Aggressive adolescents benefit from massage therapy.

Diego, Miguel A. ; Field, Tiffany ; Hernandez-Reif, Maria 等

Aggressive behavior among adolescents has become a major public health problem (Human Capital Incentive, 1997; Centers for Disease Control, 2001). Despite the recent reduction in youth violence and homicide rates occurring after 1994, these rates are still high (Brener, Simon, Krug, & Lowry, 1999; Centers for Disease Control, 2001; Federal Bureau of Investigation, 2000, 2001). For example, the incidence of arrests for violent offenses for boys younger than 18 is disproportionately high (Federal Bureau of Investigation, 2000, 2001) and homicide has become a leading cause of death for adolescents (Anderson, Kochanek & Murphy, 1997; Centers for Disease Control, 2001; Federal Bureau of Investigation, 2000). Aggression, which can be defined as "destructive behavior with intent to inflict harm or physical damage" (Marwick, 1996, p. 90), remains one of the most difficult problems to study and treat in adolescents (Malone, Luebbert, Pena-Ariet, Biesecker, & Delaney, 1994).

DSM-IV does not classify adolescent aggression as a separate disorder; rather, it is a major characteristic in a number of Axis I and II disorders, such as conduct disorder and antisocial personality (American Psychiatric Association, 1994). Research on aggression has revealed that the major type of aggression is characterized by high levels of arousal and poor modulation of behavior (Vitiello, Behar, Hunt, Stoff, & Ricciuti, 1990; Vitiello & Stoff, 1997). This type of aggression is mostly a defensive fight-or-flight response to a perceived threat.

Aggression has been associated with both high levels of arousal and underarousal, suggesting that aggressive behavior may arise from distinct biopsychosocial pathways in different individuals (Scarpa & Raine, 1997; Raine, 1996). The lack of a generally accepted treatment for aggression (Malone et al., 1994) may be explained by the different mechanisms underlying each of these two types of aggression (Vitie-llo & Stoff, 1997). These different aggressive styles complicate the assessment and treatment of aggression due to their distinct mechanisms of action.

As with some mood disorders, aggression has also been characterized by impulsivity, hostility, anger, and fear (Atkins & Stoff, 1993; Scarpa & Raine, 1997; Vitiello & Stoff, 1997). Mood disturbances including anxiety and depression have been associated with aggression (Botsis et al., 1997; van Praag, 2001), and research on aggressive behavior in children and adolescents indicates that anxiety may help predict the type of aggression being displayed. For example, while bullies show overt aggressive behavior marked by low levels of anxiety and underarousal, victims show indirect aggression marked by high anxiety levels and arousal (Craig, 1998). The absence of physical affection or the presence of neglect and physical abuse may create emotional traumas that result in either heightened sensory thresholds leading to underresponsivity to stimulation (Orbach, Mikulincer, King, Cohen, & Stein, 1997) or altered neurological development leading to physiological overreactivity (Dodge, Bates, & Pettit, 1990). These imbal ances in environmental reactivity may lead to aggressive behaviors through increased arousal-seeking behaviors or oversensitivity to stimulation.

Due to the lack of effectiveness of any single behavioral or cognitive treatment for all types of aggression (stemming from the complex mechanisms underlying aggressive behavior) and the unpleasant side effects of psychotropic drugs, complementary therapies are being investigated, such as training in aikido (a martial art similar to karate or judo) as a means of reducing aggressiveness in youth (Delva, 1995). Studies on biofeedback (Braud, 1978) and relaxation training (Dangel, Deschner, & Rasp, 1989; McPhail & Chamove, 1989), for example, suggest that these therapies are effective in diminishing aggressive behavior. Although massage therapy reduces depression and anxiety, factors noted to be related to aggression, massage therapy has not been tried with aggression. Massage therapy has, however, been tried with children and adolescents hospitalized for conduct disorder and depression (Field et al., 1992). Following 5 days of massage, the children and adolescents were less anxious and depressed and had lower s alivary cortisol (stress hormone) levels. Other studies have shown that massage therapy decreases anxiety and depression (Field, Lasko, Mundy, Henteleff, Talpins, & Dowling, 1996; Field, Seligman, et al., 1996). The present study evaluated the effects of 5 weeks of massage therapy versus relaxation therapy on aggressive adolescents.

METHOD

Participants

The participants were recruited from a child and adolescent psychiatry outpatient clinic at a major urban university hospital if they met the following criteria: (1) score of 10 or higher on the behavior component of the Overt Aggression Scale, (2) no acute suicidal threat, (3) no psychosis, (4) no mental retardation, (5) medically stable, and (6) able to understand and communicate. The sample was 53% male, ranged from 9 to 14 years old (M age 11.5), caine from predominantly lower socioeconomic status families (M = 4.6 on the Hollingshead Two-Factor Index), and was 71% African American and 29% Hispanic. The participants had an average score of 25.8 (SD = 13.9) on the Overt Aggression Scale.

Following recruitment, the participants and their guardians were contacted by a research associate who offered a financial incentive ($10 after each therapy and $30 upon completion of the study) for participating in "the study of a new therapy for reducing behavior problems." The participants were then stratified by age and aggression subtype and randomly assigned to a massage therapy group or a progressive muscle relaxation, attention control group before their first visit.

Aggression Questionnaire. This is a 10-item, yes/no questionnaire designed to assess the type of aggression (Vitiello et al., 1990). Participants were considered predatory aggressive if they had a score of 3 or greater, or affective aggressive if they had a score of -3 or lower. Participants with scores between 3 and -3 were considered mixed, predatory/affective aggressive. Examples of items include the following: "exposes self to physical harm when aggressive" and "expresses remorse after aggression." This questionnaire shows good reliability-stability (interclass correlation r = .77 for the predatory score and .71 for the affective score) and good internal consistency (Cronbach's alpha .73) (Vitiello et al., 1990).

Research on aggression has revealed two qualitatively different subtypes of aggression characterized by distinct psychophysiological mechanisms. Affective aggression is reactive to the environment and is characterized by high levels of arousal and poor modulation of behavior. Predatory or proactive aggression is goal-oriented and is characterized by low autonomic arousal and planned behavior (Vitiello, Behar, Hunt, Stoff, & Ricciuti, 1990; Vitiello & Stoff, 1997). In order to control for these subtypes of aggression, we randomly stratified individuals based on their aggressive style as determined by the Aggression Questionnaire. Participants in the massage group were 11% predatory, 22% affective, and 67% mixed, while the attention control group had 12% predatory, 25% affective, and 63% mixed adolescents. The groups did not significantly differ on this distribution or on their demographic characteristics (see Table 1 for results of chi-square tests and independent sample t tests). This distribution is consiste nt with the literature, which indicates that a large percentage of aggressive individuals show aspects of both types of aggression (i.e., mixed aggression) (Vitiello & Stoff, 1997).

Procedure

Massage therapy. The participants in this group received two 20-minute chair massage therapy sessions per week for 5 weeks. The professional massage therapist (different therapist each session) administered a standard massage procedure, with the adolescent sitting fully clothed in the special massage chair. This consisted of long, broad stroking with moderate pressure to the back: (1) compression to the back (parallel to the spine) from the shoulders to the base of the spine, (2) compression to the entire back (adding rocking), (3) trapezius squeeze, (4) finger pressure on the shoulder, (5) finger pressure along the length of the spine and back, and (6) circular stroking of the hips; arms: (1) arms dropped to the side with arms kneaded from shoulder to lower arm, and (2) pressing down on upper and lower arms; hands: (1) entire hands massaged and pulling of fingers, (2) the fleshy part of the palm pressed between the thumb and index finger for 15 to 20 seconds, and (3) pulling of the arms both in lateral and s uperior directions; and neck: (1) kneading area of cervical vertebrae, (2) finger pressure along base of skull and along side of neck, (3) scalp massage, and (4) pressing down on trapezius with finger pressure and squeezing continuing down the arms.

Relaxation therapy. The participants in this group were guided through two 20-minute progressive muscle relaxation sessions per week for 5 weeks by a researcher or a massage therapist (different therapist each session). The progressive muscle relaxation routine developed by Jacobsen (1929) was used. It consisted of a calm and soothing voice instructing the adolescents (who were sitting in the massage chair) to successfully tense and relax each of the major muscle groups in the back, arms, face, and neck.

Assessments

On the first and last days of the study, prior to the therapy, the adolescents' legal guardians were asked to complete a demographic questionnaire, the Overt Aggression Scale, and the Child Behavior Checklist. The adolescents were asked to complete the SCL-90R and the State Anxiety Inventory for Children (STAI-C).

Overt Aggression Scale (OAS). To establish recruitment criteria and to evaluate overall aggressive behavior, parents were asked to complete the OAS, a 20-item checklist on aggressive behavior (Yudofsky et al., 1986). For the purposes of this study, only the behavior component was administered. This component consists of 16 items relating to verbal aggression (i.e., curses viciously, uses foul language in anger, makes moderate threats to self or others), aggression against objects (i.e., breaks objects, smashes windows), aggression against self (i.e., small cuts or bruises, minor burns), and aggression against others (i.e., attacks others, causing mild to moderate physical injury). This scale has been shown to have good interclass correlation coefficients for reliability in children (r = .73 for verbal aggression, .86 for physical aggression against objects, 1.00 for physical aggression against self, .90 for physical aggression against others) (Yudofsky et al., 1986).

Child Behavior Checklist (CBCL). The CBCL (Achenbach, 1994; Achenbach & Edelbrock, 1987) is a 100-item questionnaire scored on a 3-point scale, from 0, indicating absence of behavior problems, to 2, indicating strong presence of problems. The instrument completed by the parents consisted of only the 47 questions corresponding to the aggression (i.e., physically attacks people), delinquent (i.e., disobedient at home), and hostility (i.e., impulsive or acts without thinking) sub-scales.

SCL-90R. The SCL-90R (Derogatis, 1983) is a 90-item questionnaire consisting of 9 subscales that measure distress experienced over the past week. For the purposes of this study, only the items corresponding to the hostility subscale were used. SCL-90R items consist of simple statements (e.g., "Having urges to break or smash things") followed by 5 choices regarding severity (0 = not at all to 4 = extremely). The SCL-90R has high internal consistency (M coefficient = .84) and test-retest reliability (M coefficient = .84) and acceptable construct validity (Derogatis, 1983).

State Anxiety Inventory for Children (STAI-C). The STAI-C (Spielberger, 1973) is an adaptation of the State Anxiety Inventory and is designed to assess the transitory level of anxiety in children and young adolescents. It consists of 20 items, each with a choice of 3 levels of severity (i.e., I feel very nervous, nervous, or not nervous). Spielberger (1973) reported alpha coefficients of .87 for females and .82 for males and an acceptable level of concurrent validity (r = .75). Based on other massage therapy and relaxation therapy studies, STAI-C scores were expected to be lower following the therapy sessions (Field et al., 1992; Platania-Solazzo et al., 1992).

RESULTS

Repeated-measures t tests revealed the following (see Tables 2 and 3): (1) only the massage therapy group showed a significant decrease in OAS aggression, t(8) = 2.42, p < .05; (2) only the massage therapy group showed a significant decrease in CBCL aggression scores, t(8) = 2.46, p < .05; (3) only the massage therapy group showed a significant decrease in hostility scores on the SCL-90R, t(8) = 2.79, p < .05; and (4) only the massage therapy group showed a significant decrease in state anxiety on the STAI-C after the first session, t(8) = 8.02, p < .001, and the last session, t(8) = 3.18, p < .05.

DISCUSSION

Anxiety has been associated with aggressive behavior (Botsis et al., 1997; van Praag, 2001) and may influence the expression and modulation of aggressive behavior through its effects on social interactions (Dadds, Marett, & Rapee, 1996). Participants receiving massage therapy reported feeling significantly less anxious after a 20-minute session on both the first and last days of treatment, but participants receiving progressive muscle relaxation did not. This is consistent with previous studies showing that, for children and adolescents, anxiety and stress hormone levels decrease following massage therapy (Field et al., 1992).

Following five weeks of treatment, the aggressive participants who received massage therapy also reported feeling significantly less hastile, but participants who received progressive muscle relaxation did not. In addition, those in the massage therapy group were perceived as being significantly less aggressive by their parents. The reduction in aggression and hostility among participants receiving massage therapy may stem from a reduction in anxiety noted in this study and in the reduction in stress hormone levels noted in previous studies (Field et al., 1992; Field, Seligman, et al., 1996), leading to lower arousal levels and therefore better control over impulsive and reactive behavior.

Unfortunately, the sample size did not allow us to explore differences between participants in the predatory, affective, and mixed aggression groups. It would also have been interesting to determine if massage therapy, through its reduction of anxiety, was more effective in reducing affective versus predatory aggressive behaviors. A future study should utilize a larger sample size that would allow for the comparison of these variables.

The interaction between the participants and the massage therapists may have contributed to the reduction in hostility. The positive rapport established during the course of treatment may have allowed these youth to experience a nonthreatening relationship with an adult, one that might not have been present in their everyday lives. The reduction in hostility and aggression noted in this study could have also derived from biobehavioral changes following the massage therapy. Further studies should document the effects of massage therapy on aggressive adolescents' relationships and their neurochemical and psychophysiological responses to this therapy.

Table 1

Means (and Standard Deviations) for Demographics

 Massage Relaxation t(15) p

Socioeconomic Status 4.6 4.6 .27 N.S.
 (0.5) (0.5)
Age 11.6 11.5 .17 N.S.
 (3.3) (1.9)

 [chi square] p

Gender .05 N.S.
 Male 56% 50%
 Female 44% 50%
Ethnicity .14 N.S.
 African American 67% 75%
 Hispanic 33% 25%
Aggressive Type .03 N.S.
 Predatory 11% 12%
 Affective 22% 25%
 Mixed 67% 63%

Table 2

Means (and Standard Deviations) for Longer Term Effects

 Message Relaxation
Type of Aggression Day 1 Day 10 Day 1 Day 10

OAS-Total 27.67 (a) 22.ll (b) 23.75 (a) 22.25 (a)
 (16.79) (14.14) (10.35) (8.73)
CBCL-Aggression 15.22 (a) 13.11 (b) 16.50 (a) 15.63 (a)
 (2.82) (2.03) (2.45) (2.39)
CBCL-Delinquent 9.78 (a) 8.89 (b) 9.13 (a) 7.88 (a)
 (2.91) (3.26) (3.94) (3.09)
SCL-90R Hostility 12.67 (a) lO.78 (b) 9.50 (a) 7.75 (a)
 (3.77) (4.21) (4.75) (2.55)

Note. Different subscripts indicate significant Day1/Day10 differences
between means at p < .05.

Table 3

Means (and Standard Deviations) for Immediate Effects

 Message Relaxation
 Pre Post Pre Post

STAI-C Day 1 37.22 (a) 31.89 (b) 35.25 (a) 34.13 (a)
 (3.07) (7.24) (5.52) (6.22)
STAI-C Day 10 35.89 (a) 27.22 (b) 33.75 (a) 31.50 (a)
 (4.54) (4.92) (5.39) (7.23)

Note. Different subscripts indicate significant pre-post differences
between means at p < .05.

REFERENCES

Achenbach, T. (1994). Child Behavior Checklist and related instruments. In M. E. Maruish et al. (Eds.), The use of psychological testing for treatment planning and outcome assessment (pp. 517-549). Mahwah, NJ: Lawrence Erlbaum Associates.

Achenbach, T., & Edelbrock, C. (1987). Manual for the Youth Self-Report and Profile. Burlington, VT: University of Vermont Department of Psychiatry.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Association.

Anderson, R. N., Kochanek, K. D., & Murphy, S. L. (1997). Report of final mortality statistics, 1995. Monthly Vital Statistics Report, 45, 11.

Atkins, M. S., & Staff, D. M. (1993). Instrumental and hostile aggression in childhood disruptive behavior disorders. Journal of Abnormal Child Psychology, 21, 165-179.

Botsis, A. J., Soldatos, C. R., & Stefanis, C. N. (1997). Suicide: Biopsychological approaches. Amsterdam: Elsevier Science.

Braud, L. W. (1978). The effects of frontal EMG biofeedback and progressive muscle relaxation upon hyperactivity and its behavioral concomitants. Biofeedback and Self Regulation, 3(1), 69-89.

Brener, N. D., Simon, T. R., Krug, E. G., & Lowry, R. (1999). Recent trends in violence-related behaviors among high school students in the United States. Journal of the American Medical Association, 282, 440-446.

Centers for Disease Control and Prevention. (2001). Deaths: Final data for 1999. National Vital Statistics Report, 49(8).

Craig, W. M. (1998). The relation among bullying, victimization, depression, anxiety, and aggression in school aged children. Personality and Individual Differences, 24, 123-130.

Dadds, M. R., Marett, P. M., & Rapee, R. M. (1996). Family process and child anxiety and aggression: An observational analysis. Journal of Abnormal Child Psychology, 24, 715-734.

Dangel, R. F., Deschner, J. P., & Rasp, R. R. (1989). Anger control training for adolescents in residential treatment. Behavior Modification, 13, 447-458.

Delva, T. J. (1995). Does brief aikido training reduce aggression of youth? Perceptual and Motor Skills, 80, 297-298.

Derogatis, L. R. (1983). SCL-90R: Administration, scoring and procedures manual (2nd ed.). Townson, MD: Clinical Psychometric Research.

Dodge, K. A., Bates, J. E., & Pettit, G. S. (1990). Mechanisms in the cycle of violence. Science, 250, 1678-1683.

Federal Bureau of Investigation. (2000). Crime report in the United States 1999. Washington, DC: U.S. Department of Justice.

Federal Bureau of Investigation. U.S. Department of Justice. (2001). Crime index trends, 2000: Preliminary figures. Washington, DC: U.S. Government Printing Office.

Field, T., Ironson, G., Scafidi, F., Nawrocki, T., Goncalves, A., Pickens, J., Fox, N., Schanberg, S., & Kuhn, C. (1996). Massage therapy reduces anxiety and enhances EEG pattern of alertness and math computations. International Journal of Neuroscience, 86, 197-205.

Field, T., Lasko, D., Mundy, P., Henteleff, T., Talpins, S., & Dowling, M. (1996). Autistic children's attentiveness and responsivity improved after touch therapy. Journal of Autism and Developmental Disorders, 27, 329-334.

Field, T., Morrow, C., Valdeon, C., Larson, S., Kuhn, C., & Schanberg, S. (1992). Massage reduces anxiety in child and adolescent psychiatric patients. Journal of the American Academy of Child and Adolescent Psychiatry, 31(1), 125-131.

Field, T., Seligman, S., Scafidi, F., & Schanberg, S. (1996). Alleviating posttraumatic stress in children following hurricane Andrew. Journal of Applied Developmental Psychology, 17, 37-50.

Hernandez-Reif, M., Dieter, J., Field, T., Swerdlow, B., & Diego, M. (1998). Migraine headaches are reduced by massage therapy. International Journal of Neuroscience, 96, 1-11.

Human Capital Incentive. (1997). Reducing violence: A research agenda (Report 5). APS Observer. Washington, DC.

Jacobsen, E. (1929). Progressive relaxation. Chicago: University of Chicago Press.

Malone, R. P., Luebbert, J., Pena-Ariet, M., Biesecker, K., & Delaney, M. A. (1994). The Overt Aggression Scale in a study of lithium in aggressive conduct disorder. Psychopharmacology Bulletin, 30(2), 215-218.

Marwick, C. (1996). Childhood aggression needs definition, therapy. Journal of the American Medical Association, 275(2), 90.

McPhail, C. H., & Chamove, A. S. (1989). Relaxation reduces disruption in mentally handicapped adults. Journal of Mental Deficiency Research, 33, 399-406.

Orbach, I., Mikulincer, M., King, R., Cohen, D., & Stein, D. (1997). Thresholds and tolerance of physical pain in suicidal and nonsuicidal adolescents. Journal of Consulting and Clinical Psychology, 65(4), 646-652.

Platania-Solazzo, A., Field, T., Blank, J., Seligman, F., Kuhn, C., Schanberg, S., & Saab, P. (1992). Relaxation therapy reduces anxiety in child and adolescent psychiatric patients. Acta Paedopsychiatrica, 55, 115-120.

Raine, A. (1996). Autonomic nervous system factors underlying disinhibited, antisocial, and violent behavior: Biosocial perspectives and treatment implications. Annals of the New York Academy of Sciences, 794, 46-59.

Scarpa, A., & Raine, A. (1997). Psychophysiology of anger and violent behavior. Psychiatric Clinics of North America, 20(2), 375-393.

Spielberger, C. D. (1973). Manual for the State-Trait Anxiety Inventory for Children. Palo Alto: Consulting Psychologists Press.

van Praag, H. M. (2001). Anxiety/aggression-driven depression: A paradigm of functionalization and verticalization of psychiatric diagnosis. Progress in Neuropsychopharmacology and Biological Psychiatry, 25, 893-924.

Vitiello, B., Behar, D., Hunt, J., Stoff, D., & Ricciuti, A. (1990). Subtyping aggression in children and adolescents. Journal of Neuropsychiatry, 2(2), 189-192.

Vitiello, B., & Stoff, D. M. (1997). Subtypes of aggression and their relevance to child psychiatry. Journal of the Academy of Child and Adolescent Psychiatry, 36(3), 307-315.

Yudofsky, S. C., Silver, J. M., Jackson, W., Endicott, J., & Williams, D. (1986). The Overt Aggression Scale for the objective rating of verbal and physical aggression. American Journal of Psychiatry, 143(1), 35-39.

The authors would like to thank the mothers and adolescents who participated in this study. This research was supported by an NIMH Senior Research Scientist Award (MH00331) and an NIMH merit award (MH46586) to Tiffany Field and funding from Johnson and Johnson.

Miguel A. Diego, Tiffany Field, and Maria Hernandez-Reif, Touch Research Institutes, University of Miami School of Medicine.

John A. Shaw, Eugenio M. Rothe, Daniel Castellanos, and Linda Mesner, Department of Psychiatry, University of Miami School of Medicine.

Reprint requests to Tiffany Field, Touch Research Institutes, University of Miami School of Medicine, Department of Pediatrics (D-820), P.O. Box 016820, Miami, Florida 33101. Electronic mail may be sent to tfield@med.miami.edu.