Youth in detention and handguns

JOURNAL OF ADOLESCENT HEALTH 1993;14:350-355

ORIGINAL ARTICLE

Youth in Detention and Handguns C H A R L E S M. C A L L A H A N * ! M . D . , M . P . H . ! FREDERICK P. R I V A R A r M . D . , M . P . H . ! A N D J A M E S A. FARROWr M . D .

The objective of this work was to describe the frequency and correlates of handgun ownership and firearm injury experiences among youth in a detention facility. A convenience sample of 89 males detained in the King County Youth Services Center, a short-term holding facility, voluntarily completed an anonymous survey. Results showed that 59% of the youth reported owning a handgun. Firearm experiences included carrying a gun to school (46%); firing at another person (68% of handgun owners); firearm injury to self (27%); death of a close friend or family member to firearm homicide or suicide (35%). Personal safety far exceeded recreational use of guns as motivation for self-arming (52% versus 4%). Handgun ownership was more common among youth who reported problem behaviors. Adjusting for age and controlling for covariation of the problem behaviors, gang membership [odds ratio (OR) 6.7; 95% confidence interval (CI) 1.7-26.1], committing assault and battery (OR 7.7; 95% CI 2.2-26.8) and selling drugs (OR 3.6; 95% CI 0.99-13.4) were associated with handgun ownership. Our conclusion was that firearm experiences for youth in detention pose health risks equal to or far exceeding other high-risk behaviors in this population.

KEY WORDS:

Handguns Adolescents Incarcerated Firearm injuries

Firearm fatalities rival motor vehicles as leading causes of death among adolescents in the United States (1,2). American adolescents, particularly males, have high rates of firearm homicide, suicide, and unintentional injuries (3,4). In one study, 41% of all urban pediatric (1-18 years of age) deaths were attributed to homicide, the majority firearm-related deaths among minority adolescents (5). The risk of violent death among youth who have been previously incarcerated is 76-fold greater than that in the general population (6). In a recent study, handgun ownership was found to be highest in youth reporting criminal behaviors (7). Handgun ownership was more prevalent in youth reporting gang membership, sentencing by a judge, selling drugs and committing assault, suggesting that handgun ownership would be more common in youth in detention facilities. Although public health reports have focused on the unmet medical and mental health needs of incarcerated youth (8,9), few reports have focused on the firearm experiences of detained youth. The purposes of this report were to describe firearm-injury experiences, the prevalence of handgun ownership and its association with problem behaviors, and reasons for handgun ownership in a group of incarcerated youth.

Methods

Sample From the Departments of Pediatrics (C.M.C., F.P.R., J.A.F.), Epidemiology (F.P.R.), and Medicine (J.A.F.), University of Washington, Seattle, Washington. Address correspondence to: Charles M. Callahan, M.D., M.P.H., Department of Emergency Medicine, University of Rochester School of Medicine and Dentistry, Strong Memorial Hospital, Rochester, NY 14642. Manuscript accepted December 14, 1992. 350 1054-139X/93/$6.00

During the fall of 1990, we conducted a survey among male adolescents (15-18 years of age) detained in the King County Youth Services Center (Seattle, Washington), a short-term (i.e., less than 30 days) holding facility for youth awaiting or appealing adjudication. Status offenders (offense as a

© Society for Adolescent Medicine, 1993 Published by Elsevier Science Publishing Co., Inc., 655 Avenue of the Americas, New York, NY 10010

July 1993

consequence of age) are not housed at the center; almost all of the residents are juvenile felons. During 1990, 3734 males (85% of admissions) were detained, 978 for the first time. The average length of stay was 9.5 days. Forty-five percent of detained youth were white, 43% African-American, and 12% other minority groups. Sixty-five percent were Medicaideligible. During 10 sessions spanning 8 weeks, volunteers from different residential units were recruited for the survey and escorted to a classroom. All four units were surveyed on a rotating basis. During a session, the population in a surveyed unit represented 10%30% of the daily facility census. Correctional officials remained outside the classroom, and participants were paid five dollars. Individuals who did not read or write English were excluded. Data were collected by anonymous questionnaire administered by one of the authors (CMC). Written and verbal assurances of confidentiality were given, and the questionnaires were collected in unlabeled envelopes in a ballot box. This study was approved by the Human Subjects Review Committee of the University of Washington.

Survey Instrument The 80-item self-report questionnaire (available upon request) measured socioeconomic status (SES), drug/alcohol use, driving behavior, group memberships, social activities, criminal and deviant behaviors, ownership and use of firearms, perceived neighborhood safety, and firearm injuries to self and family members. Social status was measured with the Hollingshead's Two-Factor Index of Social Position (10) based on the occupation and highest grade of education of household wage earners. Weighted partial scores of the occupation and educational level of the wage earner were combined to generate a social class from I (highest) to V (lowest). The wage earner in the home with the highest occupational and educational scores was coded. Deviant behaviors during the past 3 years were ascertained using questions from the Seattle SelfReport Instrument (11). This instrument was developed in 1975 and validated in 14,000 youth. The principal outcome variable was handgun ownership, which included personal ownership or shared ownership with friends. Internal consistency of responses to handgun questions with similar content was assessed.

YOUTHIN DETENTIONAND HANDGUNS

351

Data Analysis Bivariate analyses were conducted using the Yates corrected X2 statistic or two-tailed Fisher Exact Test. The ×2 test for linear trend examined the association of handgun ownership with social class and the levels of neighborhood gunfire. The main effect measure, the odds ratio (OR) with Cornfield's 95% confidence intervals (95% CI), represents the ratio of the odds of handgun ownership in those with a selected behavior to the odds of ownership in those without the behavior. Logistic regression analysis (SPSS-PC 4.01, SPSS Inc. Chicago, IL) generated adjusted odds ratios, controlling for age and adjusting for the problem behaviors that were independently associated with handgun ownership.

Results Demographic Characteristics Participation from individual units was high (more than 90%). Excluded from participation were four youths under 15 years of age and one teenager unable to read English. An undetermined number of youth were admitted and discharged between survey dates. During the 8-week period, 89 males participated. The mean age of the participants was 16.6 (SD = 0.7) years. The sociodemographic characteristics of the sample are shown in Table 1. Forty-two percent described themselves as African-American, 36% as white, 11% as Asians or Pacific Islanders. The remaining 10% included Hispanics, Native Americans/ Alaskan Natives, and individuals of multiracial/ethnic background. Thirty-five (39%) of the subjects were unwilling or unable to report the occupation and/or the education level of the primary wage earner in their home, precluding calculation of SES for them. Internal consistency of questions relating to firearm use was high. For example, all participants who reported carrying a handgun to school also reported having ever held a handgun. Nineteen participants reported firing a gun at someone in their home, and elsewhere on the questionnaire 16 (84%) of them reported "firing at another person."

Handgun Ownership Handgun ownership (Table 1) was reported by 50 of the 84 participants who answered this question. Of the 50 handgun owners, 78% reported personal ownership, and 22% had shared ownership with

CALLAHANET AL.

352

JOURNALOF ADOLESCENTHEALTHVol. 14, No. 5

Table 1. Youth in Detention Demographic

Characteristics and Associations with Handgun Ownership Parameter Age (years)

15 16 17 18 Racial/ethnicgroup White Asians/PacificIslanders African-Americans

na

89 (8) (35) (48) (9) 88 (36) (11) (42) (10)

Others

Social class

(%)

54

na 84b 6 31 41 6 84c 31 10 34 9 52b

Handgun ownership(%) (67) (74) (56) (0) (55) (70) (59) (67)

I

(0)

0

--

II Ill W V

(17) (28) (40) (15)

8 15 21 8 84d 29 40 15

(37) (73) (57) (100)

Neighborhood gunfire

Never Sometimes

Often

89 (35) (46) (19)

(41) (67) (73)

a(89-n) = Missing

bPair-wisecomparisonsand test for linear trend, p >0.05. cPair-wisecomparisons,p >0.05. dTest for linear trend, p <0.05

friends. Prevalence of reported handgun ownership was similar across age and racial/ethnic groups. Handgun ownership was least common in the higher social classes and most common in the lowest SES group 0(2 trend statistic 3.5, p = 0.06). Subjects who reported often hearing gunfire in their neighborhoods (at least 1-2 times/week) reported an almost twofold higher frequency of handgun ownership than those from neighborhoods where gunfire was never heard 0(2 statistic for trend 5.3, p = 0.02). Firearm Experiences

Youth in detention used the handguns they owned and had frequent experience with firearm injuries. Nearly one-half (41/88) of the detainees and 70% (35/ 50) of the handgun owners reported carrying a gun to school within the last 3 years. Similarly, almost one-half of the detainees and 68% of the handgun owners reported shooting at another person. Seventy-eight percent of the detainees reported having been threatened by someone with a weapon. The odds of owning a handgun was fourfold greater (OR, 95% CI; 4.0, 1.3-12.1) among detainees who

reported such a threat compared to those who had not been threatened with a weapon. Perceived improved personal safety far exceeded recreational reasons (target shooting or hunting) as the motivation given for handgun ownership (52% versus 4%). Feelings of vulnerability were based on experience as 15% of subjects were victims of driveby shootings, and 5% reported being shot as innocent bystanders. An additional 7% were injured by "accident" when handling or cleaning a firearm. Thirty percent of the participants reported the loss of a family member or close friend to firearm homicide or suicide. Almost one in five detainees reported that the possession of a handgun played a significant role in their arrest and detention.

Problem Behaviors and Handgun Ownership

Problem behaviors were more common in the youth who reported handgun ownership (Table 2). After controlling for age; gang membership, committing assault and battery, and selling drugs were found to be independently associated with handgun ownership (Table 2). Problem driving behaviors, chemical abuse, and other school problem behaviors did not increase the odds of gun ownership. Although gang members were 6.7-fold more likely to own a handgun than non-gang members, only 52% (26/50) of handgun owners reported gang membership. Most of the handgun owners reported selling drugs (86%) or committing assault and battery (84%), presumably outside the context of gang activity.

Discussion Data from this report indicate that the lives of detained youth are replete with previous experiences with firearms. Prior work has demonstrated an association of handgun ownership with deviant behaviors (7). Our data confirm that deviant youth (by nature of their incarceration) are intimately involved with handguns. Unlike the characterization of high school student handgun owners (7), within a group of detained youth, handgun ownership and deviant behaviors are so common that it is difficult to separate handgun owners from non-owners based on reported deviant behaviors or demographic characteristics. High rates of violence-related injuries (12,13) and problem behaviors like chemical dependency and high-risk sexual practices are known to be more prevalent among incarcerated youth (1416). Ownership of handguns must be added to the

July 1993

YOUTH IN DETENTION AND HANDGUNS

353

Table 2. Deviant Behaviors a n d H a n d g u n Ownership

Problem behaviors Deviant behaviors Gang member Yes No Assault and battery Yes No Robbery Yes No Sold drugs Yes No School Striking teacher Yes No Suspensions Yes No "High" in school Yes No Driving Motor vehicle violations >- 2 1 None Driving "high" Yes No Chemical abuse Cocaine Ever Never Marijuana Monthly < Few Times

n

Ownership (%)

ORa

(95% CI)

aOR ~

(95% CI)

30 54

87 44

8.1

(2.5-26.2)

6.7

(1.7, 26.1)

57 27

73 30

6.6

(2.4-18.3)

7.7

(2.2, 26.8)

30 54

73 52

2.5

(0.9-6.7)

NSc

63 21

68 33

4.3

(1.5-12.3)

3.6

19 65

79 54

3.2

(0.9-10.7)

NS

67 17

61 53

1.4

(0.5-4.1)

NS

61 23

62 52

1.5

(0.6-4.0)

NS

36 15 33

61 53 24

0.9 0.7

(0.3-2.5) (0.2-3.2)

NS NS

57 27

61 56

1.3

(0.5-3.2)

NS

20 64

40 65

0.3

(0.1-1.0)

NS

65 29

61.5 44

2.5

(0.9-7.1)

NS

(0.99,13.4)

CI, confidence interval. °Odds ratio--the ratio of the odds of handgun ownership in those with a selected problem behavior, to the odds of handgun ownership in those without the behavior. bAdjusted Odds Ratio---logistic regression model controlling for age with gang membership, assault, and selling drugs as independent variables. CNot significant--with age, gang membership, assault, and selling drugs in the model, no other problem behaviors would enter the equation, individually, or in aggregate.

highly correlated risk behaviors present in detained youth. There are a number of limitations of this descriptive study. The nonrandom small convenience sample restricts generalizability. Youth with short periods of incarceration, non-English-speaking or -reading minority groups, and youth who refused participation were not surveyed. The firearm experiences for these groups are, therefore, unknown.

The small number of subjects studied precludes precise estimation of the association of handgun ownership and problem behaviors, but the magnitude of the association and the co-variation of the problem behaviors make it unlikely that the association would disappear or reverse. Studies with more subjects in youth detention facilities across the country are needed to validate these preliminary findings and improve the precision of the effect measure.

354

CALLAHANET AL.

The lack of external validity and reliability measures suggests caution in interpretation of findings despite the internal validity observed, the cooperation we witnessed, and our efforts to promote honesty and ensure confidentiality. The tendency for deviant male youth to underreport criminal behaviors (11) suggests the association between h a n d g u n ownership and criminal behavior may be underestimates. Further, the cross-sectional design with simultaneous measurement of the outcome (handgun ownership) and exposure variables (gang membership, selling drugs, neighborhood gunfire, etc.) precludes the delineation of temporal relationships. These limitations notwithstanding, there are a number of public health implications from this report. These youth are at high risk for sustaining or inflicting firearm injuries and consequently suffering both physical and psychological harm. The frequent firearm experiences reported by youth in detention identifies them as prime targets for testing violenceprevention strategies (17,18). In the facility studied, the tragic toll of firearm injuries has been added to the other health education topics presented to the detained youth during their stay at the center. The frequency of serious injuries to self, family members, and friends has implications for the mental health needs of detained youth. Grief and posttraumatic stress disorder should be added to the list of mental health disorders likely to be found in incarcerated youth, as either primary or secondary diagnoses. Many reports make note of the high prevalence of violent traumatic experiences in the lives of urban youth (19-22), and some have called for the integration of violence risk assessment and counseling into the routine health care for adolescents (23). Data from this report would support this recommendation, especially for incarcerated youth. Having access to a lethal weapon may impart a sense of greater safety to youth who are more vulnerable to violence. The tragic cost of this contradiction (handguns imparting feelings of safety) is reflected in the firearm mortality data, placing this age group among those at highest risk for firearm death, including homicide, suicide, and unintentional firearm fatalities (1,2). The availability of handguns in the home increases the risk of firearm death (24); and, among the adolescent-young adult age group, both firearm homicide (25) and firearm suicide (26) increase with the availability of handguns. A recent report which found no association between the prevalence of handguns and homicide rates in a predominately white, non-urban population with relatively low homicide rates, discounted

JOURNAL OF ADOLESCENTHEALTHVol. 14, No. 5

youth firearm ownership with the statement "the number of handguns possessed by minors and not also considered the property of an adult in the same household must be negligible" (27). It is u n k n o w n if youth h a n d g u n ownership is highly dependent on parental firearm ownership in non-urban white populations, but the data from this report and others (7) suggest that urban youth aquire handguns independent of their parents. Detained youth, at odds with the legal and judicial systems, are likely to be the members of society with the least confidence in police ability and willingness to provide protection. Based on their experiences with gunfire in their environments, youth may believe the only recourse is to own and be willing to use a gun, abandoning collective institutionalized means of security in favor of individualistic solution (28). Given that the acquisition of handguns by anyone younger than 21 years of age is a federal crime, the prevention of h a n d g u n injuries, among adolescents will require changes beyond the legal restrictions. It must address the plethora of weapons that are so easily available and widely distributed in our society. Funding provided by the National Institute of Health, National Research ServiceAward, 5T32 PE1002, and the Centers for Disease Control, R49/CCR002570. We are grateful to Phil Robbins of the King County Youth Services Center for his support and cooperation for this project and for his tireless dedication, optimism, and sense of fairness for the troubled youth of King County.

References 1. Fingerhut LA, Kleinman JC, Godfrey MS, Rosenberg H. Firearm mortality among children, youth and young adults 1-34 years of age, trends and current status: United States, 197988. Final data. National Center for Health Statistics 1991; 39:11. 2. Guyer B, Ellers B. Childhood injuries in the United States. Am J Dis Child 1990;144:649-52. 3. Fingerhut LA, Kleinman JC. International and interstate comparisons of homicide among young males. JAMA 1990; 263:3290-5. 4. American Academy of Pediatrics, Committee on Injury and Poison Prevention. Firearm injuries affecting the pediatric population. Pediatrics 1992;89:788-9. 5. Ropp L, Visintainer P, Uman J, Treloar D. Death in the city: An American childhood tragedy. JAMA 1992;267:2905-10. 6. Yeager CA, Lewis DO. Mortality in a group of formerly incarcerated juvenile delinquents. Am J Psychiatry 1990;147: 612-4. 7. Callahan CM, Rivara FP. Urban high school youth and handguns, a school-based survey. JAMA 1992;267:3038-42. 8. Farrow J. Medical responsibility to incarcerated children. Clin Pediatr 1984;23:699-700.

July 1993

9. Jameson EJ. Incarcerated adolescents. The need for the development of professional ethical standards for institutional health care providers. J Adolesc Health Care 1989;10:490-9. 10. Hollingshead AB, Redlich F. Social class and mental illness. New York, Wiley 1958:387-97. 11. Hindelang MJ, Hirshehi T, Weis JG. Measuring delinquency. Beverly Hills, CA: Sage, 1981:123. 12. Lewis D, Shanok S, Pincus J, et al. The medical assessment of seriously delinquent boys. J Adolesc Health Care 1982; 3:160--4. 13. Woolf A, Funk SG. Epidemiology of trauma in a population of incarcerated boys. Pediatrics 1985;75:463-8. 14. DiClemente RJ, Lanier MM, Horan PF, Lodico M. Comparison of AIDS knowledge, attitudes, and behaviors among incarcerated adolescents and a public school sample in San Francisco. Am J Public Health 1991;81:628-30. 15. Nader PR, Wexler DB, Patterson TL, et al. Comparison of beliefs about AIDS among urban, suburban, incarcerated, and gay adolescents. J Adolesc Health Care 1989;10:413-8. 16. RolfJ, Nanda J, Baldwin J, et al. Substance misuse and HIV/ AIDS risks among delinquents: A prevention challenge. Int J Addict 1990-1991;25:533-59. 17. Prothrow-Stith D. Deadly Consequences. New York, HarperCollins, 1991. 18. Jenkins EJ, Bell CC. Adolescent violence: Can it be curbed. Adolesc Med State Art Rev 1992;3:71-86. 19. Gladstein J, Slatter Rusonis EJ, Heald FP. A comparison of

YOUTH IN DETENTION AND HANDGUNS

20.

21. 22. 23.

24.

25.

26.

27.

28.

355

inner-city and upper-middle class youth's exposure to violence. J Adolesc Health 1992;13:275-80. Price JH, Desmond SM, Smith D. A preliminaryinvestigation of inner city adolescents' perceptions of guns. J Sch Health 1991;61:255-9. Pynoos RS, Nader K. Children's exposure to violence and traumatic death. Psychiatric Annals 1990;20:334-344. Zinsmeister K. Growing up scared. Atlantic Monthly 1990; 6:49-66. Stringham P, Weitzman M. Violence counseling in the routine health care of adolescents. J Adolesc Health Care 1988; 9:389-93. Kellermann AL, Reay DT. Protection or peril? An analysis of firearm-related deaths in the home. N Engl J Med 1986; 314:1557-60. Sloan JH, Kellermann AL, Reay DT, et al. Handgun regulations, crime, assaults and homicide: A tale of two cities. N Engl J Med 1988;319:1256-62. Sloan JH, Rivara FP, Reay AT, et al. Firearm regulations and rates of suicide: a comparison of two metropolitan areas. N Engl J Med 1990;322:369-73. Centerwall BS. Homicide and the prevalence of handguns: Canada and the United States, 1976-1980. Am J Epidemiol 1991;134:1245-60. Young RT, McDowall D, Loftin C. Collective security and the ownership of firearms for protection. Criminology 1987; 25:47-62.