Wilderness Adventure Therapy for At-Risk Youth.

Young people in America face a multitude of risks every day. Typical developmental challenges, which every youngster must face, include identity formation, physiological changes, and school transitions. And while statistics reveal that violent crime, drug use, and a host of other adolescent problem behaviors are on the rise, academic performance, community health, home stability, and employment opportunities are plummeting.

Numerous, too, are the interventions designed to prevent, rehabilitate, and punish our youth. Traditional mechanisms include the juvenile court system, reform schools, prosecuting adolescents under adult laws, job-training programs, psychotherapy, and a plethora of social services initiatives. We have also witnessed a seemingly endless procession of new, or newly packaged, interventions designed to address the problems created and faced by at-risk youth. These efforts include creative approaches such as midnight basketball, educational/ informational campaigns, military-style boot camps, and after-school recreation programs. This article will focus on wilderness adventure therapy, or WAT, for at-risk youth, a nontraditional intervention that has been gaining popularity.

What is WAT?

Accounts of tent therapy and therapeutic camping primarily for adult psychiatric in-patients can be traced back to the beginning of this century. In the 1930s, summer camps for children began to address the psychological needs of participants. The establishment of the Dallas Salesmanship Club Boys' Camp in 1948 marked the beginning of an intense period of growth in wilderness-based interventions for at-risk youth.

Other significant events in the evolution of wilderness adventure therapy for at-risk youth include the introduction of Outward Bound to the United States in 1962 and the creation of the Eckard Foundation camps in Florida a decade later. Today there are hundreds of WAT programs serving a variety of populations with special needs. However, most of these programs are designed to serve at-risk youth.

Virtually all adolescents could be considered at-risk in one fashion or another. At-risk youth includes adolescents who are judged delinquent by the courts or commit status offenses, adolescents who abuse substances, youth with emotional or behavioral disorders, academic underachievers, adolescents who are economically or socially disadvantaged, and youth who are deemed incorrigible by school officials, parents, or social service agencies.

Wilderness adventure therapy programs are very diverse. Experiences may last anywhere from one week to one year, and range from high-adventure (peak ascents, foraging for food, survival training, Or rock climbing) to low-adventure (flat-water canoeing, shelter building, wilderness camping, or day hikes).

Equally diverse are the therapeutic elements the staff use to guide the intervention. A national survey of WAT program directors, conducted by Davis-Berman and Berman (1994), found that many of the respondents struggled to specify the therapeutic model guiding their interventions. When therapeutic elements were discussed, "small group process" and "using the wilderness experience as a metaphor for problem solving" were mentioned most frequently. Natural consequences, modeling, existential theory, reality therapy, and cognitive behavioral techniques were also common.

Staff credentials are also varied. Some programs require their staff to posses little more than prior wilderness experience and good "people skills," while others employ only college graduates. Relatively few staffers possess formal training in mental health counseling. Such staff will typically end up serving in a supervisory capacity.

Despite the aforementioned diversity in WAT, several elements are common to all programs: Intervention occurs in a natural setting, uses a group format, emphasizes the active involvement of the participants, introduces elements of perceived risk, and maintains a therapeutic focus. The following definition of WAT, synthesized from the relevant literature base, succinctly summarizes these essential elements.

Wilderness adventure therapy is a systematic experiential group intervention that occurs in a natural setting and employs therapeutic techniques and processes within the context of activities and experiences that contain elements of real or perceived risk (i.e., physical, social, and emotional) to facilitate improvements in the psychological and behavioral functioning of the participant.

Early Research

Since the late 1950s, numerous research studies of WAT have appeared in the form of journal articles, master's theses, and doctoral dissertations. This research explosion has corresponded with the growth in popularity of the programs themselves. Unfortunately, almost all of these early studies were marred by serious methodological shortcomings such as nonexistent or inappropriate control/comparison groups, small sample sizes, one-shot data collection, unproven outcome measures, absence of a theoretical context, reliance on anecdotal evidence, exaggerated claims, and weak statistical techniques (Gillis & Thomsen, 1996).

It is beyond the scope of this column to thoroughly review the entire body of WAT research, and interested readers are encouraged to visit the comprehensive literature reviews of Bandoroff (1989), Ewert (1983, 1987), Gibson (1979), Gillis (1992), Kelley (1993), Levitt (1982), Shore (1977), and Weston (1993). Gass' comprehensive Adventure Therapy: Therapeutic Applications of Adventure Programming in Mental Health Settings (1993) also provides considerable information that should prove relevant to students, scholars, and practitioners. Cole, Erdman, and Rothblum edited a special issue of Women & Therapy in 1994 that dealt with wilderness therapy for women.

WAT's Hot Today?

Davis-Berman and Berman (1994) surveyed the directors of 31 therapeutic wilderness programs that provided mental health services to a variety of populations, the majority of which were at-risk adolescents. The programs were identified through their membership in the Association for Experiential Education. Behavioral problems, school and family problems, conduct disorders, self-esteem problems, depression, and suicidal ideation were among the common issues that gained individuals admission into a program.

The authors were interested in examining client type, staff qualifications, treatment modality, and activity type. Forty-eight percent of the programs consisted of private, for-profit operations serving at-risk adolescents. The majority of the programs reported the use of intensive wilderness experiences as the primary activity, while 10 relied mainly on ropes courses. Length of stay ranged from one day to 18 months, and cost varied from 850 to 8400 per day.

Respondents were not very specific when asked to identify their primary therapeutic modality, but the majority mentioned the therapeutic qualities of the group process and the use of the wilderness experience as a metaphor for real-world problem solving. Finally, 11 programs reported that front-line program staff were not required to hold a bachelor's degree. While 58 percent did employ master's-level staff, many of these worked in supervisory positions.

The results of this study were somewhat alarming. Many of the programs surveyed were using minimally qualified staff to work with one of the most challenging and vulnerable populations on very difficult issues, with little theoretical guidance. It is impossible to generalize WAT programs, given that the 31 surveyed were drawn from the AEE membership roles. But their membership in this professional association signifies that they are professional in nature. It also raises the possibility that they may enjoy a higher level of professionalism in staffing and design than most WAT programs.

Pommier and Witt (1995) studied the effects of an Outward Bound course that had a family-training component. Adolescents who had committed status offenses --acts such as alcohol use, truancy, curfew violations, and school misconduct--were referred to the experimental group by a state social services agency. The Self-Perception Profile for Adolescents, Family Adaptability and Cohesion Evaluation Scales-II, and Eyberg Child Behavior Inventory were used to collect data from 39 experimental subjects before the program started, then again at four weeks and four months.

Self-perception and family adaptability and cohesion data were collected from a non-equivalent control group composed of 40 local students who participated in no special programming. The treatment group showed significantly greater gains in self-perception and family adaptability at four weeks, but only global self-perception and five of the eight self-perception sub-scales remained significantly different at four months. Furthermore, the treatment group showed significant declines in self-perception, family adaptability, and behavior scores from four weeks to four months.

Thus, it appears that participation in the Outward Bound program was associated with some positive outcomes, though these were not uniformly maintained. It should be noted that the design of the study did not allow for a direct examination of the family training component. Furthermore, several methodological shortcomings (nonrandom convenience sample, limited demographic data, modest sample size, and use of a non-equivalent control group) limited the validity of the findings.

Bandoroff and Scherer (1994) also conducted a study of wilderness family therapy that involved a 26-day wilderness experience followed by a three-day family training component, which included themes such as emotional repair, trust-building, negotiation, and communication. Twenty-seven families constituted the experimental group; the 39 families who comprised the comparison group participated in the 26-day wilderness experience only.

The adolescents in the program presented a variety of situations including substance abuse, behavior problems, poor academic performance, and delinquent activity. Many had prior arrest, detention, and treatment histories. The Family Assessment Measure III, Self-Reported Delinquency Checklist, Revised Child Behavior Checklist, and Self Description Questionnaire III were used to collect data immediately before and following the experience, and once again six weeks after completion.

Caution for Encouraging Results

Following the intervention, the authors reported improvements among both groups on all indices. Further, for those participating in the family component, family functioning improved from the clinically impaired range at the time of the pretest to the normal range at the time of the follow-up. While encouraging, these results must be interpreted cautiously. Only means scores and standard deviations were reported, and data regarding norms and scoring procedures for the various measures were incomplete, hindering full interpretation of the outcomes and precluding any inferences beyond this sample. The small sample and lack of random assignment to the two conditions further compromise external validity.

Davis-Berman and Berman (1994) reported a longitudinal study of 23 adolescent therapy outpatients who participated in a two-week backpacking trip combined with an intensive therapeutic component. Using a mail survey, method data were collected with the Behavioral Symptoms Inventory, the General Self-Efficacy Inventory, and the Internal-External Locus of Control Scale immediately following the trip, and at four months, 12 months, and 24 months.

Only descriptive statistics were presented because of the low response rate (65 percent) at the final data-collection period. The results indicated that self-efficacy, behavior, and locus of control scores improved immediately following the trip, then regressed toward their pretest levels when measured at four months. At 12 months, scores again approached their initial posttest levels and remained stable at 24 months.

The attempts by the authors to provide a strong therapeutic component and long-term follow-up data collection are commendable. However, the small sample (approximately 15), the absence of a control group, the limited power of the statistical analysis, and the potential for the participants to become sensitized to the outcome measures warranted a cautious interpretation of the results. For example, the rebound in scores seen at 12 months could be the result of sample maturation rather than the WAT intervention (Campbell & Stanley, 1963).

Participants Improve

Cason and Gillis (1994) conducted a meta-analysis of 43 studies of outdoor adventure programming for adolescents (predominantly adjudicated youth) and reported a .31 overall average effect size. The authors interpreted this effect to mean that the average adolescent adventure program participant improved 12.2 percent and that he or she was better off than 62 percent of nonparticipants.

Larger effect sizes (i.e., greater positive outcomes) were found when clinical scales such as the MMPI were used as outcome measures as opposed to locus of control or serf-esteem scales. Longer adventure programs and younger participants were also associated with larger effect sizes. Finally, Cason and Gillis reported that "normal" adolescents were just as successful (i.e., equivalent effect sizes) as "labeled" adjudicated or at-risk adolescents.

Cason and Gillis' research did not provide convincing support for the efficacy of WAT, nor did it allow for meaningful insight into the salient features of these programs. The authors claimed that the results of adventure programming research were generally positive, but their meta-analysis provided only limited support for this assertion. The authors acknowledged that the average effect size of .31 was small to moderate, but they apparently failed to understand that their large standard deviation (.62) meant that the observed effect size did not differ significantly from zero.

Furthermore, the authors found that the more stringent the research methodology employed, the smaller the observed effect size. Rigorous research designs generally allowed for greater confidence in the validity of the results. Thus, it's possible that the well-designed studies in the Cason and Gillis meta-analysis, those most likely to produce valid results, did not uncover larger positive effects because they did not exist. Regardless, it seems likely that the current body of WAT research is insufficient for conducting a meaningful meta-analysis, a point made by one of the authors in a subsequent publication (Gillis & Thomsen, 1996).

Summary

The literature offers little of substance to further our understanding of WAT for at-risk youth. The results reported by Pommier and Witt (1995) follow a familiar pattern in WAT research: Some positive outcomes are identified immediately following participation in wilderness adventure activities, particularly in the area of self-perception, followed by a regression toward pretest scores when long-term follow-ups are conducted.

Minor and Elrod (1994) failed to find any significant differences between or within groups when adolescent self-concept and locus of control were examined. In their study of family functioning and delinquent behavior, Bandoroff and Scherer (1994) reported significant improvement following participation in WAT, but their limited data make clear interpretations difficult and preclude generalizations beyond their sample.

The longitudinal results of the impact of WAT on adolescent locus of control, self-efficacy, and behavioral symptoms, reported by Davis-Berman and Berman (1994), may stem from maturation or some other uncontrolled factor. The average effect size of .31 found by Cason and Gillis (1994) was not significantly greater than zero, and the number of poorly designed studies included in their analysis contributed to a lack of confidence in their findings. Other results appeared to be consistent with those of Cason and Gillis, but Gillis and Thomsen (1996) argued that there was not sufficient emphasis on research to make a meta-analysis in adventure therapy meaningful.

While some of those studying WAT for at-risk youth have responded to the frequent calls for more rigorous research, the work is still plagued by serious methodological limitations, which include the use of small convenience samples, the failure to isolate independent variables, the use of simplistic methodologies, and the absence of appropriate comparison or control groups.

Many of these limitations are directly tied to the difficulties of conducting field research with persons who have special needs, but they can be overcome with thoughtful effort. More disturbing are the limitations that appear to be the result of investigator negligence. These include overinterpreting or misinterpreting results (e.g., emphasizing limited positive outcomes and ignoring negative outcomes), partial reporting of data, making sweeping but unsubstantiated claims about the efficacy of WAT for at-risk youth, and failing to state clearly the limitations of the methodology used and the resulting data.

Recommendations

The most immediate need is for punctilious reporting of WAT research on at-risk youth, a more critical evaluation of what is said, how it is said, and how various groups of readers -- academics, practitioners, and consumers -- will interpret and use research. A complete and clear explication of our methods, variables, analyses, limitations, and findings will allow investigators to build upon previous works systematically, practitioners to use the research findings to improve their interventions, and consumers to make informed choices.

The present review makes clear the need for more rigorous WAT research, as well as greater care in the selection, operationalization, and measurement of variables. Random selection of our research participants and assignment to appropriate control/comparison groups are essential to all inferences beyond the immediate participants. The development of systematic programs of research would help ensure that meaningful questions are addressed in an informative manner.

Despite the limited empirical support, it appears that wilderness activities can have a positive effect on at-risk youth. However, a careful review of the literature reveals how little we know about why or how WAT works, for whom it works, under what conditions it works, and for how long it works. Some experts believe that the natural setting is the most important therapeutic element, while others argue that it is the activities themselves and how they are experienced.

Some proponents of the natural setting suggest that it is the unambiguous nature of the wilderness environment that is therapeutic, while others claim it is nature's inherent ambiguity that is conducive to change. Those who believe adventure activities are the essential therapeutic feature lack consensus regarding which are the most efficacious for a given population. Some argue for quiet, reflective experiences, while others recommend stressful, challenging activities.

Regardless, it appears that agreement over what makes WAT "therapeutic" for at-risk youth will not be forthcoming until a substantial body of rigorous, systematic, empirical research has been accumulated. A careful examination of the process of WAT for at-risk youth will allow us to begin providing answers to how, why, for whom, and under what conditions.

Finally, investigators must continue to expand at-risk youth WAT research to incorporate other areas of study. Serious questions are being raised about the usefulness of boosting self-esteem to promote desirable behaviors (Bandura, 1997; Baumeister, Smart & Boden, 1996). It seems timely for WAT researchers to reevaluate the appropriateness of this construct, a perennial favorite in WAT studies.

Ulrich's application of Wilson's biophilia hypothesis (Kellert & Wilson, 1993) might provide a useful framework for examining the role of setting in WAT. Ulrich has argued that humans have an innate affiliation for certain types of natural environments and that a variety of benefits may be derived from them (stress recovery, facilitation of learning, and post-surgical recovery).

Zuckerman's theory of sensation-seeking (Zuckerman, 1979), and the theory of delinquency Farley built from it (Farley & Sewell, 1976), might provide direction for investigations into the relation between personality type and activity focus in the WAT process.

According to Farley's theory, adolescents who have a high need for arousal yet lack socially acceptable opportunities to meet this need will engage in high-risk behaviors to satisfy their sensation-seeking needs. Theories of counseling (Prochaska, 1984; Yalom, 1985) might help guide investigations into the role staff play in facilitating the therapeutic process

Those conducting WAT research with at-risk youth should raise their investigative standards to the level of their dedication and enthusiasm. Only then will wilderness adventure therapy begin to move from an intuitively appealing alternative intervention to a validated mainstream therapy.

References

Bandoroff, S. 1989. "WAT for delinquent and pre-delinquent youth: A review of the literature." Unpublished manuscript. University of South Carolina at Columbia.

Bandoroff, S., and D.G. Scherer. 1994. "Wilderness family therapy: An innovative treatment approach for problem youth." Journal of Child and Family Studies 3: 175-91.

Bandura, A. 1997. Self-efficacy: The exercise of control. New York: Freeman Press.

Baumeister, R. F., L. Smart, L., and L.M. Boden. 1996. "Relation of threatened egotism to violence and aggression: The dark side of high self-esteem." Psychological Review 103: 5-33.

Campbell, D. T., and J.C. Stanley. 1963. Experimental and quasi-experimental design for research. Boston: Houghton Mifflin.

Cason, D., and H.L. Gillis. 1994. "A meta-analysis of outdoor adventure programming for adolescents." The Journal of Experiential Education 17: 40-7.

Cole, E., E. Erdman, and E.D. Rothblum. 1994. "Wilderness therapy for women: The power of adventure." Women & Therapy 15.

Davis-Berman, J., and D.S. Berman. 1994a. "Research update: Two-year follow-up report for the wilderness therapy program." The Journal of Experiential Education 17: 48-50.

Davis-Berman, J., and D.S. Berman. 1994b. "Therapeutic wilderness programs: A national survey." The Journal of Experiential Education 17: 49-53.

Ewert, A. 1983. Outdoor adventure and self-concept: A research analysis. Eugene, Ore.: University of Oregon Center for Leisure Studies.

--. 1987. "Research in outdoor adventure: Overview and analysis." Bradford Papers Annual 2: 15-28.

Farley, F.H., and T. Sewell. 1976. "Test of an arousal theory of delinquency: Stimulation-seeking in delinquent and non-delinquent black adolescents." Criminal Justice and Behavior 3: 315-20.

Gass, M.A. 1993. Adventure therapy: Therapeutic applications of adventure programming in mental health settings. Dubuque, Iowa: Kendal Hunt.

Gibson, P.M. 1979. "Therapeutic aspects of wilderness programs: A comprehensive literature review." Therapeutic Recreation Journal 13: 21-33.

Gillis, H.L. 1992. "Therapeutic uses of adventure-challenge-outdoor-wilderness: Theory and research." Coalition for education in the outdoors, research symposium proceedings. 35-47.

Gillis, H.L., and D. Thomsen. 1996. "A research update of adventure therapy (1992-1995): Challenge activities and ropes courses, wilderness expeditions, and residential camping programs." Coalition for education in the outdoors, research symposium proceedings. 77-90.

Kelley, M.P. 1993. "The therapeutic potential of outdoor adventure: A review with a focus on adults with mental illness." Therapeutic Recreation Journal (2nd quarter): 110-25.

Levitt, L. 1982. "How effective is wilderness therapy: A critical review." In F.E. Bolteler (Ed.) Proceedings: Wilderness Psychology Group Third Annual Conference. 81-9.

Minor, K.L., and P. Elrod. 1994. "The effects of a probation intervention on juvenile offenders' self-concepts, loci of control, and perceptions of juvenile justice." Youth and Society 25: 490-511.

Pommier, J.H., and P.A. Witt. 1995. "Evaluation of an Outward Bound school plus family training program for juvenile status offenders." Therapeutic Recreation Journal (2nd quarter): 86-103.

Prochaska, J.O. 1984. Systems of psychotherapy: A transtheoretical analysis. Pacific Grove, Calif.: Brooks/Cole.

Shore, A. 1977. Outward Bound: A reference volume. Greenwich, Conn.: Outward Bound Inc.

Ulrich, R.S. 1993. "Biophilia, biophobia, and natural landscapes." In S.R. Kellert, and E.O. Wilson (Eds.) The biophilia hypothesis. Washington, D.C.: Island Press.

Weston, R. 1993. "The relationship of selected input variables to success in a therapeutic wilderness program." Unpublished master's thesis. Clemson University, Clemson, S.C.

Yalom, I.D. 1985. The theory and practice of group psychotherapy. New York: Basic Books.

Zuckerman, M. 1979. Sensation seeking: Beyond the optimal level of arousaL Hillsdale, N.J.: Lawrence Erlbaum.

RELATED ARTICLE: Research into Action: Adventure as Therapy

Research Into Action is published monthly by the Society of Park and Recreation Educators, National Recreation and Park Association. As an accompaniment to "Research Update," its goal is to turn research findings into field action by highlighting management strategies. Founding editors are Dr. Ruth Russell and Dr. Daniel D. McLean, Department of Recreation and Park Administration, Indiana University.

Introduction

Risks facing young people today are numerous, as are the interventions designed to prevent, rehabilitate, and punish them. Wilderness adventure therapy for at-risk youth is a nontraditional intervention that has experienced a steady growth in popularity. The research into WAT continues and holds the potential for future enhancement of understanding.

Impact of this Research

The impact of the research is very limited for several reasons. First, while positive outcomes may frequently be reported, particularly in the areas of self-perception, long-term studies do not indicate a lasting effect, thus eroding the potential positive effect of WAT. Second, uncontrolled variables may affect results of research that are not understood, not measured, or not measurable. Meta-analysis of WAT research suggests there is not a sufficient body of research to make such an analysis. Finally, such research is plagued by serious methodological limitations.

Research into WAT outcomes still holds an unsubstantiated promise for those who are committed to it. Ongoing research will provide important answers for WAT practitioners. Issues of self-perception, family communication and cohesiveness, self-control, and expansion of adolescent impacted theory (sensation seeking and delinquency) continue to receive attention from researchers.

How to Use this Research

The existing research holds promise but few answers for practitioners. It is suggested:

* Practitioners work jointly with researchers to develop and implement well-designed studies focusing on expanding understanding and theory.

* While claims of the positive effect of WAT need to be tempered in light of existing research, we must acknowledge that WAT does have some positive effect on certain at-risk youth.

* A more rigorous and coordinated research program be undertaken.