Injuries to Children and Adolescents by Law Enforcement

Abstract

This observational study evaluates the demographic and temporal distributions of hospital-treated injuries perpetrated by law enforcement among youth at the intersection of age, sex, and race and ethnicity.

Police violence is a critical issue associated with adverse health throughout the lifecourse.¹ Exposure to police violence in youth may be uniquely harmful, as adverse experiences during sensitive developmental stages have health implications that compound into adulthood.^2,3 Long-standing work by activists and organizers to address police violence has led to calls for documentation of health outcomes among youth that are associated with policing practices.^1,2,3

To date, the few health studies on police violence among youth have elucidated mechanisms of exposure to policing practices, including direct contact (eg, stop and frisk), anticipatory contact (eg, surveillance), vicarious contact (eg, witnessing police violence), and in utero exposure.^2,3 They have also found associations between policing practices and adverse health outcomes, including anxiety and posttraumatic stress disorders, injury, and death, demonstrating the health harms associated with policing even when it is operating as designed.^3,4 Further, police violence against youth is patterned by experiences of structural marginalization, with racially minoritized youth disproportionately targeted and harmed by policing practices.^2,3,4 However, this literature groups together youth of all ages or reports findings for boys alone, potentially obscuring important patterns. We examined demographic and temporal distributions of hospital-treated injuries perpetrated by law enforcement among youth at the intersection of age, sex, and race and ethnicity.

Methods

We used statewide emergency department and hospitalization data from California between January 2005 and December 2017. The California Health and Human Services Agency and University of California, Berkeley Committees for the Protection of Human Subjects approved this study. Informed consent was not required because in California, confidential deidentified patient-level hospital data sets are available for research purposes through the Information Practices Act. Injuries caused by law enforcement among patients aged 0 to 19 years were identified using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and ICD-10-CM codes for legal intervention injury (eMethods in the Supplement). We calculated legal intervention injury rates per 100 000 person-years overall and by age, sex, and race and ethnicity (identified in the patient record), and at their intersections. We estimated rate ratios and rate differences with 95% CIs to examine inequities by demographic characteristics. Data set cleaning and creation were performed using SAS version 9.4 (SAS Institute) and analyses were conducted in R version 3.6.1 (the R Foundation). We conducted a sensitivity analysis to assess whether results were robust to changes in injury codes across the ICD transition (eMethods in the Supplement).

Results

There were 15 967 youth treated for legal intervention injury in California hospitals from January 2005 to December 2017 (Table). The overall rate of injury was 11.9 per 100 000 person-years. Black youth experienced higher injury rates than youth of other races and ethnicities. Black boys aged 15 to 19 years had the highest rate (200.9 per 100 000 person-years), experiencing 143.2 additional injuries per 100 000 person-years (95% CI, 134.8-151.6) compared with White boys of the same age (rate ratio, 3.5; 95% CI, 3.3-3.7). Compared with White girls aged 15 to 19 years, Black girls of the same age experienced 4.3 times the injury rate (95% CI, 3.7-4.9). Relative inequities between Black and White youth were even greater among those aged 10 to 14 years. Black boys had 5.3 times (95% CI, 4.3-6.5) the injury rate of White boys, and Black girls experienced 6.7 times (95% CI, 4.8-9.5) the injury rate of White girls. The rate among Black girls was higher than all other groups except Black boys.

Table. Pooled Legal Intervention Injury Rate, Overall and by Demographic Characteristics, and Relative and Absolute Inequities, Among Youth Aged 0 to 19 Years in California, 2005 to 2017.

Characteristic	LI injury, No. (%)	Rate per 100 000 PYs	Rate ratio (95% CI)	RD per 100 000 PYs (95% CI)
Total	15 967 (100.0)	11.9	NA	NA
Race and ethnicitya
American Indian or Alaska Native	68 (0.4)	12.4	1.2 (0.9 to 1.5)	2.0 (−0.9 to 5.0)
Asian or Pacific Islander	281 (1.8)	1.9	0.2 (0.2 to 0.2)	−8.4 (−8.8 to −8.1)
Black	3035 (19.0)	39.3	3.8 (3.6 to 4.0)	29.0 (27.6 to 30.5)
Latinx or Hispanic	7382 (46.2)	10.9	1.1 (1.0 to 1.1)	0.6 (0.2 to 1.0)
White	3902 (24.4)	10.3	1 [Reference]	1 [Reference]
Other or multiple raceb	849 (5.3)	15.2	1.5 (1.4 to 1.6)	4.8 (3.8 to 5.9)
Sex
Boys	14 154 (88.6)	20.6	7.5 (7.2 to 7.9)	17.9 (17.5 to 18.2)
Girls	1791 (11.2)	2.7	1 [Reference]	1 [Reference]
Age, y
0-4	49 (0.3)	0.1	1 [Reference]	1 [Reference]
5-9	45 (0.3)	0.1	0.9 (0.6 to 1.4)	0.0 (−0.1 to 0.1)
10-14	1227 (7.7)	3.6	24.3 (18.3 to 33.1)	3.5 (3.3 to 3.7)
15-19	14 646 (91.7)	41.6	278.7 (210.7 to 376.9)	41.5 (40.8 to 42.2)
10-14 y c
Race and ethnicitya and sex
Boys
American Indian or Alaska Natived	NA	NA	NA	NA
Asian or Pacific Islander	17 (1.4)	0.9	0.2 (0.1 to 0.4)	−3.0 (−3.7 to −2.4)
Black	211 (17.2)	20.7	5.3 (4.3 to 6.5)	16.8 (13.9 to 19.6)
Latinx or Hispanic	440 (35.9)	5.1	1.3 (1.1 to 1.6)	1.2 (0.5 to 1.9)
White	194 (15.8)	3.9	1 [Reference]	1 [Reference]
Other or multiple raceb	50 (4.1)	7.4	1.9 (1.4 to 2.6)	3.5 (1.4 to 5.7)
Girls
American Indian or Alaska Natived	NA	NA	NA	NA
Asian or Pacific Islanderd	NA	NA	NA	NA
Black	84 (6.8)	8.6	6.7 (4.8 to 9.5)e	7.3 (5.5 to 9.2)e
Latinx or Hispanic	106 (8.6)	1.3	1.0 (0.7 to 1.4)	0.0 (−0.4 to 0.4)
White	60 (4.9)	1.3	1 [Reference]	1 [Reference]
Other or multiple raceb,d	NA	NA	NA	NA
15-19 y
Race and ethnicitya and sex
Boys
American Indian or Alaska Native	51 (0.3)	62.7	1.1 (0.8 to 1.4)	5.0 (−12.3 to 22.3)
Asian or Pacific Islander	236 (1.6)	11.3	0.2 (0.2 to 0.2)	−46.4 (−48.9 to −43.9)
Black	2314 (15.8)	200.9	3.5 (3.3 to 3.7)	143.2 (134.8 to 151.6)
Latinx or Hispanic	6275 (42.8)	72.9	1.3 (1.2 to 1.3)	15.2 (12.5 to 17.9)
White	3189 (21.8)	57.7	1 [Reference]	1 [Reference]
Other or multiple raceb	714 (4.9)	112.6	2.0 (1.8 to 2.1)	54.9 (46.4 to 63.4)
Girls
American Indian or Alaska Natived	NA	NA	NA	NA
Asian or Pacific Islander	19 (0.1)	1.0	0.1 (0.1 to 0.2)	−7.6 (−8.5 to −6.7)
Black	397 (2.7)	36.5	4.3 (3.7 to 4.9)e	28.0 (24.3 to 31.7)e
Latinx or Hispanic	506 (3.5)	6.2	0.7 (0.6 to 0.8)	−2.3 (−3.3 to −1.4)
White	439 (3.0)	8.5	1 [Reference]	1 [Reference]
Other or multiple raceb	70 (0.5)	11.3	1.3 (1.0 to 1.7)	2.7 (0.0 to 5.5)

Abbreviations: LI, legal intervention; NA, not applicable; PYs, person-years; RD, rate difference.

^a Using data on race and ethnicity from patient records, the racial categories were operationalized such that American Indian or Alaska Native, Asian or Pacific Islander, Black, White, and other or multiple race excluded patients identifying as Latinx or Hispanic. A total of 450 (2.8%) and 22 (0.1%) legal intervention injury patients missing race and ethnicity and sex, respectively, were excluded from relevant subgroup analyses.

^b The other or multiple race category includes patients who reported a race other than American Indian or Alaska Native, Asian or Pacific Islander, Black, or White, as well as patients who reported more than 1 of these categories.

^c Counts among children younger than 10 years were too low to report by sex and race and ethnicity.

^d Not reportable because of cell count <15.

^e Black girls aged 10 to 14 years vs White boys as reference: rate ratio (95% CI), 2.2 (1.7 to 2.9); rate difference (95% CI), 4.7 (2.8 to 6.6) per 100 000 person-years; Black girls aged 15 to 19 years vs White boys as reference: rate ratio (95% CI), 0.6 (0.6 to 0.7); rate difference (95% CI), −21.2 (−25.3 to −17.1) per 100 000 person-years.

Legal intervention injury rates increased and then declined between 2005 and 2017, with Black boys experiencing a much sharper increase and later decline (Figure). The trajectory for Black girls was closer to that for White boys and Latinx or Hispanic boys than that of girls of any other race or ethnicity, and their rate of injury was higher than that of White and Latinx or Hispanic boys by 2017.

Figure. Trends in the Annual Rate of Legal Intervention Injury Among Youth Aged 0 to 19 Years by Race and Ethnicity and Sex in California, 2005 to 2017.

Using data on race and ethnicity from patient records, the racial categories were operationalized such that Black, White, and other or multiple race excluded patients identifying as Latinx or Hispanic. Annual counts were too low (less than 15) to report rates for American Indian or Alaska Native and Asian or Pacific Islander boys and girls, and for other or multiple race girls. The other or multiple race category included patients who reported a race other than American Indian or Alaska Native, Asian or Pacific Islander, Black, or White, as well as patients who reported more than 1 of these categories. A total of 450 (2.8%) and 22 (0.1%) legal intervention injury patients missing data on race and ethnicity and sex, respectively, were excluded.

Discussion

While youth are generally less likely than adults to be injured by policing practices, our findings note the protections of childhood are not afforded to all children.⁵ Black youth in California experience a substantially greater burden of injuries perpetrated by law enforcement than youth of other races and ethnicities. This is consistent with evidence that police violence is a pathway through which structural racism operates in young people’s lives, primarily impacting racially minoritized youth and contributing to health inequities.^2,3,4 Our intersectional analyses document the unique impact on Black girls, reflecting literature on how Black girls are more likely than White girls to be adultified—that is, perceived as older than they are, less innocent, and in need of less protection—with serious repercussions for more aggressive legal system targeting.⁶

This study has limitations. Research has found that death certificate data underreport when Black individuals in the US are killed by law enforcement.⁴ If similar patterns are present in hospital administrative data, the racial inequities documented in this study may be underestimated. Although sensitivity analyses confirmed result robustness across the ICD transition (eMethods in the Supplement), it is possible that ICD coding changes influenced post-transition rates. While California has a large, diverse population, future work should document legal intervention injury among youth in other contexts.

Joining in the precedent set by organizing movements, leading health organizations have recently issued guidance on systems-level interventions to address police violence and its implications for youth.^1,3 Concurrently, a new pediatric framework recommends well-child questions about police interactions.³ Clinicians can serve an important role in documenting these incidents, providing compassionate care and connection to services, and advocating for structural intervention.

Supplement.

eMethods

eReferences