James, an 18-year-old boy, is in your office for a physical examination required for starting college in the fall. He was the valedictorian of his high school class, has a college scholarship, was president of his school's National Honor Society chapter, and was captain of the football team. You have known James since his family moved into the area 7 years ago, but you have not seen him for several years. He has no specific complaints today. Just before the end of the visit you ask, "Is there anything more you would like to discuss?" Taking a deep breath, he reveals that he was kicked out of his home after graduation 3 weeks ago and has been living in his car ever since. The reason is that he told his parents that he is gay. He admits to feeling depressed, and has had thoughts of suicide. However, he still has plans to go to college.
Toni, a 16-year-old girl, is also on your schedule this afternoon. This is her first visit. She has not been in school since she was 14 years old, and was failing most of her classes for as long as she could remember before that. She is the mother of two children, ages 8 months and 2½z years. Both are currently in the care of her maternal aunt. Toni was incarcerated when her boyfriend, the father of her children, was arrested for selling drugs, but she was released on parole. During your interview, you learn that Toni is working toward her GEO by taking night courses. She says she is working two minimum-wage, part-time jobs, and that is why her children spend much of their time at her aunt's home. She has not used drugs since breaking up with her last boyfriend more than a year ago, and she hopes to be a computer programmer.
Both of these adolescents demonstrate resilience. James is able to "stay on track" with his goal of starting college in the fall, despite being rejected by his parents and being depressed. Toni is working hard toward improving her life circumstances and those of her children through schooling and employment.
RESILIENCE, RISK, AND PROTECTION
Resilience is commonly defined as the capacity to bounce back after adversity.1 Certainly, both Toni and James are faced with different forms of adversity. Another characterization of the resilient individual is the desire to "make it," to "overcome the odds." However, success so defined is not synonymous with happiness. Clearly, neither Toni nor James are as happy as they otherwise would be without comparable life stresses.
Another way to think of resilience is in terms of protective factors. Protective factors are those individual, family, peer, and environmental influences that reduce negative health or social outcomes.2 A resiliency paradigm strives to identify protective, nurturing factors in the lives of adolescents that can strengthen their capacity to resist problem behaviors or bounce back from adversity.3
In contrast to protective factors, risk factors are those factors that limit the likelihood of successful development.4 For example, risk factors might include poverty, a history of mental illness in the family, or frequent moves during childhood. These risk factors are not to be confused with health risk behaviors that may lead to negative health outcomes, such as unprotected sexual intercourse.
A framework for intervention based on protective factors and risk factors assumes that resilience is not a trait that some have and others lack. Instead, resilience represents the interaction between each individual and his or her environment.4 The framework also assumes that stress is a subjective response to our experience that is universal. For example, one individual may experience an anticipated examination as stressful, whereas another does not. How one interprets an event is a mix of biology and history.
Much of the recent research on resiliency and protective factors is grounded in a youth development perspective.3 The youth development framework assumes that adolescents have fundamental needs for healthy development; these needs might be experiences or circumstances that permit the achievement of developmental tasks. More than 25 years ago, the U.S. Department of Health, Education, and Welfare commissioned a study detailing the fundamental elements of healthy adolescent development.5 Eight conditions set the precepts of healthy youth development. Konopka argued that young people need to:
1. Participate as citizens, household members, workers, and responsible members of society;
2. Gain decision-making experience;
3. Interact with peers and acquire a sense of belonging;
4. Reflect on self in relation to others, and discover self by looking outward as well as inward;
5. Discuss conflicting values and formulate one's own value system;
6. Experiment with one's own identity, with relationships with others, and with ideas to try out various roles without having to commit oneself irrevocably;
7. Develop a sense of accountability in the context of a relationship among equals; and
8. Cultivate a capacity to enjoy life.
Intervention programs that are effective at reducing the major morbidities of adolescence incorporate elements of healthy youth development. Programs that strive simply to provide information to youth have generally been ineffective.6 Similarly, programs designed to solely improve access to services have been inadequate. We believe that applying an understanding of those factors that lead to positive outcomes for adolescents is far more effective than focusing exclusively on the reduction of problem behavior (R. W. Blum, MD, MPH, PhD, unpublished data, March 2000).
THE NATIONAL LONGITUDINAL STUDY OF ADOLESCENT HEALTH (ADD HEALTH)
Add Health was the largest study of its type ever undertaken in the United States.2 The goal of this comprehensive, school-based study was to identify the family, school, and individual factors most strongly associated with diminished health risk behaviors. Two of three planned phases were conducted in 1995 and 1996. In the first phase, 90,118 students in grades 7 through 12 from 134 schools around the United States responded to a brief, in-school survey. School administrators completed a half-hour survey on school policies, provision of health services, and school environment and characteristics.
The second phase involved two 90-minute, inhome interviews (wave 1 and wave 2), administered approximately 1 year apart. The first wave was completed by more than 20,000 adolescents and the second wave by more than 15,000 adolescents. Adolescents listened to sensitive questions through earphones and responded by entering their answers into laptop computers. Data were collected on health risk behaviors, such as drug and alcohol use, sexual behavior, and interpersonal violence, as well as on health status, use of health care, family dynamics, peer networks, romantic relationships, decision making, aspirations, and attitudes.
In addition, a parent living in the home, usually a mother, was invited to complete a half-hour interview at the time of the first wave. With the use of linked identifiers, the in-school and inhome data sets of adolescents, the responses of school administrators, and the responses of parents were merged. Extensive precautions were taken to ensure confidentiality.
A number of Add Health analyses have been published.7 Several have illuminated protective factors in the lives of adolescents. These are divided into family, school, and individual factors.
Among family factors, parent-family connectedness was highly protective.2 Adolescents who perceived that at least one parent cared about and loved them were less likely than others to participate in every risk behavior studied. These adolescents reported that they could talk to a parent and that a parent was psychologically available to them. Adolescents who reported a high level of parent-family connectedness also reported less emotional distress (feeling depressed, lonely, sad, fearful, or moody, crying, or having decreased appetite), suicidality (suicidal ideation or suicide attempts), violence (had been in a physical fight, had been injured or had threatened another with a weapon, had used a weapon in a fight, or had shot or stabbed someone), substance use (use of cigarettes, alcohol, and marijuana), and sexual behaviors (young age at first sexual activity and pregnancy history).
Parental presence was a second family-level factor protective against a number of risk behaviors, especially those related to substance abuse.2 Adolescents who reported that at least one of their parents was home at certain times during the day (before or after school, at bedtime, or at dinner) were less likely to report smoking cigarettes or using alcohol or marijuana. This protective effect increased when parental presence was combined with the perceived sense that a parent cared about the adolescent.
A third family-level protective factor was an adolescent's perception that school achievement was important to his or her parents.2 Adolescents who believed that their parents were concerned about school performance not only did better in school, but also were less likely to experience emotional distress or violence or to smoke cigarettes.
Similar to parent-family connectedness, school connectedness was found to be highly protective against all risk behaviors.2 Adolescents who perceived that there was an adult in school .who cared or was concerned about them and felt that they were treated fairly (without discrimination) by peers were less likely to experience emotional distress, suicidality, violence, substance use, or risky sexual behaviors. This association appeared to be independent of school performance, and was protective even in the absence of parental connectedness.
An individual factor found to be protective was a sense of spirituality or religiosity; this was independent of the specific faith of the individual.2 Religion alone had no apparent protective effect. It appeared that the belief in something beyond oneself was associated with less risk behavior.
A second individual factor associated with less risk behavior was related to pubertal development. Those adolescents whose physical maturation was slower were less likely to participate in risk behaviors.2 Conversely, those with precocious pubertal development were at the highest risk.
Another factor associated with less risk behavior was doing well in school.2 Academic success was protective against every risk behavior studied, whereas poor school performance and having repeated a year of school were associated with increased health risk behavior.
Other analyses of Add Health data focused on the relationships among race and ethnicity, income, and family structure.8
Add Health data revealed that white adolescents were more likely to smoke cigarettes, use alcohol, and attempt suicide than were black or Hispanic adolescents. Black adolescents were more likely to have early sexual intercourse, and black and Hispanic adolescents were more likely to engage in violence. However, when gender, race and ethnicity, income, and family structure were controlled, only 10% or less of the variance in each of these adolescent risk behaviors could be explained. This suggests that race and ethnicity, income, and family structure are insufficient to understand risk behaviors at the individual level.
Rather, we must understand the individual, family, peer, and environmental factors that are associated with risk and that protect adolescents from harm if we are going to work with them effectively in our clinics.
Another Add Health analysis looked more closely at risk and protective factors for adolescent suicide.9 The presence of three protective factors reduced the risk of a suicide attempt by 70% to 85% for each gender and racial or ethnic group. These protective factors included family factors such as parent-family connectedness and parental presence, school and community factors such as school connectedness and the presence of a caring adult, and individual factors such as emotional well-being and grade point average. This marked reduction in suicide risk was seen for adolescents with and without identified risk factors.
A different analysis looked prospectively at the perpetration of violence approximately 1 year after examining risk and protective factors.10 Except for Hispanic boys, the presence of protective factors (perceived family caring, high parental expectations for school performance, higher grade point average, and perceived importance of religion and prayer) reduced the proportion of adolescents involved in violence by at least half.
THE APPLICATION OF A RESILIENCY FRAMEWORK TO CLINICAL PRACTICE
A resiliency framework allows us, as pediatricians, to move beyond a focus on the behaviors that predispose adolescents to morbidity and mortality to understanding the antecedents of risk behaviors. Even more important is understanding the factors - the strengths and capacities an adolescent has to draw on - that will allow the adolescent to deal with the adversities with which he or she is faced.
The application of such an approach requires that we understand the multiple contexts in the lives of adolescents - family, school, peers, and community. It requires that we identify those aspects within each domain that may be deficient and therefore may predispose to negative outcomes. For example, a risk factor in the family domain may be a family member's suicide attempt, and in the school domain, it may be poor academic achievement.
So far this sounds remarkably like a traditional "history," but approaching the adolescent within a resiliency framework requires going beyond identification of the problem to identification of the protective factors. Within the family domain, this may include identification of an adult with whom the adolescent feels close. What are the parents' expectations for behavior? For school completion? How available is the parent emotionally to the adolescent? In the school domain, it involves identification of individuals with whom the adolescent feels close, school activities in which he or she feels a part, and clubs and sports in which he or she participates. On the individual level, applying a resiliency framework requires understanding the adolescent's belief system, spirituality and religiosity, ability to use support systems, and level of developmental maturation.
Pediatricians can offer considerable anticipatory guidance to the parents of adolescents to reduce the risk of harm. First, parents should be reminded not to believe that they do not matter. There is extensive evidence that parents are centrally important in the lives of their children throughout the adolescent years.
Second, parents should be reminded that even if their adolescent does not talk to them, they should not stop talking to their adolescent. Based on the information we have, even when adolescents appear not to be listening, they are watching and listening intently to their parents.
Third, parents should be encouraged to be explicit about their expectations. We know that parents who have high expectations for their adolescent's performance in school have an adolescent who not only performs better in school, but also is involved with fewer risk behaviors than other adolescents. Additionally, we know that adolescents who have come to internalize their parents' values and beliefs about sexual mores tend to have sexual intercourse later than do other adolescents.
Fourth, pediatricians should discuss what it means to be connected. This does not mean just playing football or going shopping at the mall with their adolescent. Although parental connectedness is critically important, it is not reflected in shared activities. Rather, it is based on a parent's psychological and emotional availability to their adolescent. Parents should be reminded that this means being available to listen, even if they disagree with what their adolescent has to say, and then providing him or her with an adult "mirror," reflecting their beliefs and values.
Pediatricians need to be a bridge between adolescents and their parents, while being careful not to become entrapped by siding with one or the other. They may be able to facilitate open conversation and discussion between adolescents and parents in conflict when others may be unable or unwilling to do so.
Currently, there are no clinical questionnaires or guidelines that tap these dimensions as others11'12 have done for the identification of adolescent problems. As we increasingly understand the role of protective factors in developing effective interventions for adolescent patients, we will move in that direction. However, knowing what we do about what is associated with risk factors and protective factors in adolescence, we should not wait. Our patients need our help now.
1. Rutter M. Resilience: some conceptual considerations. J Adolesc Health. 1993; 14:626-631, 690-696,
2. Resnick MD, Bearman PS, Blum RW, et al. Protecting adolescents from harm: findings from the National Longitudinal Study on Adolescent Health. iAMA. 1997;278:823-832.
3. Resnick MD. Protective factors, resiliency, and healthy youth development. In: Strasburger VC, Greydanus DE, eds. At-Risk Adolescent: An Update for the New Century. Philadelphia: Hanley & Belfus; 2000:157-164.
4. Blum RW. Healthy youth development as a model for youth health promotion. J Adolesc Health. 1998;22:368-375.
5. Konopka G. Requirements for healthy development of adolescent youth. Adolescence. 1973;8:291-316.
6. Kirby D. No Easy Answers. Washington, DC: National Campaign to Prevent Teen Pregnancy; 1997:17-44.
7. Udry JR, Bauman KE, Bearman PS, et al. The National Longitudinal Study of Adolescent Health. Chapel Hill, NC: Carolina Population Center at the University of North Carolina at Chapel Hill; 1998. Available at: www.cpc.unc. edu/ projects / addhealth / home.hrml. Accessed August 29, 2000.
8. Blum RW, Beuhring T, Shew ML, Bearinger LH, Sieving RE, Resnick MD. The effects of race, income, and family on adolescent risk-taking. Am J Public Health. 2000; 90:1879-1884.
9. Borowsky IW, Ireland M, Resnick MD. Adolescent suicide attempts: risks and protectors. Pediatrics. In press.
10. Resnick MD, Borowsky IW, Ireland M. Adolescent violence perpetration: What predicts? What protects? J Adolesc Health. 1999;24:128.
11. Neinstein LS, ed. Adolescent Health Care, 3rd ed. Baltimore: Williams & Wilkins; 1996:50-52.
12. Elster AB, ed. Guidelines for Adolescent Preventive Services (GAPS): Recommendations and Rationale. Baltimore: Williams & Wilkins; 1994:1-182.