Reducing Weapon-Carrying Among Urban American Indian Young People

Journal of Adolescent Health 47 (2010) 43–50

Original article

Reducing Weapon-Carrying Among Urban American Indian Young People Linda H. Bearinger, Ph.D.a,b,*, Sandra L. Pettingell, Ph.D.a, Michael D. Resnick, Ph.D.b, and Sandra J. Potthoff, Ph.D.c a Center for Adolescent Nursing, School of Nursing, University of Minnesota, Minneapolis, Minnesota Division of Adolescent Health and Medicine, Department of Pediatrics, Medical School, University of Minnesota, Minneapolis, Minnesota c Division of Health Policy and Management, School of Public Health, University of Minnesota, Minneapolis, Minnesota Manuscript received September 4, 2009; manuscript accepted December 30, 2009

b

Abstract

Purpose: To examine the likelihood of weapon-carrying among urban American Indian young people, given the presence of salient risk and protective factors. Methods: The study used data from a confidential, self-report Urban Indian Youth Health Survey with 200 forced-choice items examining risk and protective factors and social, contextual, and demographic information. Between 1995 and 1998, 569 American Indian youths, aged 9–15 years, completed surveys administered in public schools and an after-school program. Using logistic regression, probability profiles compared the likelihood of weapon-carrying, given the combinations of salient risk and protective factors. Results: In the final models, weapon-carrying was associated significantly with one risk factor (substance use) and two protective factors (school connectedness, perceiving peers as having prosocial behavior attitudes/norms). With one risk factor and two protective factors, in various combinations in the models, the likelihood of weapon carrying ranged from 4% (with two protective factors and no risk factor in the model) to 80% of youth (with the risk factor and no protective factors in the model). Even in the presence of the risk factor, the two protective factors decreased the likelihood of weaponcarrying to 25%. Conclusions: This analysis highlights the importance of protective factors in comprehensive assessments and interventions for vulnerable youth. In that the risk factor and two protective factors significantly related to weapon-carrying are amenable to intervention at both individual and population-focused levels, study findings offer a guide for prioritizing strategies for decreasing weapon-carrying among urban American Indian young people. Ó 2010 Society for Adolescent Health and Medicine. All rights reserved.

Keywords:

Adolescents; American Indian; Urban; Weapon-carrying; Young people

Violence, manifested as intentional injuries, continues to be one of the nation’s most urgent health problems of young people. The availability of weapons in the context of violent confrontations or suicidal thinking markedly boosts the

The contents of the article are solely the responsibility of the authors and do not necessarily represent the official views of the National Center for Injury Prevention and Control. *Address correspondence to: Linda H. Bearinger, Ph.D., M.S., Center for Adolescent Nursing, School of Nursing, University of MN, 5-140 WeaverDensford Hall, 308 Harvard Street, S.E., Minneapolis, MN 55455. E-mail address: [email protected]

likelihood of fatal injury [1]. Weapon-carrying substantively contributes to mortality and constitutes a critical dimension of interpersonal violence [2–4]. It is also a marker for community-level and school-based violent, aggressive behavior [3]. Overall, prevalence of weapon-carrying among high school youth (grades 9–12) declined between 1991 (26.1%) and 1997 (18.3%), leveling off through 2003 (17.1%) [5]. However, reported in the 2005 Youth Risk Behavior Survey, the percentage of young people who carried a weapon (e.g., gun, knife, or club within the last 30 days) returned to 1997 levels: 18.5% overall (29.8% males, 7.1% females) [6]. Weapon-carrying in school was

1054-139X/$ – see front matter Ó 2010 Society for Adolescent Health and Medicine. All rights reserved. doi:10.1016/j.jadohealth.2009.12.033

44

L.H. Bearinger et al. / Journal of Adolescent Health 47 (2010) 43–50

less prevalent: 6.5% overall (10.2% boys, 2.6% girls) had carried a gun ‘‘1 of the last 30 days’’ [6]. As with most studies on weapon-carrying, gender differences were evident for all grades and within black, white, and Hispanic subpopulations [7, 8]. No Youth Risk Behavior Survey data are available, specifically for American Indian adolescents [6]. In another nationally representative sample of middle and senior high school students, one in five middle school and more than one in four high school students reported easy access to firearms at home [9]. Ease of access was associated significantly with increased self-directed and interpersonal violence, after controlling for gender, race, ethnicity, poverty status, and family structure. Weapon-carrying at school, far more common among boys, is associated with fighting behavior and substance use, including tobacco, alcohol, marijuana, and other illicit drugs [2, 10, 11, 12, 13]. Nationwide, 8.9% of students had been threatened or injured with a weapon on school property more than once in a year [7]. Other issues such as early sexual activity [2], seat-belt nonuse [14], poor academic performance [12, 14], and skipping school [2] also have been associated with weapon-carrying. Although some studies have found prior victimization or fear of violence as significant motivators for weapon-carrying [3, 8, 10, 15, 16], others conclude that weapon-carrying more strongly associates with violent perpetration and other delinquent behaviors [12, 13]. Many of the Healthy People 2010 objectives for selfdirected and interpersonal violence specifically target youth, including the goal of reducing in-school weapon-carrying from a baseline of 6.9% (in 1999) to 4.9% among 9th–12th graders (Objective, 15–39) [17]. The youth violence focus in Healthy People 2010 also emphasizes the disproportionate risks among youth of color, including American Indian and Alaska Native. Although a plethora of information focuses on adolescent weapon-carrying and violence involvement in the general population, data on urban American Indian youth are lacking. What is known is that suicide rates among American Indian adolescents are more than twice the national rate for all adolescents [18, 19]. Homicide rates for American Indian or Alaska Natives represent >3 times the rate for all teenagers and >5 times the rate among Caucasians [20]. High levels of violence among American Indian adolescents have received growing attention; however, most extant research has been conducted on reservations [20–22]; little is known about those living in urban areas. A mid-1980s study in one metropolitan area, comparing health indices among urban youth of color, showed that American Indians had higher rates of physical or sexual abuse as well as tobacco, alcohol, and crack/cocaine use [23]. They also expressed more negative body image and worry over a parent’s death. In terms of the presence or absence of protective factors in their lives, urban American Indian boys and girls alike indicated the lowest levels of connectedness with family, and felt the least cared for by parents, teachers, and other adults. Coupled with a sense of disaffection from adults in their

primary social systems, they reported lower grade point averages and a greater dislike of school than any other racial/ ethnic group [23]. With few exceptions, this mid-1980s profile of urban American Indian adolescents was one of elevated risk, diminished protective factors, and negative health indices exceeding any other racial/ethnic group of young people. A decade later, data from an urban American Indian youth health survey (the survey used in the present study) [18, 24] revealed prevalence rates for self-directed and interpersonal violence and risk of injury from firearms in excess of levels documented in either the reservationbased survey [20] or the comparative urban youth health study described earlier in the text [23]. A resiliency paradigm that identifies factors protecting against health-jeopardizing behaviors and adverse outcomes guided our study. In the paradigm, health and social outcomes emanate from health and social behaviors. These, in turn, result from the dynamic interplay of environmental attributes (e.g., parent and peer norms), behavior attributes (e.g. competence, coping), and personal attributes (e.g., cognitions, self-beliefs, connectedness). Seminal works on resilience, see Werner [25] and Rutter [26], have identified relationships between these characteristics, health and social behaviors, and subsequent health and social outcomes. Using this paradigm, we explored the most significant risk and protective factors correlated with weapon-carrying. With these factors identified, we determined the likelihood of weapon-carrying given various combinations of risk and protective factors. We hypothesized that the probability of weapon-carrying would increase as the number of risk factors increased, but that in profiles with one or more protective factors, the protective effect would off-set the increased risk and provide a decrease in the probability of weapon-carrying. Methods Participants Detailed previously [18, 24], 577 participants completed the Urban Indian Youth Health Survey (UIYHS) with a usable sample of 569. All participants, self-identified as American Indian, either attended schools within a single metropolitan school district, or were enrolled in a nearby culturally-focused after-school youth program. Survey administration (during school hours or after-school program hours) occurred between 1995 and 1998. The University of Minnesota Institutional Review Board and the district school board approved study protocols. Parents granted permission for their children (96.4%); all youth assented. Instrument Along with social, contextual, and demographic information, the UIYHS assessed an array of health-compromising behaviors and assets, strengths, or protective factors. Adolescent health experts in consultation with representatives from

L.H. Bearinger et al. / Journal of Adolescent Health 47 (2010) 43–50

the American Indian community developed the 200-item instrument, with many items adapted from a national survey of reservation-based youth created by the University of Minnesota in collaboration with the U.S. Indian Health Service [20, 27]. Items were revised after focus group and pilot testing, gearing toward a second grade reading-level to be suited for the youngest participants. For those requiring reading assistance, survey administrators read aloud the items from a separate survey copy (to maintain confidentiality) [18, 24]. Variables Guided by prior analyses with UIYHS [18, 24], we developed scales measuring risk and protective factors within broad domains of the resilience paradigm. Seven scales measured protective factors: (a) connectedness to others, (b) family connectedness, (c) parent prosocial behavior norms, (d) peer prosocial behavior norms, (e) perceived self-image, (f) positive affect, and (g) school connectedness. Three scales represented risk factors: (a) substance use, (b) suicidal thoughts or behaviors, and (c) violence perpetration. To receive a standardized score (range: 0–1), participants had to endorse 75% of the items in the scale. Factor analyses and Cronbach’s alpha confirmed scale dimensionality and internal consistency. Table 1 shows scales, items, and alphas [18, 24]. Participants were considered as ‘‘having carried a weapon in the past month’’, if they responded ‘‘yes’’ (i.e., ‘‘mark all that are true for you’’) to one or more of the following five response options: handgun; other guns, such as a rifle or shotgun; knife or razor; club, stick, bat, or pipe; some other weapon. Data analysis Wald chi-square tests examined bivariate relationships between weapon-carrying and risk or protective factors. Next, risk or protective factors scales achieving a significance level of p  .05 and showing potential clinical significance (e.g., with an odds ratio [OR] of <0.50 for a protective factor and >2.0 for a risk factor) were entered into the initial multivariate logistic regression models: one model for risk factors and a second for protective factors. These separate models provided the ORs identifying the risk and protective factors used for determining the probability of weapon-carrying. Then, a final multivariate model was constructed using statistically significant (defined above) variables [28, 29]. Probability profiles provided estimates of the likelihood of an outcome, given the presence or absence of varying combinations of risk and protective factors [28]. Thus, to create profiles, using the criteria of inclusion described earlier in the text, one risk and two protective factors associated with weapon-carrying were entered into a multivariate logistic regression model, providing the beta weights to form the probability profiles. Then, the estimated likelihood of

45

carrying a weapon was computed, where combinations of 0 or 1 risk factors and 0, 1, or 2 protective factors were present [28, 29]. All bivariate and multivariate analyses included age and gender as covariates, given the anticipated relationship between weapon carrying and age and gender.

Results Descriptive findings Most participants self-identified as Ojibwa/Chippewa/ Anishinabe (66.1%) with 22.9% Lakota/Dakota/Sioux, and 1.6% Winnebago; 17.2% belonged to tribes not listed in the instrument or were unsure of their tribal heritage. (Marking multiple tribal affiliations meant that percentages exceeded 100%). A total of 51.7% were female [18, 24], ranging in age from 9 to 15 years (mean age ¼ 11.9 years). Specific to the present analysis, 25.1% reported carrying a weapon in the past month; of these, 35.2% carried a gun or gun with another weapon, 35.3% a knife, 9.4% either a club, stick, bat, or pipe; 20.1% multiple weapons (not including a gun) or another weapon not specified in the open-ended response. Weapon-carrying increased with age, 8.9%–38.3% from 9 to 15 years of age, and more males carried a weapon (57.6%). Thus, age and gender were used as covariates in all analytic models. As previously published [18, 24], study participants reported other indicators of violence. Half of them (49.2%) had witnessed someone getting shot or stabbed (‘‘in real life, not on TV’’); 19.0% had witnessed this level of violence 3 times. Two of every three reported that 1 family member had been shot or stabbed (65.5%). Over half (56.6%) had a friend or family member in a gang; fewer had ever been in a gang themselves (19.7%). Although higher among boys, overall 9.6% had shot at someone with a gun, 11.5% had cut or stabbed someone, and 3.7% had done both. Nearly one-fifth (19.1%) reported 1 gun in their home. Specific to risk factors in this analysis, 28.2% reported using cigarettes, alcohol, marijuana, or a combination of these substances within the previous month; use was higher among girls. Protective factors presented a contrasting picture [18]. Though younger adolescents felt more connected to school than their older counterparts, overall 69.4% indicated liking school ‘‘some’’ or ‘‘a lot’’. A comparable proportion (72.5%) said they knew a teacher ‘‘really well’’; fewer (49.6%) ‘‘got along’’ with their teachers. Adults outside of family, which may have included teachers, provided a source of support for nearly two-thirds (63.2%). Several survey items tapped into young people’s perceptions of their parents’ perspectives, that is, perceptions of parental norms. Most (93.3%) thought that their parents would be upset if they dropped out of school. Likewise, 90.4% said their parents would be upset if they got arrested or beat up another person (63.0%). They also believed that their parents would be upset if they carried a weapon (gun [89.1%], knife [76.4%]).

L.H. Bearinger et al. / Journal of Adolescent Health 47 (2010) 43–50

46

Table 1 Independent variables with descriptions, number of items, and alphas [18, 24] Variable Connectedness to others Family connectedness Parent prosocial behavior norms Peer prosocial behavior norms Perceived self-image Positive affect School connectedness

Variable description

No. items (a)

Protective Factors Feeling other adults care about you, have other adults outside your family that you can talk to, your friends care, you can make friends, and you have friends you can talk to and/or go to for safety if needed Feeling that your mom, dad, and others in your family care about you and your feelings, praises you, understands and pays attention to you, talks with you about your future goals, and you and your family eat meals together and have fun with each other Perception of your parents’ approval or disapproval of anti-social behavior: getting arrested, beating someone up, carrying a gun or knife, getting (someone) pregnant, dropping out of school, drinking alcohol, and smoking marijuana Perception of your peers’ approval or disapproval of anti-social behavior: getting arrested, beating someone up, carrying a gun or knife, getting (someone) pregnant, dropping out of school, drinking alcohol, and smoking marijuana General health and healthiness compared to peers, life satisfaction and liking yourself, perceived weight (underweight, overweight, about right), and weight satisfaction Positive feelings about yourself with respect to mood and/or mental status, e.g., bored, sad, stressed, tired Perceptions of your relationships with teachers and other school staff (care about you, know you well, expect you to do well, you get along with them), importance of school attendance and doing well in school

6 (a ¼ .57) 10 (a ¼ .78)

8 (a ¼ .77)

8 (a ¼ .91)

6 (a ¼ 0.50) 5 (a ¼ .64) 6 (a ¼ .73)

Risk Factors Substance use Suicidal thoughts/behaviors Violent perpetrationa a

Number of days in the past month in which you used alcohol, marijuana, and cigarettes Worry about killing yourself, thoughts about killing yourself in the past month, and ever tried to kill yourself Pick a fight to feel better if have problem, when angry hit someone, number of times in the past year you hit or beat up another person, number of times in the past year you’ve been in a fight where a group of your friends fought another group

3 (a ¼ .83) 3 (a ¼ .49) 4 (a ¼ .67)

The violence perpetration scale was modified from that listed in the previously published table cited below.

The participants’ perceptions of peer norms did not parallel perceptions of parental norms, that is, peer tolerance exceeded parents. Varying numbers of youth thought their peers would be upset if they got arrested (42.4%), beat up another person (26.9%), and carried a gun (50.1%) or a knife (45.2%). Regarding dropping out of school, 63.2% thought this would upset their peers. Protective and risk factors associated with weapon-carrying Bivariate relationships (Table 2) indicated family, personal, and community characteristics related to weapon-carrying. Study participants were less likely to carry a weapon if they Table 2 Bivariate logistic regressions for protective and risk factors on carrying a weapon, with age and gender as covariates Factor Protective factor Connectedness to others Family connectedness Parent prosocial behavior norms Peer prosocial behavior norms Perceived self-image Positive affect School connectedness Risk factor Substance use Suicidal thoughts/behaviors Violence perpetration

b

Standard error

p

–.79 –1.97 –2.74 –1.95 –.72 –1.99 –5.51

.53 .55 .44 .34 .55 .57 .89

.137 <.001 <.001 <.001 .193 <.001 <.001

.456 .140 .065 .143 .487 .136 .004

2.61 2.31 4.37

.40 .43 .52

<.001 <.001 <.001

13.612 10.043 79.324

Odds ratio

felt connected to school or family, felt good emotionally, and perceived peers and/or parents as having prosocial attitudes. The scales anticipated to be risk factors, that is, associated with increased weapon-carrying, included substance use, suicidal thoughts or behaviors, and perpetrating violence. That said, perpetrating violence had a large OR indicating overlap between violence perpetration and weapon-carrying. Therefore, violence perpetration was not included in the models. On basis of the specified inclusion criteria for bivariate and initial multivariate models (detailed above), two protective factors related to weapon-carrying (school connectedness and peer prosocial behavior norms) and one risk factor (substance use) were entered into the final multivariate model (Table 3). Probability profiles To weigh the overall contribution of each risk and protective factor in the final multivariate logistic regression models, weapon-carrying probabilities were computed using scores reflecting the 10th and 90th percentile for each scale [29]. In Figure 1, among youth characterized by the two protective factors (school connectedness and peer prosocial norms), and no risk factor (substance use), 4% would be expected to carry a weapon. Likelihood of carrying a weapon increased as the risk factor was added to the profile with two protective factors still in the model. The presence of the risk factor in combination with the two protective factors resulted in an almost sixfold increase (4%–25%) in the probability of carrying a weapon.

L.H. Bearinger et al. / Journal of Adolescent Health 47 (2010) 43–50 Table 3 Final multivariate logistic regression model for protective and risk factors on carrying a weapon, with age and gender as covariates Factor

b

Standard error

p

Odds ratio

Peer prosocial behavior norms School connectedness Substance use Age Gender Constant

–1.40 –3.18 2.18 .06 .65 –.34

.46 1.21 .46 .10 .32 1.44

.002 .008 <.001 .546 .039

.247 .042 8.874 1.061 1.924

47

and Orpinas et al. [30], in the presence of risk, other forces protect adolescents from harm, particularly the critical role of both family and school contexts. Even after controlling for demographic characteristics, social contexts still count [9, 30]. Likewise, in the present study, school connectedness, characterized by the presence of caring teachers and high expectations for student performance, was protective for these urban American Indian youth. So, too, was the positive effect of perceived peer prosocial norms that discouraged violence involvement, substance use, pregnancy, and dropping out of school. Our findings mirror previously published works that examined overall profiles for suicidal attempt and violence perpetration [18, 24], respectively, within this urban American Indian youth sample. In these analyses, substance use was significantly associated with suicidal attempts and violence perpetration. Similarly, peer and/or parental prosocial behavior norms and school connectedness played significant protective roles, associated with lower violence perpetration and suicidal attempts (males only). Initial analyses of this urban American Indian dataset revealed disturbingly high levels of violence involvement— violence against others (perpetration) [24] as well as selfdirected violence (suicide) [18]. Likewise, teachers, counselors, and youth leaders in the schools and communities serving the young people in our study expressed growing concern about violence-related issues. In the initial analysis on violence perpetration [24], we excluded weapon-carrying as a potential risk factor for three reasons. First, given the specific survey items used to measure weapon-carrying, we, like others using survey data, could not distinguish between offensive or defensive reasons for having a weapon [31]. Secondly, weapon-carrying was moderately correlated

Final model c2 (5 degrees of freedom) ¼ 103.433, p < .001.

The fewer protective factors in the models, the higher the probability for weapon-carrying. Again, in Figure 1, with no risk factor in the profile, the reduction in protective factors raised the probability of weapon carrying from 4% to 31%. Likewise, with fewer protective factors and the risk factor in the profiles, probabilities of weapon-carrying increased from 25% to 80% (worst-case scenario). For the profile with no risk factor yet no protective factors, there was still a 31% chance a person would carry a weapon. That was almost eight times greater than the probability for those who had no risk but had two protective factors (4%). Discussion As found in the probability profiles, urban American Indian young people’s perceptions of their peers’ prosocial norms and their connections to school served as primary protections against weapon-carrying. In other words, connections to school, and, from the participants’ perspectives, having peers with prosocial attitudes substantially lowered probability of weapon-carrying. As in Resnick et al. [9]

Probability Profiling for Weapon Carrying

80%

Protective Factors

No Protective Factors

31%

58%

School Connections

13%

Peer Prosocial Norms

49% 10%

Risk Factors

Substance use Both Protective Factors

No substance use

25% 4%

0

0.1

0.2

0.3

0.4

0.5

Probability of Carrying a Weapon

Figure 1. Probability profiling for weapon carrying.

0.6

0.7

0.8

0.9

48

L.H. Bearinger et al. / Journal of Adolescent Health 47 (2010) 43–50

with other variables of interest in our analysis of violence perpetration such that its exclusion from the models in this prior analysis allowed for testing the salience of those other variables. Third, and perhaps most importantly, the focus of the present analysis on factors associated with weaponcarrying represents a distinct area for developing prevention approaches. In other words, perhaps the most effective strategies for reducing violence will be to focus interventions on reducing the number of young people who carry weapons, for whatever reasons. A companion approach to violence reduction, then, might be to focus separately on prevention of violence perpetration and suicide, considering salient risk and protective factors for each [18, 24]. Whether examining salient influences for weaponcarrying, violence perpetration, or suicide risk, the attitudes and beliefs of those in their social contexts of young people are vitally important. After an adolescent’s sense of self is steered toward the attitudes and beliefs of a specific, and often admired, peer referent group, young people’s behaviors often reflect the values (both positive and negative) needed to associate and bond with members of that peer group [32]. Similarly, parents, siblings, extended family, and other adults, play a central role in the development of attitudes and in guiding life- choice decisions [32]. In the present study, both peer and parent attitudes/beliefs, as perceived by the study participants, played a protective role against weapon-carrying. In contrast to protective factors, using substances was associated with an elevated risk of weapon-carrying. Consistently, studies have shown similar associations between access to weapons and violence involvement against self and/or others [3, 9, 12, 33, 34]. Also revealed in our findings, levels of violence involvement among urban American Indian youth who carried a weapon exceeded those found in a national survey of rural, reservation-based American Indian youth [33]. Moreover, our findings substantiate the commonly-documented role of substance use in potentiating risk, in this case the likelihood of weapon-carrying [34–36]. From a public health perspective, if young people use alcohol, marijuana, and/or other illicit drugs, and have a history of violence involvement, the presence of weapons significantly elevates cause for concern. With their potential for lowering inhibitions and increasing aggression, substance use combined with access to a weapon can create a lethal situation that might otherwise have been prevented [31, 37]. Limitations As previously detailed [18, 24], several limitations should be considered. First, our results report on group findings and should not replace careful clinical assessment. Second, only in-school youth enrolled in this study; findings cannot be generalized to all American Indian young people. Third, cross-sectional data cannot predict. Fourth, lower alphas were likely the consequence of two factors: few items in some scales [3-5] and limited response options [2-3] for

some items [38]. Reducing items and response options were deliberate survey adaptations responding to feedback from focus groups and community advisors, who cautioned against a lengthy and complex instrument for those aged <9 years. Finally, although our findings can offer direction for intervention development, all interventions should be appropriately evaluated and replicated so that generalization to broader groups is possible. Implications Probability profiling is an approach that can significantly improve the assessment of vulnerability. Here, it identified a set of risk and protective factors for weapon-carrying in urban American Indian youth, most of which are amenable to intervention at both individual and population-based levels. Our results emphasize that those who are more likely to carry a weapon also are more apt to use substances [31]. They also show, however, that school connectedness and peer prosocial norms can buffer against this behavior. From a programmatic perspective, probability profiling provides a guide for prioritizing interventions most likely to succeed, in this case, those which are grounded in the dual strategy of reducing risk factors, while enhancing protective factors [24, 39]. This analysis underscores the value of focusing on protective factors in comprehensive assessments, particularly with highly vulnerable youth, such as those in this study. In light of this, health care professionals have a responsibility to inquire about school attitudes and experiences, academic expectations, and interactions with teachers and classmates. Likewise, the sense that young people have about behavioral expectations, from parents and/or peers, is an essential part of clinically assessing current health status and potential vulnerabilities. Beyond individual guidance, population-focused interventions must respond to the growing knowledge about effective strategies for lowering risk and promoting protection [30]. Given the co-occurrence of substance use and violence involvement, programs focused on reducing weapon-carrying should address all of these behaviors rather than aim at a single, categorical risk [27, 40]. Furthermore, prevention and intervention programs, schools, religious institutions, and other youth-focused organizations need to invest in adding resources to the lives of youth that may help to effectively counterbalance risk factors present in their lives.

Conclusion Current research clearly indicates that the availability of weapons escalates the lethality of violent actions [28, 33]. Unquestionably, having weapons available during violent confrontations markedly boosts the likelihood of severe or fatal injury. Restricting the accessibility of guns has been

L.H. Bearinger et al. / Journal of Adolescent Health 47 (2010) 43–50

identified as the best strategy to prevent suicide and other violent outcomes. Moreover, greater effort can be given to sending messages about the negative effect of using substances and the damaging consequences of irresponsible use of firearms [15]. However, equally important, connections to school need to be promoted, and youth need to be directed to peer groups that advocate prosocial behavior. Our findings suggest that these protective factors can make a difference in how young people navigate the stressful experiences they encounter in everyday life.

Acknowledgments The authors thank the following for their roles and contributions to the original R01 study co-investigator Marcia Shew, M.D., M.P.H. formerly of the Division of Adolescent Health and Medicine, Medical School, University of MN; R01 study community program specialists, John Eichhorn, B.A., Margaret Dexheimer Pharris, Ph.D., M.P.H., M.S. The authors also appreciate the statistical and editorial assistance of Carol Skay, Ph.D., Elizabeth Saewyc, Ph.D., and Lisa Martin-Crawford, M.S. This study and its authors were supported in part by grants NIH/1R01-NR03562-01A1, (L.H.B.) from the National Institute of Nursing Research, NIH; NIH/1R03-MN601-02, (L.H.B.) from the National Institute of Mental Health, NIH; MCJ279185 (L.H.B.) from the Maternal and Child Health Bureau, (Title V, Social Security Act) Health Resources and Services Administration; T01-DP000112 (L.H.B.) from the Centers for Disease Control and Prevention (CDC); and R49/CCR511638-03-2 (M.D.R.) from the National Center for Injury Prevention and Control, CDC.

References [1] Borowsky IW, Widome R, Resnick MD. Young people and violence. In: Heggenhougen K, Quah S, eds. International Encyclopedia of Public Health, Vol. 6. San Diego, CA: Academic Press, 2008:675–84. [2] Kulig J, Valentine J, Griffith J, et al. Predictive model of weapon carrying among urban high school students: Results and validation. J Adolesc Health 1998;22:312–9. [3] Durant RH, Getts AG, Cadenhead C, et al. The association between weapon carrying and the use of violence among adolescents living in and around public housing. J Adolesc Health 1995;17:376–80. [4] Resnick MD, Ireland M, Borowsky IW. Youth violence perpetration: What protects? What predicts? Findings from the National Longitudinal Study of adolescent health. J Adolesc Health 2004;35:424–33. [5] Brener N, Lowry R, Barrios L, et al. Violence-related behaviors among high school students—United States, 1991-2003. JAMA 2004;292: 1168–9. [6] Eaton DK, Kann L, Kinchen SA, et al. Youth risk behavior surveillance—United States, 2005. MMWR Surveill Summ 2006;55:1–108. [7] Grunbaum JA, Kann L, Kinchen SA, et al. Youth risk behavior surveillance—United States, 2001. MMWR Surveill Summ 2002;51:13–5. [8] Rudatsikira E, Singh P, Job J, et al. Variables associated with weaponcarrying among young adolescents in Southern California. J Adolesc Health 2007;40:470–3.

49

[9] Resnick MD, Bearman PS, Blum RW, et al. Protecting adolescents from harm. Findings from the National Longitudinal Study on adolescent health. JAMA 1997;278:823–32. [10] Kingery PM, Pruitt BE, Heuberger G. A profile of rural Texas adolescents who carry handguns to school. J Sch Health 1996;66:18–22. [11] Kann L, Kinchen SA, Williams BI, et al. Youth risk behavior surveillance—United States, 1999. MMWR CDC Surveill Summ 2000;49: 1–32. [12] DuRant RH, Kahn J, Beckford PH, et al. The association of weapon carrying and fighting on school property and other health risk and problem behaviors among high school students. Arch Pediatr Adolesc Med 1997;151:360–6. [13] Bailey SL, Flewelling RL, Rosenbaum DP. Characteristics of students who bring weapons to school. J Adolesc Health 1997;20:261–70. [14] Bergstein JM, Hemenway D, Kennedy B, et al. Guns in young hands: A survey of urban teenagers’ attitudes and behaviors related to handgun violence. J Trauma 1996;41:794–8. [15] Simon TR, Crosby AE, Dahlberg LL. Students who carry weapons to high school: Comparison with other weapon-carriers. J Adolesc Health 1999;24:340–8. [16] Leeb RT, Barker LE, Strine TW. The effect of childhood physical and sexual abuse on adolescent weapon carrying. J Adolesc Health 2007; 40:551–8. [17] United States Department of Health and Human Services. Healthy people 2010: With understanding and improving health and objectives for improving health. 2nd edition. Washington, DC: USDHHS, 2000. [18] Pettingell SL, Bearinger LH, Skay CL, et al. Protecting urban American Indian young people from suicide. Am J Health Behav 2008;32: 465–76. [19] United States Indian Health Service, Division of Program Statistics. Trends in Indian Health, 1998-1999. Rockville, MD: United States Department of Health and Human Services, Public Health Service, Indian Health Service, Office of Planning, Evaluation, and Legislation, Division of Program Statistics, 1999. [20] Blum RW, Harmon B, Harris L, et al. American Indian–Alaska Native youth health. JAMA 1992;267:1637–44. [21] Inouye DK. Our future in jeopardy: The mental health of Native American adolescents. J Health Care Poor Undeserved 1992;4:6–8. [22] HeavyRunner I, Morris JS. Traditional native culture and resilience. Research/Practice 1997;5:28–33. Available at, http://education.umn. edu/CAREI/Reports/Rpractice/Spring97/traditional.htm. [23] Blum RW, Resnick MD, Harris LJ. The Next Generation: The Health and Well Being of Young People of Color in the Twin Cities (monograph). Minneapolis, MN: University of Minnesota, National Adolescent Health Resource Center, 1990. [24] Bearinger LH, Pettingell SL, Resnick MD, et al. Violence perpetration among urban American Indian youth. Arch Pediatr Adolesc Med 2005; 159:270–7. [25] Werner EE. The children of Kauai: Resiliency and recovery in adolescence and adulthood. J Adolesc Health 1992;13:262–8. [26] Rutter M. Psychosocial resilience and protective mechanisms. Am J Orthopsychiatry 1987;57:316–31. [27] Potthoff SJ, Bearinger LH, Skay CL, et al. Dimensions of risk behaviors among American Indian youth. Arch Pediatr Adolesc Med 1998; 152:157–63. [28] Borowsky IW, Ireland M, Resnick MD. Adolescent suicide attempts: Risks and protectors. Pediatrics 2001;107:485–93. [29] Rubenstein JL, Heeren T, Housman D, et al. Suicidal behavior in "normal" adolescents: Risk and protective factors. Am J Orthopsychiatry 1989;59:59–71. [30] Orpinas P, Murray N, Kelder S. Parental influences on students’ aggressive behaviors and weapon carrying. Health Educ Behav 1999; 26:774–87. [31] Suris JC, Thurnherr J, Berchtold A, et al. To use it or not to use it: Differences between adolescent males carrying a weapon and those using it in a fight. J Adolesc Health 2008;42:S21.

50

L.H. Bearinger et al. / Journal of Adolescent Health 47 (2010) 43–50

[32] Leventhal H, Keeshan P. Promoting the healthy alternatives to substance abuse. In: Millstein SG, Petersen AC, Nightingale EO, eds. Promoting the Health of Adolescents: New Directions for the Twenty-First Century. New York, NY: Oxford University Press, 1993:260–84. [33] Borowsky IW, Resnick MD, Ireland M, et al. Suicide attempts among American Indian and Alaska Native youth: Risk and protective factors. Arch Pediatr Adolesc Med 1999;153:573–80. [34] Barkin SL, Smith KS, DuRant RH. Social skills and attitudes associated with substance use behaviors among young adolescents. J Adolesc Health 2002;30:448–54. [35] Duncan SC, Alpert A, Duncan TE, et al. Adolescent alcohol use development and young adult outcomes. Drug Alcohol Depend 1997;49:39–48.

[36] LaFromboise TD, Bigfoot DS. Cultural and cognitive considerations in the prevention of American Indian adolescent suicide. J Adolesc 1988; 11:139–53. [37] Mezzich AC, Giancola PR, Tarter RE. Violence, suicidality, and alcohol/drug use involvement in adolescent females with a psychoactive substance use disorder and controls. Alcohol Clin Exp Res 1997; 21:1300–7. [38] Carmines EG, Zeller RA. Reliability and Validity Assessment. Quantitative Applications in the Social Sciences (series no. 07-017). Beverly Hills, CA: Sage Publications, 1979. 70. [39] Resnick MD. Protective factors, resiliency and healthy youth development. Adolesc Med 2000;11:157–65. [40] Garbarino J. Lost boys: Why Our Sons Turn Violent and How We Can Save Them. New York, NY: Free Press, 1999. xiii, 274.