1.21; 95% CI, 1.01-1.44; P = .04).46 Police exposure was associated with stress in 4 of 5 associations (β = 0.723; SE = 0.340; P < .05; β = 0.862; SE = 0.421; P < .05; β = 1.003; SE = 0.427; P < .05; β = 1.614; SE = 0.417; P < .001),42 with one revealing no association.41 There were mixed findings for the association between police exposure and depression, with 4 positive associations (β = 0.161; SE = 0.044; P < .01; β = 0.146; SE = 0.041; P < .01; β = 0.034; SE = 0.009; P < .0149; β = 1.12; SE = 0.43; P < .0136) and 3 showing no association.36,41,48 Most studies (6 [67%]) were of moderate or strong quality.
Four studies (27%) examined 6 distinct associations between police exposures and sexual risk behaviors.38,39,45,47 Positive associations were seen with intentions to have sex (OR, 1.52; P < .05),38 number of anal sex partners (event rate ratio, 1.89; 95% CI, 1.41-2.55; P < .001),47 and number of overall sex partners (event rate ratio, 1.43; 95% CI, 1.13-1.81; P < .01).47 No association was seen with unprotected anal intercourse.39,45,47 There were no negative associations. Overall, the studies were of moderate or strong quality, with the exception of a single study examining police exposure and sexual intentions, which was of weak quality.38
Three studies (20%) examined 8 associations between police exposures and substance use.40,43,51 Positive associations were seen with smoking for boys (OR, 2.0; 95% CI, 1.2-3.4) but not for girls.51 There were mixed findings for the association between police exposure and drug use, with 5 positive associations (men: β = 0.268; P < .001; women: β = 0.357; P < .00140; β = 0.40; SE = 0.16; P < .05; β = 0.64; SE = 0.17; P < .001; β = 0.47; SE = 0.18; P < .01),43 and 1 showing no association (β = 0.15; SE = 0.17).43 Most studies of risk behaviors were of moderate or strong quality (4 [57%]).51
Table 3 summarizes the themes identified for the included qualitative studies.26,29,31,32,52-55 Qualitative synthesis resulted in the identification of descriptive themes,25 which were grouped into analytic themes and were ultimately organized within 3 health domains, including mental health, risk behaviors, and safety.
The mental health domain includes the following analytic themes: psychological distress, maladaptive coping, vicarious impact, and role confusion.
Eleven studies (85%) explored police exposures and psychological distress, which comprises the descriptive themes of internalizing symptoms26,29,31,32,52-58 (ie, feelings of depression, anxiety, withdrawal, hypervigilance) and externalizing symptoms29,31,32,52,57-59 (ie, feelings of anger, aggression, resentment).
Twelve studies (92%) explored police exposures and maladaptive coping, which includes the descriptive themes of identity negotiation32,53,54,56,60 (ie, fear influences self-acceptance), self-policing one’s behavior29,31,32,55,56,60 (ie, adjusting behavior to avoid feeding into stereotypes), normalizing29,31,55,57,59 (ie, perceiving negative police interactions as normal, minimizing resultant symptoms), and resignation to injustice26,29,32,52,56-58 (ie, negative police interactions are inevitable).
Seven studies (54%) explored police exposures and vicarious impact, which includes the descriptive themes of personal impact of publicized killings26,31,32,52,56-58,60 (ie, impact even if not the victim or connected to victim), messaging to loved ones26,52,56 (ie, negative police encounters influence messaging to family), imposing fear on family members52 (ie, guilt related to family’s fear for their lives), and fear for family members31,56,57 (ie, fear for family and resultant negative toll).
Six studies (46%) explored police exposures and role confusion, which includes the descriptive themes of cannot be a kid29,32,52,56,58 (ie, policed while waiting for the bus, studying, playing sports) and cannot be a concerned family member31 (ie, danger in defending loved ones to police).
The risk behaviors domain includes the following analytic themes: risk taking, impairment of future orientation, and inability to engage in prosocial activities.
Two studies (15%) explored police exposures and risk taking,54,55 which includes the descriptive themes of sexual risk behaviors54 (ie, sexual behaviors predisposing youth to adverse health outcomes) and substance use (ie, use of substances, including alcohol and drugs).55
Impairment of Future Orientation
Twelve studies (92%) explored police exposures and future orientation, which includes the descriptive themes of fear for life26,32,52-61 (ie, fear of being killed by police regardless of actions) and hopelessness29,32,55,60 (ie, hopeless regarding police interactions).
Inability to Engage in Prosocial Activities
Eight studies (62%) explored police exposures and inability to engage in prosocial activities, which includes the descriptive themes of social spaces for destressing off limits29,31,32,52,58,61 (ie, inability to be with friends in public spaces, such as parks), group gatherings deemed suspicious31,32,61 (ie, policed when in groups with an assumption of gang or criminal activity), overpolicing discouraging family visits26 (ie, family avoid visiting or will not pick up youth because of policing in neighborhood), and social isolation or exclusion32,53,54 (ie, hypersurveillance plus discrimination leads to feelings of loneliness and feeling unwanted).
The safety domain includes the following analytic themes: maltreatment and vulnerability to violence.
Ten studies (77%) explored police exposures and experiences of abuse, which includes the descriptive themes of verbal abuse29,31,53,56,58,59 (ie, use of racial epithets, name calling, and aggressive language), physical abuse29,31,53,55,56,58,59 (ie, kicked, choked, punched, gun placed in mouth), sexual abuse31,53 (ie, sexual assault), and dehumanization29,31,32,52,53,55,58,60 (ie, humiliation; labeled as stupid, worthless).
Vulnerability to Violence (Police and Community)
Five studies (39%) explored police exposures and vulnerability to police and community violence, which includes the descriptive themes of interactions impact compliance26 (ie, previous interactions instill fear and lead to flight response), treatment impacts compliance26 (ie, being antagonized leads to defiance), concerns ignored29,31,53,58,59 (ie, not being believed by police or concerns not taken seriously), victim blaming53,59 (ie, blamed when reporting concerns), avoiding help from police26 (ie, avoidance of seeking help because of previous experience or desire to not be associated with police within neighborhood), and purposeful abandonment by police29,31 (ie, instances in which police knowingly abandon youth in unsafe neighborhoods far from home).
Overall, the included qualitative studies were of moderate to strong quality, with the exception of 2 studies (15%) of weak quality.52,57
Synthesis of Quantitative and Qualitative Studies
Juxtaposition of the health outcomes measured in quantitative studies with the themes explored in the qualitative studies resulted in a combined analysis.
Supportive and Contradictory Findings
There were corroborating and contradictory findings across the health domains. Within mental health, qualitative studies consistently identified themes about police exposure and internalizing symptoms, such as stress and depression. However, quantitative findings were mixed with a largely positive association with stress but mixed findings regarding depression. Contradictory findings were also seen for feelings of safety vs fear. One quantitative study showed police exposure was associated with feelings of safety, which was not supported by the 1 quantitative study44 assessing fear nor by the multiple qualitative studies26,32,52-61 showing police exposure’s association with fear for life. Quantitative and qualitative studies, however, both showed associations and identified consistent themes between police exposures and anger/aggression (referred to as externalizing symptoms) and vicarious impact.
Within the domain of risk behaviors, while both quantitative and qualitative evidence showed associations and identified consistent themes with sexual risk behaviors, findings were contradictory for unprotected anal sex. Quantitative studies showed positive associations but were not statistically significant, while qualitative evidence indicated transgender adolescent girls and young women reported avoiding carrying condoms because of fear of police. Lastly, both syntheses showed associations and identified themes with substance use. Quantitative literature studied drug use and smoking, while qualitative literature additionally explored alcohol use.
Qualitative Evidence That Explains Quantitative Evidence
Qualitative evidence helps explain the quantitative associations between police exposures and mental health. Participants reported anger as a result of having to respond to racism and interact with a system in place to protect that instead caused them harm. Repeated negative exposure and the inability to resolve confrontation with police were also cited as leading to symptoms of depression. Marginalization and being assumed to be guilty contributed to frustration, anxiety, and hopelessness.
Qualitative evidence also helps explain the quantitative associations with risk behaviors. Youth reported avoiding carrying condoms owing to fear of police confiscation and subsequent assertion that condoms are evidence of sex work. Participants also noted that lack of support from social institutions, including police, diminished self-worth, and made them more vulnerable to risky sexual behaviors and substance use.
Qualitative Evidence That Explains Differences Across Studies
While the qualitative studies identified themes connecting police exposures and mental health, quantitative studies that examined depression had mixed findings. The conflicting direction of these results may be explained by the qualitative evidence, showing that Black boys and young men may not label their depressed mood as a mental health issue secondary to normalizing or avoiding stigma. Furthermore, qualitative evidence indicated themes connected with intermediary symptoms, such as worry, stress, and nervousness, which may be precursors to the outcomes measured in many of the quantitative studies, such as depression and anxiety.
Outcomes and Phenomena Explored in Only One Methodology
There were analytic themes from the qualitative literature that were not examined in any of the quantitative studies. Within the mental health domain, this included maladaptive coping and role confusion. Within the risk behaviors domain, this included inability to engage in prosocial activities. Lastly, within the safety domain, this included maltreatment and vulnerability to violence.
Our review found that the current body of evidence, although limited, reveals a positive association between police exposure and adverse health outcomes for Black youth. To our knowledge, this is the first study that comprehensively synthesizes this research and furthermore uses qualitative evidence to contextualize quantitative findings. Of the 29 studies, 25 reported at least 1 association or thematic link between police exposure and adverse health outcomes. Of the 8 studies of strong quality (3 quantitative and 5 qualitative), all found associations or themes regarding policing and adverse health outcomes. The 3 studies, all quantitative, that did not reveal this association were of weak or moderate quality. While there has been a growing focus on the most serious consequences of police exposure, including death, this systematic review brings to light that seemingly trivial police contact or even the fear of contact may be detrimental to the health of Black youth.
Although the quantitative associations and qualitative themes linking police exposure to adverse health outcomes were highlighted across both methodologies, qualitative studies illuminated the specific role of racism within this association. Similarly, the wide-range of settings—grade schools, historically White colleges, neighborhoods of varying racial compositions, streets, and parks—emphasizes the ubiquity of such encounters as well as the role of racial bias in policing Black youth. Prior research linking racism to adverse mental health,62 substance use,63,64 and sexual risk behaviors65 supports the notion that racism is a key contributor to police exposure’s association with adverse health.62
Implications for Research
This review bares important methodological considerations for future research. First, there is no consensus on police exposure measurement. Future research may use standardized tools, such as the Police and Law Enforcement Scale,66 or draw from adult literature that uses various data sets67,68 to perform ecological analyses. For example, data from Mapping Police Violence69 was used to show that participants living in the same states as recent police killings had an increase in adverse mental health.68 With increasing national attention on police violence,70 ecological analyses will be an important consideration for forthcoming studies.
Second, while frequency of exposure and use of force have been shown to have dose-related effects on health, few studies interrogated this.36,42 To fully capture policing’s impact on health, we recommend including these details when measuring exposure.
Third, using nonracially diverse samples or race as a covariate creates difficulty in drawing comparisons between groups. The racial gradient theory purports that Black individuals are treated worse than Latinx individuals who are treated worse than White individuals.71 Future research may use diverse samples and race stratification or effect modification to probe racial disparities more accurately.
Additionally, this review shows that police exposures are associated with intermediary health outcomes, such as fear and hopelessness, which impede a child’s ability to thrive72 and may later manifest as depression or anxiety. By studying health outcomes along the continuum, we will gain a more comprehensive understanding of police exposure’s true impact. Also, although adult literature has explored physical health, including physiological impacts of policing on cortisol levels and telomere length,73 physical health was not studied in any of the included quantitative studies, a gap in pediatric literature.
Lastly, there is a dearth of literature examining police exposure in additional sectors, such as within schools. While one study included in this review reveals police presence was associated with feeling safe, this same study shows that Black students were more likely than students of other races to be disciplined despite their positive perceptions of police.46 Similarly, policing in schools is associated with poor test scores and dropout.74 As the presence of police officers increases in schools,75 educational settings are an important setting to conduct future research.
Implications for Practice and Policy
Our research suggests that it is vitally important to consider the influence that policing has on Black youth across multiple sectors. Pediatric clinicians must use a standardized approach to screen youth for not only exposure to police but also the quality of such exposure (eg, use of force, racially tinged language, other forms of harassment). They then can use tools to screen for associated health manifestations and referral to ancillary services (eg, mental health resources) to ensure high quality of care when children experience negative police encounters. Pediatric clinicians can also partner with public health practitioners and policy makers to support community and legislative reforms that focus on mitigating the health impact of police exposures on youth, which may include youth-specific limits to touch and force, youth development training for police, or alternative approaches to safety that limit the use of police in spaces that youth frequent.
The results from this review must be considered in light of its limitations. No studies included youth younger than 9 years, suggesting the difficulty in extrapolating this evidence to children of all ages. However, research showing police exposure’s association with low birth weight corroborates that children are impacted by policing even while in the womb.19 Future work may consider innovative approaches, such as use of drawings to interrogate police’s impact on younger children.76
Additionally, most studies used cross-sectional analyses that cannot assess temporality and may have confounders. However, the included longitudinal analyses did reveal robust associations with police exposure and adverse health outcomes.
Lastly, only 6 of 15 quantitative studies adjusted for delinquent behavior. While previous literature debunks the perception that Black youth’s contact with police is linked to differential wrongdoing, some continue to overestimate criminal involvement for Black youths. Further prospective analyses that adjust for delinquent activity will serve to bolster current evidence that decouples racial disparities in police contact and criminal activity. It is also important to note that youth’s experiences of discrimination and abuse within police encounters renders the presence of criminal activity potentially immaterial to the association between police exposure and adverse health.
This systematic review is an important step in understanding the association between police exposures and adverse health outcomes for Black youth. Despite limitations, it fills a critical gap in the literature and highlights the need for additional, rigorous prospective research assessing the association between police exposure and health among Black youth, who are disproportionately exposed to police interactions. Building the cumulative evidence base is vital to inform the efforts of pediatric clinicians as we partner with other scientists, public health practitioners, and policy makers to continue to shine a bright light on the long-standing health inequities that result from unequal and unfair treatment.