Youth‒Police Contact: Burdens and Inequities in an Adverse Childhood Experience, 2014‒2017

Abstract

Objectives. To assess police contact as a potential adverse childhood experience by measuring its prevalence, nature, and distribution among urban adolescents.

Methods. Detailed US population-based data on youth‒police contact were collected in the Fragile Families and Child Wellbeing Study (n = 2478) from 2014 to 2017. Using regression modeling, I assessed adolescents’ police exposure and the magnitude and robustness of racial disparities in police contact. Sensitivity analyses examined disparities by behavior and socioeconomic context.

Results. Urban youths are heavily policed, beginning in preadolescence. Exposure to policing is unevenly distributed, with non-White adolescents—particularly Black boys—reporting more, and more aggressive, contact than their White counterparts. Hispanic‒White differences and disparities in girls’ experiences were less pronounced but present, particularly in how intrusive stops were. Intrusion disparities were robust to most behavioral controls, but not observed among youths with higher socioeconomic status.

Conclusions. Given extant literature documenting adverse health consequences of police encounters, findings implicate policing as a driver of health disparities in adolescence and throughout the life course. Public health infrastructure dedicated to the prevention and treatment of adverse childhood experiences is well suited for mitigating these harms and inequities.

Adverse childhood experiences (ACEs) are potentially traumatic events that occur in childhood or environmental circumstances that can undermine a child’s sense of safety, stability, and bonding.^1,2 ACEs may contribute to toxic stress that can harm development through changes to the nervous, endocrine, and immune systems.² These, in turn, influence adult health and health risk behaviors and may be transmitted intergenerationally.² The Centers for Disease Control and Prevention have made the identification of ACEs, and mitigation of their harms, priorities for public health.²

Recent incidents of police violence against Black people have brought widespread attention to policing as a threat to health and driver of health disparities.^3–9 Ethnographic research finds that in urban communities and communities of color, traumatic experiences with police begin early in life,^10–12 suggesting that they warrant consideration as an ACE. However, little is known about whether these news stories and qualitative accounts fit a broader population pattern¹³ or if they highlight the most aggressive contexts of police surveillance and racial inequity, providing a misleading portrait of youth experiences. I used national data from the Fragile Families and Child Wellbeing Study (FFCWS) to measure police contact experienced by urban adolescents, and racial disparities in these experiences, providing a population perspective on youth‒police contact.

POLICE CONTACT, PUBLIC HEALTH, AND HEALTH DISPARITIES

Police encounters carry a significant threat to physical and psychological health. A long history documents police violence in American life, particularly in Black communities.^14,15 For decades, police brutality and law enforcement homicides were not systematically documented¹⁶; however, the recent proliferation of violent incidents recorded on video and widely publicized has brought police violence to the forefront of national and public health discourse.^16,17 Recent data indicate that police officers have killed approximately 1000 people annually since 2015.¹⁸ Individuals experiencing frisks, searches, and other police intrusion report elevated symptoms of anxiety and posttraumatic stress disorder,⁴ stresses that may manifest physically.^3,5,6,9 Entrenched racial inequality in criminal justice^15,19,20 suggests that police encounters may also trigger stresses associated with exposure to racism.^21,22 Police contact can also harm the health of individuals not personally stopped by police,⁵ particularly if “vicarious contact,” such as witnessing an encounter or knowing someone stopped,^23,24 signals one’s personal vulnerability to police violence. These effects are racially patterned, reflecting the historical and ongoing traumas of systemic racism.^2,8,22

Recent evidence suggests that adolescents—particularly non-White adolescents—face considerable police exposure. A survey of Chicago, Illinois, students found that approximately half had been stopped by a police officer, questioned, and “told off or told to move on” by ninth or tenth grade.¹² A study of Black and Latino boys in “a large city in the Southern United States” found substantial police contact and adverse developmental consequences.²⁵ Ethnographic and journalistic work also documents substantial police contact among girls of color, who experience police contact in qualitatively different ways than their male counterparts—including elements of sexual harassment or assault.^11,26 The early ages at which many Black and Hispanic youths encounter the police,^10–12,25 coupled with the developmental importance of ACEs^1,2 and adolescence,²⁷ suggest that police contact could drive health inequality throughout the life course, necessitating public health intervention. However, to date, we know little about the burdens or distribution of youth‒police contact on a population level.

MEASUREMENT CHALLENGES

The vast majority of police‒public encounters,²⁸ including physically intrusive encounters such as “stop and frisk” activity,²⁹ do not lead to arrests and are less systematically measured.¹³ What we know about involuntary contacts between minors and the police has been inferred from older populations,^4,28 based on group-level analyses³⁰ or individual-level data limited in scope or generalizability.

Single-city surveys and qualitative studies^{10–12,23,25,26} may not generalize nationally. The National Longitudinal Study of Adolescent to Adult Health (AddHealth) asked basic questions on police contact of a national sample³¹; however, the AddHealth cohort came of age in the 1990s, before the rise in “stop and frisk” and other “proactive” policing.²⁹ AddHealth also captured limited information about the quantity, timing, and intrusion of reported encounters, information critical for understanding their health consequences.^4,32 Analyses of administrative data are further limited; administrative data tend to be incident-level rather than person-level, deidentified, and unsuitable for measuring repeated police contact or its potential consequences.

CURRENT CONTRIBUTIONS

As the first population survey of youth‒police contact in the proactive policing era, the FFCWS measures adolescents’ experiences of personal and vicarious police contact, alongside detailed longitudinal information about their behavior, family background, and broader well-being. These data place policing in context and further understanding of its role as a social determinant of health.

This study provides detailed measures of police contact among boys and girls, across race and ethnicity, and estimates the extent to which observed disparities exceed what might be predicted by behaviors that generate police attention. Aggressive policing is increasingly recognized as a contributor to premature morbidity and mortality. Measuring the prevalence and distribution of youth‒police contact in a national sample furthers our understanding of how policing and its individual consequences shape population health inequality.

METHODS

The FFCWS is a population-based cohort survey that follows 4898 children born in 20 large cities between 1998 and 2000, along with their families. Sixteen of these cities were selected using a multistage random sampling process,³³ providing a sample that, when weighted, represents urban births nationwide (national sample baseline n = 3442).

Families were recruited at the hospital following the child’s birth, using a systematic oversample of nonmarital births. The resulting sample was socioeconomically disadvantaged, with high proportions of Black and Hispanic families and high rates of criminal justice involvement. Parents were contacted for follow-up interviews 5 times, most recently between 2014 and 2017. In the 2 most recent waves (year [Y] 9 and Y15, when the children were aged approximately 9 and 15 years), the children were also interviewed. Sampling weights adjusted for family selection into the baseline sample and the child’s retention and participation at Y15.^33,34

Key Measures

Police contact

The FFCWS adolescent survey measured several aspects of adolescents’ experiences with the police, including whether they had personally been stopped by police and whether they had vicarious police contact—witnessing a stop or hearing about a stop of somebody they knew. (Less than 1% of the Y15 national sample “didn’t know” or “refused” to answer the question on police contact or was missing for unknown reasons.) Adolescents personally stopped provided details of their experiences, including the number of stops they experienced, their age when first stopped, and officer behavior in the incident that most stood out in their mind (which I refer to as their “critical stop”). Analyses focused on 2 measures: a binary indicator of personal police contact and an additive scale summarizing critical stop intrusion: whether the officer frisked them, searched their bags or pockets, used harsh language, handcuffed them, used racial slurs, threatened physical force, or used physical force (α = 0.85). Items were coded to zero for adolescents not personally stopped and for those who answered “don’t know” or who “refused” to answer a question. (Only 1% of adolescents stopped by the police responded “don’t know” or “refused” to answer any of the intrusion questions.) This provided a conservative estimate of disparities with alternative model choices explored in sensitivity analyses.

Demographic background

Adolescents’ self-reported racial/ethnic backgrounds were coded into 5 categories (White, Black, Hispanic, other race, and multiple races). Adolescents’ sex was recorded at birth, and their age was measured on their interview date.

Adolescent behavior and social environment

Adolescents’ experiences with police were interpreted in the context of their participation in and exposure to behavior that might attract police attention. Behavior at Y15 was measured using self-reported past-year participation in a series of delinquent activities. The adolescents also reported illegal behaviors their peers engaged in, which comprised a second additive measure, and their own behaviors at Y9, which comprised a third. These indicators, detailed in Appendix A (available as a supplement to the online version of this article at http://www.ajph.org) were used as covariates, for sample stratification, and as predictors of adolescents’ propensity for police contact. In sensitivity analyses, I examined socioeconomic moderation of racial disparities, measuring socioeconomic status (SES) by using mothers’ baseline educational attainment.

Analytical Approach

Analyses examined the extent to which police contact among the FFCWS adolescents might contribute to racial disparities in health. I began with a detailed description of urban adolescents and their exposure to the police, nationally and by race and sex.

Assessing racial disparities

I assessed racial disparities in adolescents’ police contact and critical stop intrusion by using weighted logistic, negative binomial, and ordinary least squares regression models, separately for boys and girls. Models adjusted for adolescent age and self-reported and peer delinquency.

Sensitivity analyses

I examined the sensitivity of findings to several analytic choices and explored variation across the socioeconomic spectrum. As an alternative to regression adjustment for adolescent behavior, I ran models using a sample stratification approach, defining 3 subsamples based on Y15 self-reported delinquency: adolescents reporting none, 1 delinquent behavior, and 2 or more listed behaviors. (Intrusion was modeled by using complete case and ordinary least squares models in the sample stratification analysis, because negative binomial models did not converge in the imputed data set. Details are provided in Appendix B, available as a supplement to the online version of this article at http://www.ajph.org.) To understand the robustness of intrusion disparities to modeling choices, I estimated 2 additional sets of models: limiting the analysis sample to adolescents personally stopped by the police and combining the sampling weights with inverse probability of treatment weighting to control for selection into police contact.³⁵

Analysis sample and sample description

Adolescents were included in the analysis sample (n = 2478) if they were part of the national sample, interviewed at Y15, and reported on personal and vicarious contact with police. Missing predictors were imputed using multiple imputation. (Less than 1% of the analysis sample was missing data on adolescent age, Y15 delinquency, or maternal education. Five percent of the sample was missing peer delinquency data, and 12% was missing Y9 delinquency data.)

Analyses were weighted to represent adolescents born in large cities between 1998 and 2000. A description of the analysis sample is provided in Table 1.

TABLE 1—

Description of Analysis Sample: Based on National Sample From the Fragile Families and Child Wellbeing Study, United States

Variables	% or Mean (SD)
Demographic description
Adolescent race
White	36
Black	24
Hispanic	30
Other	6
Multiple races/ethnicities	5
Adolescent sex
Male	56
Female	44
Adolescent age	15.2 (0.48)
Mom’s baseline education
< High school	25
High school only	30
Some college	20
College graduate	25
Adolescent behavior
Y9 delinquency (range = 0‒17)	1.03 (1.82)
Y15 delinquency (range = 0‒16)	1.09 (1.85)
Y15 peer delinquency (range = 0‒11)	1.20 (2.07)
Police contact
Ever stopped (personally)	19
Vicarious contact	69
Summary of stop experience
First stopped age 10 y or younger	3
First stopped age 11–12 y	3
First stopped age 13–15 y	13
First stopped after age 15 y	< 1
Age first stopped not reported	< 1
Not personally stopped	81
Average number of stops reported	0.44 (1.73)
Ever arrested?	3
Critical stop experience: any critical stop intrusion?	9
Did the officer . . .
Frisk you?	5
Search your pockets or bags?	6
Use harsh language?	3
Use racial slurs?	< 1
Threaten physical force?	2
Use physical force?	1
Handcuff you?	4
Intrusion index (0‒7)	0.22 (0.83)
Any intrusion among adolescents stopped	48
Intrusion index among adolescents stopped	1.17 (1.60)

Note. Y = year. Statistics are weighted to represent urban births between 1998 and 2000 nationwide. The total sample size was n = 2478. Means and standard deviations are based on observed, rather than imputed, data (n = 2187 for Y9 delinquency, n = 2456 for Y15 delinquency, n = 2343 for Y15 peer delinquency, n = 2478 for the number of stops experienced and critical stop intrusion, and n = 677 for critical stop intrusion among adolescents stopped). Percentages may not total 100 because of rounding.

Source. The Fragile Families and Child Wellbeing Study is a longitudinal study in which participants were born over a 3-year time period (1998–2000) and families were reinterviewed 5 times over approximately a 15-year period.

The analysis sample and population it represents are predominantly non-White: more than half the weighted sample was Black or Hispanic; just over one third was White, and approximately 5% were “other” or multiple races. Because of the small sample size, disparities involving “other race” and multiracial adolescents are not reported.

RESULTS

Approximately 19% of adolescents reported having been stopped by the police, and 69% reported vicarious contact. As detailed in Table 1, adolescents reported diverse experiences, beginning at young ages. The population average of 0.44 stops per adolescent included several adolescents stopped 10 or more times. Although frisks, searches, and other police intrusion were rare overall, their prevalence in adolescents’ critical stops (e.g., 23% involved frisks, 30% involved searches, 19% involved handcuffing) suggest that many adolescents’ experiences were far from benign. Notably, 30% of adolescents reporting personal contact were first stopped as preadolescents (ages 8–12 years).

Race differences in adolescents’ experiences with the police are presented in Table 2. Police exposure was common across race, with most Black, White, and Hispanic adolescents reporting vicarious contact. In personal experiences, however, disparities were pronounced. Black boys and girls were each more likely than their White counterparts to report being stopped: 39% of Black boys and 14% of Black girls reported police contact, while only 23% of White boys and 10% of White girls did (P < .001 among boys). Hispanic‒White differences were statistically insignificant.

TABLE 2—

Unadjusted Racial Disparities in Police Exposure and Contact by Sex: Analysis Sample Based on Fragile Families and Child Wellbeing Study National Sample, 2014‒2017

	Boys, No., %, or Mean (SD)				Girls, No., %, or Mean (SD)
	Total	White	Black	Hispanic	Total	White	Black	Hispanic
No.	1283	264	560	348	1195	259	509	322
Police exposure
Vicarious contact	73	76	77	72	66	58	69	70
Ever stopped (personally)	26	22	40*	21	10	10	14	6
Number of stops reported	0.66 (2.21)	0.60 (2.69)	0.84 (1.97)	0.57 (1.58)	0.17 (0.66)	0.13 (0.47)	0.24 (0.78)	0.13 (0.67)
Any critical stop intrusion?	14	5	26***	15**	3	2	6	3
Ever arrested?	4	< 1	9	3	1	< 1	2	1
In your critical stop (if any), did the officer . . .
Frisk you?	7	2	13***	9**	1	0	3**	1
Search your pockets or bags?	9	3	14**	12**	2	2	4	2
Use harsh language?	5	< 1	10*	6*	< 1	0	2***	< 1
Use racial slurs?	2	< 1	3*	2	< 1	0	< 1*	< 1
Threaten physical force?	4	< 1	6**	4	< 1	0	1**	< 1
Use physical force?	2	< 1	2*	4*	< 1	0	1**	< 1
Handcuff you?	6	1	13*	5*	2	< 1	3*	2
Intrusion index (0‒7)	0.34 (1.02)	0.08 (0.40)	0.60*** (1.27)	0.41*** (1.14)	0.07 (0.45)	0.03 (0.19)	0.15** (0.75)	0.06 (0.41)
Among those stopped	473	76	240	116	204	34	107	39
Age first stopped, y	12.84 (1.64)	12.55 (1.76)	12.93* (1.54)	13.23*** (0.15)	12.99 (1.94)	13.02 (2.24)	12.82 (2.00)	13.39 (1.35)
Any critical stop intrusion?	53	21	67***	68***	31	20	42	44
Intrusion index (0‒7)	1.33 (1.66)	0.35 (0.80)	1.53*** (1.65)	1.91*** (1.78)	0.66 (1.27)	0.26 (0.56)	1.05** (1.72)	1.00 (1.31)

Note. Statistics are weighted to represent urban births between 1998 and 2000 nationwide. N = 1277 for number of stops reported by boys, n = 1189 for girls. N = 463 for age boys first stopped, n = 197 for girls, with missing observations distributed across racial groups.

*P < .05; **P < .01; ***P < .001 based on comparisons of Black to White, and Hispanic to White adolescents (within sex) using ordinary least squares and linear probability models.

Table 2 indicates racial disparities in critical stop intrusion, most pronounced among boys: more than two thirds of Black and Hispanic boys stopped reported intrusion in their critical stops, while fewer than one quarter of White boys did. Approximately 12% of Black boys reported frisks, 14% reported searches, 10% reported harsh language, and 12% reported being handcuffed, experiences extremely rare among White boys. Hispanic‒White differences were less pronounced but also significant. Police intrusion was less common among girls; however, intrusion was predominantly reported by Black girls and virtually nonexistent for White girls. Black‒White differences in the additive index of stop intrusion were substantial and statistically significant for boys and girls, as were Hispanic‒White differences among boys.

Table 3 presents racial disparities in police contact for boys and girls, unadjusted and regression-adjusted for adolescent behavior.

TABLE 3—

Relative Odds of Reporting Personal Police Contact: Analysis Sample Based on Fragile Families and Child Wellbeing Study National Sample, 2014‒2017

	Boys, OR (95% CI)		Girls, OR (95% CI)
Variables	Unadjusted Model (n = 1283)	Adjusted Model (n = 1283)	Unadjusted Model (n = 1195)	Adjusted Model (n = 1193)
Adolescent race (Ref = White)
Black	2.29 (1.12, 4.71)	2.04 (0.96, 4.31)	1.55 (0.72, 3.36)	1.23 (0.53, 2.84)
Hispanic	0.96 (0.48, 1.89)	0.81 (0.39, 1.68)	0.64 (0.25, 1.65)	0.48 (0.15, 1.50)
Other	0.37 (0.09, 1.50)	0.46 (0.11, 1.93)	0.01 (0.00, 0.08)	0.01 (0.00, 0.10)
Multiracial	1.65 (0.45, 5.98)	1.31 (0.42, 4.07)	3.06 (0.99, 9.41)	2.49 (0.70, 8.80)
Y9 delinquency		0.96 (0.84, 1.08)		1.20 (0.87, 1.65)
Y15 delinquency		1.49 (1.26, 1.77)		1.56 (1.26, 1.94)
Y15 peer delinquency		1.05 (0.88, 1.26)		0.95 (0.80, 1.13)
Age		1.14 (0.72, 1.82)		0.86 (0.41, 1.81)
Constant	0.29 (0.16, 0.50)	0.02 (0.00, 30.85)	0.11 (0.06, 0.21)	0.65 (0.00, 57 986.76)

Note. CI = confidence interval; OR = odds ratio; Y = year. Analyses are weighted to represent urban births between 1998 and 2000 nationwide.

Black boys had odds of reporting police contact that were more than twice those of White boys. This difference narrowed slightly when adjusting for age and behavior, but this finding was not significant (P = .063). Differences among girls were smaller in magnitude and statistically nonsignificant.

Table 4 presents unadjusted and adjusted racial differences in reported critical stop intrusion. The rate ratios indicate significant, substantial, and robust Black‒White differences among both boys and girls. Differences slightly widened when adjusting for adolescent behavior. Hispanic‒White differences were smaller in magnitude and also widened slightly when adjusting for adolescent behavior, but were only statistically significant for boys. Adjusting for behavior did not significantly change either Black‒White or Hispanic‒White racial gaps; confidence intervals around the rate ratio estimates overlapped substantially between the unadjusted and adjusted models.

TABLE 4—

Incidence Rate Ratios From Negative Binomial Models Predicting Critical Stop Intrusion: Analysis Sample Based on Fragile Families and Child Wellbeing Study National Sample, 2014‒2017

	Boys, IRR (95% CI)		Girls, IRR (95% CI)
Variables	Unadjusted Model (n = 1283)	Adjusted Model (n = 1283)	Unadjusted Model (n = 1195)	Adjusted Model (n = 1193)
Adolescent race (Ref = White)
Black	7.83 (3.71, 16.53)	10.19 (4.92, 21.11)	5.980 (1.45, 24.68)	6.22 (1.72, 22.44)
Hispanic	5.31 (2.36, 11.95)	6.35 (2.90, 13.91)	2.57 (0.52, 12.67)	2.87 (0.63, 12.98)
Other	1.89 (0.46, 7.83)	4.17 (0.87, 19.93)	0.23 (0.02, 2.71)	0.63 (0.07, 6.15)
Multiracial	11.41 (3.51, 37.10)	6.30 (2.08, 19.08)	2.68 (0.37, 19.56)	0.74 (0.13, 4.10)
Y9 delinquency		0.99 (0.89, 1.10)		1.45 (0.86, 2.47)
Y15 delinquency		1.34 (1.14, 1.59)		1.70 (1.28, 2.26)
Y15 peer delinquency		1.15 (0.96, 1.39)		1.03 (0.78, 1.36)
Age		1.23 (0.84, 1.79)		0.80 (0.29, 2.22)
α	7.26 (4.46, 11.82)	4.21 (2.49, 7.14)	30.52 (17.28, 53.89)	10.79 (4.23, 27.51)
Constant	0.08 (0.04, 0.15)	0.00 (0.00, 0.41)	0.03 (0.01, 0.10)	0.14 (0.00, 919 593.42)

Note. CI = confidence interval; IRR = incident rate ratio; Y = year. Analyses are weighted to represent urban births between 1998 and 2000 nationwide.

Sensitivity analyses are provided in Appendix B. When stratified by self-reported delinquency, racial disparities in stop experience were most pronounced among boys reporting more delinquent behaviors. Black boys in this group had more than 3 times greater odds of reporting police contact than their White counterparts. Black‒White differences among boys reporting fewer delinquent activities were smaller in magnitude and statistically nonsignificant. Black‒White disparities in critical stop intrusion, on the other hand, were robust and statistically significant among boys reporting no, 1, or multiple delinquent behaviors. Hispanic‒White differences were significant among boys reporting 1 or more delinquent behaviors. Among girls, Black‒White and Hispanic‒White disparities were concentrated in those reporting multiple delinquent behaviors.

When the critical stop intrusion analysis was limited to adolescents stopped and inverse probability of treatment weights were applied, Black‒White and Hispanic‒White differences remained large and statistically significant. However, sensitivity analyses examining socioeconomic context suggested considerable SES moderation: Black‒White disparities were largely robust across adolescents of less-educated mothers, but not observed among children of college graduates. Hispanic‒White differences in stop intrusion were significant among children of mothers with a high-school education or less, but not children of more educated mothers.

DISCUSSION

Findings suggest that urban adolescents face broad, potentially toxic exposure to police, beginning as early as childhood. Most adolescents in the FFCWS reported vicarious police contact, and nearly one fifth reported personal police contact. Many reported frisks and physical force, and verbal indignities including harsh language and, for some, racial slurs. This exposure to aggressive policing has the potential for lasting harm to the health of a new generation.

Although vicarious police contact was common across race, personal experiences were racially disparate and patterned by class. These findings—particularly the robust disparities in critical stop intrusion—suggest that police encounters with non-White adolescents are qualitatively different, substantially more aggressive than those with White adolescents, and potentially traumatic. Notably, disparities were concentrated among children of less educated mothers, and not observed among the children of college graduate mothers. These findings stand in contrast to previous literature that has found high-SES minority youth to experience particularly disparate policing³⁶ and underscore the salience of policing in the lives of already-vulnerable young people.

Limitations and Directions for Future Research

Although the FFCWS advances our understanding of interactions between urban adolescents and the police, these analyses have limitations. Like all longitudinal surveys, the FFCWS suffers from attrition, raising generalizability concerns. However, attrition was greatest among more disadvantaged families; the findings therefore likely understate the prevalence and severity of youth‒police interactions.

The sample was also too small for detailed examinations of within-group differences such as those between Black and White Latinos, or of adolescents of “other” or multiple races. Descriptive statistics indicated that multiracial adolescents reported more intrusive police encounters than other adolescents. More research is needed to understand these experiences, which likely vary by both adolescents’ physical presentation and social contexts.

Analyses adjusted for adolescents’ self-reported behavior, and conclusions largely depended on the validity of these self-reports. However, the vast majority of results suggest that the intrusive police experiences of Black and Hispanic adolescents extend beyond a reflection of behavioral differences. One likely contributor to these disparities is the disparate social contexts in which adolescents function. Structural racism in the United States³⁷ has contributed to residential and school segregation, with predominantly Black neighborhoods particularly heavily policed.^29,30 The lack of observed racial disparities among children of college graduate mothers suggests that educational attainment may provide Black families a pathway out of segregated, heavily policed neighborhoods. Linking health surveys to external data on neighborhood conditions can advance our understanding of these dynamics.

Although beyond the scope of this analysis, research is also needed to examine disparities in policing across other dimensions of social identity, including skin tone, religion, sexual orientation, gender presentation, disability status, and intersections of each of these with race and sex. Future studies would also benefit from the measurement of sexual misconduct in police encounters, which is likely to occur disproportionately along several of these dimensions as well.

Finally, the extensive exposure to aggressive policing faced by young people, through their own contact, that of their peers, and media exposure to high-profile events, has the potential to have an impact on their immediate and long-term well-being beyond the effects currently documented. Public health researchers should follow today’s youths prospectively with an eye toward these experiences, to understand and treat their effects now and in adulthood.

Public Health Implications

Adverse health outcomes associated with police contact, both early in life⁵ and in adulthood,^4,9 implicate aggressive policing as an ACE that requires institutional attention toward prevention efforts, in the immediate aftermath of contact, and throughout the life course.

Prevention efforts may come from multiple sources. Police departments have several avenues for reform: a reduced reliance on aggressive tactics, equitable treatment of community members, and both individual and institutional accountability for unjust and harmful practices. Public health approaches can be integrated into policing to improve community safety without the harms of more aggressive practices.³⁸ Legislation and public education campaigns can provide material and political support for this integration.² Public health also has the potential to address the vulnerabilities that bring young people to police attention, including substance misuse, mental illness, and behavioral challenges that might be addressed with educational or therapeutic services. Other institutions, such as schools, afterschool programs, and mentorship programs, can also support healthy youth development without the harms of aggressive policing.²

When police contact cannot be preempted among their patients, health professionals should be prepared to treat the resulting harms: to ask about police experiences alongside other ACEs and to help patients process any associated stress and trauma. An initial pediatric screening question, asked of the child or a caregiver, could simply be, “Have you/Has your child ever been stopped by the police?” Follow-up questions related to whether the encounter was physical (e.g., involving a frisk or physical force) or verbally aggressive (e.g., involving racial or homophobic invective, sexual harassment, the threat of force, or other harsh language) could help to guide subsequent care. Particularly for Black adolescents, whose police experiences are significantly more intrusive than those of White adolescents, treatment should deal explicitly with the potential for racialized trauma.^22,39 School personnel should be similarly attentive to the possibility that their students have experienced police contact. Providing an environment in which students feel comfortable disclosing such experiences can help to connect affected youths to support.

Because the health consequences of childhood trauma may persist into adulthood, physicians treating adults should also inquire about experiences with police. If an adult patient reports a history of police contact, their age at first contact, as well as their specific experiences, should determine their treatment needs.

As with other ACEs, screening for police contact and treatment of its effects must be done with care and without stigma. Trauma-informed care must be appropriately nuanced for the unique reactions that individuals have to their experiences. Physicians, educators, and others must also treat the topic with the sensitivity it deserves, to avoid inflicting additional trauma in the retelling of an incident.³²

Police contact has been referred to as a “fact of urban life” that young people must tolerate to stay safe.⁴⁰ This, however, ignores the context in which policing takes place—the potential for police aggression and adverse health consequences and the documented racial inequity in police practices. Designating early police contact as an ACE would provide institutional recognition of the potential for harm in police encounters and draw on an established literature and policy and practice framework for the prevention, identification, and treatment of these harms. In heavily policed communities, such care is critical to public health.