A Predictive and Moderating Model of Psychosocial Resilience in Adolescents

By Tusaie, Kathleen; Puskar, Kathryn; Sereika, Susan M

Purpose: To identify point prevalence of psychosocial resilience (PR) and to test moderating and predictive relationships among optimism, chronological age, gender, perceived family and friend support, number of bad life events, and PR in rural adolescents.

Design: A secondary analysis of a cross-sectional survey of 624 rural adolescents aged 14 to 18 in an eastern U.S. state.

Method: Descriptive statistics were used to identify point prevalence, and stepwise logistic regression was used to identify which variables alone or in combination had significant effects upon PR.

Findings: The point prevalence of PR was 17% with the largest percentage of students reporting medium levels of resilience. Level of adolescent PR was partially predicted by cognitive factors (optimism, perceived family support), number of bad life events, age, and gender. Perceived support of friends and optimism modified the level of PR.

Conclusions: The predictive and moderating model was useful for building knowledge about the process of PR in rural adolescents.

JOURNAL OF NURSING SCHOLARSHIP, 2007; 39:1, 54-60. 2007 SIGMA THETA TAU INTERNATIONAL.

[Key words: resilience, adolescents, predictive model]

Resilience is the capability to adapt better than expected in the face of significant adversity or risk. This capability changes over time, is developmentally specific, influenced by risk and protective factors in the person and the environment, and it also contributes to the maintenance and enhancement of health. Resilience can be described for a specific domain (psychosocial, physical, work or school) or as overall general resilience. This study was focused on psychosocial resilience (PR) of rural adolescents.

Background

A large and diverse body of literature indicates that all people have some level of resilience, and approximately onethird of any population have high levels of resilience (Hauser, 1999; Luthar, 1991; Masten & Coatsworth, 1998; Resnick, 2000; Rutter, 1993; Tusaie & Dyer, 2004; Werner & Smith, 1992). Resilience has been documented across a wide range of environmental and developmental stressors, diseases, and catastrophic events. Those findings add to the understanding of individual diversity in responses to the same stressor (Al-Naser & Sandman, 2000; Hess, 2002; Norris, 2002; Sigal & Weinfeld, 2001). Many protective factors have been identified as potential shapers of individual resilience: cognitive factors such as cognitive reframing, problem-solving abilities, optimism, a sense of meaning or a cohesive narrative about the stressor, high intelligence level, reading skills, resourcefulness in seeking social support; and environmental factors such as having fewer negative life events, history of competence or successes, positive attachments between parents and child, participation in school activities, and expansion of support system outside the family (Chang, 2001; Geanellos, 2005; Rutter, 1993; Scheir & Carver, 1987; Werner & Smith, 1992).

Despite considerable research on outcomes and protective factors, little is known about the complex interactions among resilience and contributing variables in adolescents. Models of resilience are needed to expand understanding from factor identification to resilient processes. For example, if an adolescent has a negative relationship with parents, can other relationships compensate? Which internal factors are more influential in enhancing resilience? This understanding can be important in designing interventions to enhance internal as well as environmental factors that contribute to higher levels of resilience.

Today’s adolescents are bombarded with internal and environmental stressors. The leading causes of death and illness among adolescents are potentially preventable. Injuries, homicide, and suicide, with alcohol or drugs frequently involved, are the primary reasons adolescents die in the US today (U.S. Department of Justice, 2000). Rural adolescents have been reported to be at an even higher risk for emotional distress than are urban or suburban adolescents (Hartley, Bird, & Dempsey, 1999; Resnick et al., 1997; U.S. Department of Health and Human Services, 2003). However, information is lacking about interactions among protective factors and resilience to direct interventions for adolescents, and even less for rural adolescents.

This study was designed as a step toward filling gaps in knowledge of the process of adolescent resilience. The design for this secondary analysis was a cross-sectional survey used in the first phase of the parent study. The purpose of this study was twofold: first, to identify point prevalence (adolescents having PR at one time) as well as levels and gender differences in psychosocial resilience in rural adolescents; and second, to derive a model depicting predictive and moderating relationships among optimism, chronological age, gender, perceived family and friend support, number of bad life events, and psychosocial resilience.

The theoretical framework for this study was based on concepts from Lerner’s model of developmental contextualism, including plasticity, embeddedness, intraindividual diversity, and biology (Lerner et al., 1996) and Lazarus’ theory of stress and coping; including coping, social resources, person, and genetics; Lazarus & Folkman, 1984. To be considered resilient, individuals must meet two criteria: highrisk status and adaptation better than expected for the population (Masten & Coatsworth, 1998). Rural adolescents are considered to be at risk for emotional distress. Adolescent suicide rates are nearly 80% greater in rural than urban counties (seekins, 2002). Furthermore, rural residents have fewer health resources, are underrepresented in prevention research, and have similar or higher rates of poverty and substance use than do urban residents (Eberhardt, Ingram, & Makuc, 2001). Adaptation better than expected for this adolescent population is defined as depressive symptoms below the population mean, substance use below the sample mean, and cognitive coping above the sample mean. The rationale for inclusion of the study variables is both theoretical and research based (Chang, 2001; Masten & Coatsworth, 1998; Rutter, 1993; Shapiro, 2002; Tusaie-Mumford, 2001; Werner & Smith, 1992).

Methods

Sample

A convenience sample of 624 students was selected from four rural schools in western Pennsylvania for the parent study. All school districts had populations under 25,000 and the average income for each family was approximately $24,000. Most of the families were supported by fathers who were unskilled or semiskilled workers such as laborers, mill workers, machine operators, and bus or truck drivers. Although most families had large property lots, only a small percentage (5%) were able to support their families by farming. The sample was 60% (n=376) female, predominantly Caucasian (97%, n=606; which is comparable to the population), and ranged in age from 14 to 19 years (mean=15.8, SD=0.99).

Design

This study was a secondary analysis of data from the first phase of a larger study by Puskar and colleagues, Intervention to Promote Mental Health in Rural Youth (NIH/NINR ROI NRO03616). The purpose of the parent study was to assess the effect of an intervention on multiple psychiatric symptoms and coping skills in rural adolescents. Results from that grant included articles about the intervention (Puskar, Sereika, & Tusaie, 2003), optimism in adolescents (Puskar, Sereika, Lamb, Tusaie-Mumford, & McGuinness, 1999), and substance abuse (Puskar, Sereika, Lamb, & Tusaie- Mumford, 2000).

The secondary analysis reported here was focused on psychosocial resilience in rural adolescents. Point prevalence, i.e., the number of people having a certain attribute at one time, levels of psychosocial resilience, and internal factors that might predict or moderate PR in adolescents were examined.

Instruments

Psychosocial resilience (PR) was a composite variable measured by the Reynolds Adolescent Depression Scale (RADS), Drug Use Screening Inventory (DUSI), and the four cognitive subscales of the Coping Response Inventory-Youth Form (CRI-Y).

RADS (Reynolds, 1986) is a 30-item self-report scale designed to measure cognitive, motor, somatic, and interpersonal symptoms associated with depression. Responses were ranked on a 4-point Likert scale with higher scores indicating pathology. Reliability estimates have been reported for over 2,000 students in 7th through 12th grade. In the literature, Cronbach’s alphas were consistently high and ranged from .91 to .94 with a total sample alpha of .92 (Reynolds, 1986). Three studies were conducted by Reynolds (1986) to examine test-retest reliability at 6 weeks, 3 months, and 1 year. The coefficients were .80, .79, and .63, respectively. Overall, the reliability of the RADS was supported in the literature. Internal consistency and validity scores were high. To date RADS has been administered to more than 10,000 adolescents. For this sample of rural adolescents, Cronbach’s alpha was .93. Scores lower than the mean for this sample population were included in the psychosocial resilience variable. These scores indicated adolescents experiencing less intense depressive symptoms than did the average adolescent in the sample.

DUSI (Tartar, 1990) is used to measure substance use and was the second m\easure of PR. It has 15 questions to measure frequency of lifetime use, and 15 commonly used substances are reported for times of use. Responses are five categories ranging from 0 to 20 times. A cutoff of 30 on problem density is considered normal, indicating fewer than six times used. In this rural sample, the internal consistency coefficient was .79. Problem density scores lower than 30 were included in the variable psychosocial resilience. These scores indicated adolescents using substances less than the average adolescent in this sample.

The final measure of the composite variable, PR, was the CRI-Y (Moos & Schaeffer, 1993). Part 1 of this instrument includes 10 questions to describe a stressor and Part 2 measures coping response with 48 items; the eight scales each contain six items. Approach and avoidance subscales are included, with behavioral and cognitive strategies under each. High Cronbach’s alpha coefficients have been reported for each subscale (Moos & Schaefer, 1993). Scores have moderate stability over time and validity has been well established (Ebata & Moos, 1994; Moos & Schaefer, 1993). The cognitive coping subscale used as a portion of the measurement of psychosocial resilience in this study was derived through confirmatory factor analysis of the 48 items (Tusaie-Mumford, 2001). Only the cognitive items were used because higher levels of cognitive coping by adolescents have been associated with higher levels of adaptive coping (Puskar et al., 2003, Werner & Smith, 1993). Cronbach’s alpha coefficient for the cognitive coping subscale was estimated at .86 for this rural adolescent sample. At this high level, correlations are attenuated very little by random measurement error. Therefore, this cognitive coping subscale had high reliability (Carmines & Zeller, 1979). Cognitive coping scores above the mean for this population were included in the variable psychosocial resilience.

Optimism was measured by the Life Orientation TestRevised (LOT- R; Scheir, Carver, & Bridges, 1994). It contains generalized outcome expectancies with 10 items, 4 of which are fillers. It is scored on a 5-point Likert scale. The literature indicates an acceptable level of internal consistency, with Cronbach’s alpha of .78. Temporal stability was demonstrated with a test-retest correlation of .51 over 10 weeks, and validity was established by comparison to other scales in the literature (Scheier et al., 1994). In this rural study sample, Cronbach’s alpha was .75, indicating adequate internal consistency.

Perceived social support was measured by the Perceived Social Support Scale (PSS; Procidano & Hellers; 1983). It is a 40-item, self-report list to measure perceived social support by family and friends with a separate scale for each. Procidano and associates (1990) reported internal consistency as .89 for family and .86 for friends. In this rural sample, internal consistency of the family scale was .88 and .84 for the friend scale. Factor analysis confirmed the one-factor validity of each scale.

Bad life events were measured by the Life Event Checklist (LEC; Johnson, 1993). A positive and a negative event score can be obtained. It is a self-report scale to measure life events for older children and adolescents. It has 46 items plus four spaces for adding events that were not listed. Testretest correlations over 6 weeks were .69 for positive score and .72 for negative scale. This scale has been widely used in child and adolescent research. In this study only negative life events scores were used. Cronbach’s alpha for this rural sample was .82, and the average number of bad life events experienced was six.

Data Analysis

Both descriptive statistics to report point prevalence and regression analysis to determine predictive and moderating capability of the variables were completed. Although PR was a continuous variable, cut-off points were assigned to allow comparison to point prevalence and levels of resilience reported in other populations. Consistent with other studies, psychosocial resilience was described as the absence of depression and substance abuse, with cognitive coping subscale scores above the mean of this sample (Luther, 1991; Masten & Coatsworth, 1998). Those students with Reynolds Adolescent Depression Scale scores < 77, Drug Use Screening Inventory density scores>30, Logical Analysis score>8.6, Cognitive Reframing score>8.3, Cognitive Avoidance scores>8.7, and Acceptance/Resignation scores>8.6 were considered psychosocially resilient. These cut-off points were determined in relation to the mean scores in this rural adolescent sample. The remaining students were classified as not psychosocially resilient. To determine levels of PR, data from the PR scores were divided into low, medium, and high from the total distribution. Scores below the median of one were low, scores at the median of one were medium, and scores above one were high.

Next, continuous data were used to investigate relationships among variables. Stepwise logistical regression was used to determine which variables alone or in combination were significant to psychosocial resilience. The variables included bad life events, optimism, perceived social support of family, perceived social support of friends, gender, and chronological age. A significance level of .05 was established for all statistical tests.

Procedures

The parent study procedure for data collection involved meeting with all 9th, 10th, and 11th grade students at four rural schools during assembly to discuss the project and provide consent and assent forms to the students. All students who returned the signed consents and assents were assigned to large groups in the cafeteria of each school and given 1 hours to independently complete the paper-and-pencil questionnaires. Fifty-two percent of the potential participants completed the questionnaires. An exempt institutional review board approval was obtained for the secondary analysis and the statistician from the parent study completed computer encryption of identifiers before the researcher obtained data.

Results

Psychosocial resilience was present in this sample of rural adolescents as proposed in Lerner’s concept of plasticity and Lazarus’ concept of adaptive coping. Point prevalence of psychosocial resilience in the total sample was 17% (n=104, 95% CI=102-106). Low levels of psychosocial resilience were reported by 30.6% (n=191) of the total sample, with the largest number of students, 45% (n=281), at medium levels, and the lowest number of students, 21.3% (n=133) at high levels of psychosocial resilience.

Biological influences proposed to influence the resilient process were indicated by the gender differences. Results showed a higher percentage of girls with low levels of psychosocial resilience (female=36%, n=133; male=24.6%, n=58) and a higher percentage of boys with medium (female=44.2%, n=163; male=50%, n=118) and high levels of resilience (female=19.5%, n=73; male=25.4%, n=24.4). Therefore, gender differences as well as level differences reached statistical significance (χ^sup 2^=30.712, df=1, p<.01; (χ^sup 2^=36.633, df=2, p<.01, respectively).

Stepwise logistic regression was used to identify which variables alone or in combination had significant effects on psychosocial resilience. Forward and backward entry of variables produced the same results. Only the significant bivariate correlations were entered into a stepwise linear regression analysis. Five variables directly predicted psychosocial resilience: optimism, bad life events, gender, age, and perceived support by family. Four interactions: optimism age; optimism bad life events; perceived social support of friends bad life events, perceived social support of friends age, were significant modifiers of adolescent psychosocial resilience.

The proposed influence of intraindividual differences and social embeddedness were indicated in the findings about optimism, perceived support, and life events. Optimism was the strongest direct positive influence, followed by perceived social support of family. Bad life events was the strongest direct negative influence, followed by age. Although the direct influence of age was consistent with the literature, the negative direction indicating that older adolescents had lower resilience was not expected. Perceived social support of friends interacted with bad life events and age to shift negative influences to positive but became significant only as age and number of bad life events increased. Thus, perceived support of friends became more important to older adolescents experiencing more negative life events. Optimism interacted with age to also shift negative influences to positive. When optimism interacted with bad life events, the negative influence upon psychosocial resilience was decreased. When adolescents experienced bad life events, perceived social support of friends was the most powerful moderator in this model. See Figure for the complete model and Table for the statistical predictors and moderators in the model.

Discussion

The importance of identifying point prevalence and levels of psychosocial resilience was to determine the existence of PR in rural adolescents. Recognition that at-risk rural adolescents can adapt better than expected contributes to a view of health beyond the absence of illness and disease to the idea that people can achieve high levels of wellness during adversity.

This predictive and moderating model showed that younger, male adolescents who experienced fewer bad life events, with higher levels of optimism and perceived family support, were more likely to have higher levels of PR. Older male adolescents with higher levels of optimism and higher levels of perceived support of family and friends also showed higher levels of PR even if they experienced multiple bad life events.

These findings are important in several ways. First, adolescent PR was str\ongly influenced by cognitive patterns. Optimism significantly decreased the effects of bad life events and shifted older adolescents to higher levels of PR in spite of negative age effects on levels of resilience. Adolescents who had more positive expectations for their future were less distressed when adverse events occurred and they pushed ahead more actively than those who expected worse outcomes (Carver & Scheir, 1998). This approach to living has the potential to co-create an external supportive environment to buffer adverse events and promote resilience. Often, bad life events are not preventable, but perceptions about the event and types of coping can be modified (Seligman, 1991; Tusaie & Patterson, 2006).

Second, although perceived support of family had direct effects on level of PR, no interactions with other variables were found in this model. Although friends are generally regarded as important to normal adolescent development, these findings indicated that perceived family support as well as optimistic thoughts were more powerful than were perceived friend support in the development of resilience. Although these findings were consistent with studies in other populations, (Belgrave, Chase-Vaughn, Gray, Addison, & Cherry, 2000: Werner & Smith, 1992), the lack of interaction of family support with other variables is different from previous studies, and it shows that perceived friend support could not be substituted for perceived family support. This explanation must be considered with caution because not all possible relationships among variables were explored. A more complete analysis would require structural equation modeling to identify mediating in addition to main and moderating effects.

Third, psychopathology co-occurred with psychosocial resilience. The presence of PR did not exclude the existence of psychiatric symptoms. But inverse relationships were found between resilience (PR) and depression, suicide attempts, and substance use. As psychosocial resilience increased, level of depression, number of suicide attempts, and substance use decreased. These relationships raised several questions. The bulk of research studies and funding has been directed to study human pathology. However, if interventions were done to promote resilience, would pathological symptoms such as suicide attempts, depression, and substance use decrease? Would universal (all adolescents) mental health promotion programs decrease the prevalence of psychiatric symptoms in adolescents? This positive perspective of the human condition has recently been organized under the banner of positive psychology, a term introduced by Seligman, who inaugurated the Positive Psychology Center at the University of Pennsylvania (Seligman & Csikszentmihalyi, 2000). Discussing a positive psychological approach does not negate or minimize the overwhelming social compromises or anguish that many adolescents face as they grow. However, understanding some of the interactions among natural characteristics and environments that promote growth of the resilience process has the potential to buttress the lives of less resilient people. Furthermore, beginning to understand and appreciate the grace, determination, and achievements of resilient individuals enhances the optimism of educators, clinicians, and researchers, as well as people experiencing social compromise, poverty, or abuse. Perhaps a focus on resilience and unexpected strengths would promote mobilization of the power of the human adaptation system in addition to reducing or eliminating pathological states.

Fourth, this study showed a direct effect of gender, the only biological factor in this model, on level of PR, but no interactions. Minimal information exists about the relationship of psychological and physical resilience, including the effects of genetics. Therefore, this area requires more attention to identify paths by which positive cognitions influence and are influenced by physiology.

Finally, the use of the construct of resilience might lead to more systematic thinking about concepts of positive psychology. This approach can decrease reductionistic thinking about the dynamic process of human adaptation. Shifting to a more positive approach to adolescents will require shifts in research, education of healthcare providers, current diagnostic procedures, and health promotion programs.

When scientific methods are applied to phenomena as complicated as human beings, the results come with many qualifiers and more questions. This study is only a beginning in the complex process of understanding adolescent resilience. Limitations of this work include the lack of longitudinal data and lack of generalizebility beyond this rural, primarily Caucasian populations.

Conclusions

The implications of research on resilience center on the ability to recognize positive adaptation in addition to pathology following adversity. Recognizing psychosocial resilience and identifying interactions among contributing factors has the potential to improve overall health through evidence-based programs designed to promote resilience. An increased focus on resilience also may assist health care professionals maintain optimism when working with severely traumatized people. Future research directions include analyzing data with structural equation modeling, completing longitudinal studies to establish causation, and exploring genetic contributions to adolescent resilience.

References

Al-Naser, F., & Sandman, M. (2000). Evaluating resiliency patterns using the ER89: A case study from Kuwait. Social Behavior and Personality, 28(5), 505-514.

Beck, A. (1976). Cognitive therapy and the emotional disorders. New York: New American Library.

Belgrave, E, Chase-Vaughn, G., Gray, E, Addison, J., & Cherry, V. (2000). The effectiveness of a culture- and gender-specific intervention for increasing resiliency among African-American preadolescent girls. Journal of Black Psychology, 26(2), 133-147.

Carmines, E., & Zeller, R. (1979). Reliability and validity assessment. Newbury Park, CA: Sage.

Carr, A. (Ed.). (2000). Cognitive therapy and the emotional disorders. New York: New American Library.

Carver, S.C., & Scheir, M. (1998). On the self-regulation of behavior. New York: Cambridge University Press.

Chang, E. (2001). Optimism & pessimism: Implications for theory, research, and practice. Washington, DC: American Psychological Association.

Ebata, A., & Moos, R. (1994). Personal, situational, and contextual correlates of coping in adolescents. Journal of Research on Adolescence, 4(1), 99-125.

Eberhardt, M., Ingram, D., & Makuc, D. (2001). Urban and rural health chartbook: Health, United States, 2001. Hyattsville, MD: National Center for Health Statistics.

Geanellos, R. (2005). Adversity as opportunity: Living with schizophrenia and developing a resilient self. International Journal of Mental Health Nursing, 14(1), 7-16.

Hartley, D., Bird, D., & Dempsey, P. (1999). Rural mental health and substance abuse. In T.C. Ricketts (Ed.), Rural health in the United States (pp. 159-178). New York: Oxford University Press.

Hauser, S.T. (1999). Understanding resilient outcomes: Adolescent lives across time and generations. Journal of Research on Adolescence, 9(1), 1-24.

Hess, C.R. (2002). Resilience among African American adolescent mothers: Predictors of positive parenting in early infancy. Journal of Pediatric Psychology, 27, 619-629.

Johnson, J.H. (1993). Lift events as Stressors in childhood and adolescence (Vol. 8). Newbury Park, CA: Sage.

Lazarus, R., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer.

Lerner, R., Lerner, J., von Eye, A., Ostrom, C., Nitz, K., Talwar- Soni, R., et al. (1996). Continuity and discontinuity across the transition of adolescence: A developmental contextual perspective. In J. Graber, J. BrooksGunn, & A. Peterson (Eds.), Transirions through adolescence (pp. 3-19). Mahwah, NJ: Lawrence Erlbaum.

Luthar, S. (1991). Vulnerability and resilience: A Study of high risk adolescents. Child Development, 62, 600-616.

Masten, A., & Coatsworth, J.D. (1998). The development of competence in favorable and unfavorable environments: Lessons from research on successful children. American Psychologist, 53(2), 205- 220.

Moos, R., & Schaefer, J. (1993). Coping resources and processes. In L. Goldberger & S. Breznitz (Eds.), Handbook of stress (pp. 234- 257). New York: The Free Press.

Norris, EH. (2002). 60,000 disaster victims speak: An empirical review 1981-2001. Psychiatry, Fall(65), 207-239.

Procidano, M., Sc Hellers, K. (1983). Perceived social support. American Journal of Community Psychology, 11, 1-24.

Puskar, K., Sereika, S., Lamb, J., & Tusaie-Mumford, K. (2000). Substance use among high school students in rural Pennsylvania. Journal of Addiction Nursing, 12(2), 55-63.

Puskar, K., Sereika, S., & Tusaie, K. (2003). Effect of the Teaching Kids to Cope Program on outcomes of depression and coping among rural adolescents. Journal of Child and Adolescent Psychiatric Nursing, 16(2), 71-80.

Puskar, K.R., Sereika, S., Lamb, J., Tusaie-Mumford, K., Sc McGuinness, T. (1999). Optimism and its relationship to depression, coping, anger, and life events in rural adolescents. Issues in Mental Health Nursing, 20, 115-130.

Resnick, M.D. (2000). Resilience and protective factors in the lives of adolescents. Journal of Adolescent Health, 27(1), 1-2.

Resnick, M., Bearman, P., Blum, R., Bauman, K., Harris, K., Jones, J., et al. (1997). Protecting adolescents from harm: Findings from the National Longitudinal Study on Adolescent Health. Journal of the American Medical Association, 278(10), 823-832.

Reynolds, W. (1986). A model for the screening and identification of depressed children and adolescents in school settings. Professional School Psychology, 1(2), 117-129.

Rutter, M. (1993). Resilience: Some conceptual considerations. Journal of Adolescent Health, 14, 626-631.

Scheir, M., & Carver, C.(1987). Dispositional optimism and physical well-being: The influence of generalized outcome expectancies on health. Journal of Personality, 55(2), 169-210.

Scheir, M., Carver, C., & Bridges, M. (1994). Distinguishing optimism from neuroticism (and trait anxiety, self-mastery, and self- esteem): A reevaluation of the Life Orientation Test. Journal of Personality and Social Psychology, 67(6), 1063-1078.

Seekins, T. (2002). Rural facts: Rural mental health. Retrieved May 1, 2006, from http://rtc.ruralinstitute.umt.edu/ MentalHealth.htm

Seligman, M. (1991). Learned optimism. New York: Random House.

Seligman, M., & Csikszentmihalyi, M. (2000). Positive psychology. American Psychologist, 55(1), 5-14.

Shapiro, E. (2002). Chronic illness as a family process: A social- developmental approach to promoting resilience. Journal of Clinical Psychology, 17(11), 832-838.

Sigal, J.J., & Weinfeld, M. (2001). Do children cope better than adults with potentially traumatic stress? A 40-year follow-up of Holocaust survivors. Psychiatry, Spring(64), 69-80.

Tartar, R. (1990). Evaluation and treatment of adolescent substance abuse: A decision tree method. American Journal of Drug and Alcohol Abuse, 16(1-2), 1-46.

Tusaie, K., & Dyer, J. (2004. Jan/Feb). A historical review of resilience. Holistic Nursing Practice, 3-8.

Tusaie-Mumford, K. (2001). Psychosocial resilience in rural adolescents: Optimism, perceived social support, and gender differences. [Unpublished dissertation]. Pittsburgh, PA: University of Pittsburgh.

Tusaie, K., & Patterson, K. (2006). A comparison of trait, situational, and comparative optimism: Clarifying concepts for a theoretically consistent evidence-based intervention to maximize resilience. Archives of Psychiatrie Nursing, 20(3), 144-150.

U.S. Department of Health and Human Services. (2003). Youth risk behavior surveillance-United States. Retrieved May 23, 2006, from http://www.ced.gov/mmwr/preview/mmwrhtml/ss5302al.htm

U.S. Department of Justice, F.B.I. (2000). Crime in the United States. Washington, DC: Author.

Werner, E., & Smith, R. (1992). Overcoming the odds: High risk children from birth to adulthood. Ithaca, NY: Cornell University Press.

Kathleen Tusaie, PhD, APRN, BC, Delta Omega, Assistant Professor, The University of Akron, College of Nursing, Akron, OH; Kathryn Puskar, DrPH, FAAN, Professor; Susan M. Sereika, PhD, Associate Professor; both at University of Pittsburgh, School of Nursing, Pittsburgh, PA. Parent study funded by NIH NINR Grant #R01 NR 03616- 01, K. Puskar Pl. Correspondence to Dr. Tusaie, The University of Akron, College of Nursing, 209 Carroll Street, Akron, Ohio 44325- 3701. E-mail: ktusaie@uakron.edu

Accepted for publication September 13, 2006.

Copyright Blackwell Publishing First Quarter 2007

(c) 2007 Journal of Nursing Scholarship. Provided by ProQuest Information and Learning. All rights Reserved.