Adolescent female vomiters and alcohol abuse: differences on violence measures.

Abstract: Numerous studies on adolescent risk behaviors have defined bulimic behavior by examining responses to paper-and-pencil measures on vomiting behaviors. Alcohol abuse among adolescent populations is a common practice and can be a factor in vomiting behaviors. Few studies have differentiated between groups that abuse alcohol and also display vomiting behavior. The purpose of this study was to compare vomiting and alcohol abuse female groups (N = 1170) on violence measures using data from the 1998 version of the Alabama Adolescent Survey. Results indicate a relationship between the incidence of violence with both vomiting behavior and alcohol abuse.

**********

Eating disorders are relatively common in the United States. Conservative estimates indicate that 5-10% of women have an eating disorder of some kind (Eating Disorder Awareness and Prevention, 2000). While most of the attention has focused on college women, 0.5-1% of adolescent girls also have an issue with eating disorders (K-Seidenfeld & Ricket, 2001). A co-morbid risk factor associated with eating disorders is substance abuse. This may be due to personality characteristics that may be complicated by anxiety and impulsivity (Grau & Orter, 1999; Loxton & Dawe, 2001; Vitousek & Manke, 1994). The need to control is common to both substance abuse and disordered eating behavior and if one examines the characteristics of women being clinically treated for eating disorders, more than 50% displayed alcohol abuse (Watts & Ellis, 1992).

Substance abusing women with eating disorders seem to follow the same course of addiction as those abusers without eating disorders (Specker, Westermeyer, & Thuras, 2000). However, women with both afflictions tend to be younger, more often single, and more likely to be living with family or friends (Specker, Westermeyer, & Thuras, 2000). Cooley and Toray (2001) found that bulimic symptoms increased as alcohol use increased. Other studies have shown that female adult children of alcoholics are at risk for eating disorders, interpersonal problems, and sexual abuse (Ackerman & Gondolf, 1991; Berkowitz & Perkins, 1988; Bulik, 1987; Cooley & Toray, 2000). These individuals also showed prevalence of a childhood environment of violence (Stout & Mintz, 1996). Some researchers have concluded that both disordered eating behavior and substance abuse are used as a coping mechanism for girls dealing with depression and difficult family and peer relationships (Watts & Ellis, 1992).

Potential triggers of disordered eating include anxiety associated with the threat of emotional and physical violence or other traumatic experiences (Lyubomirsky, Casper, & Sousa, 2001). One traumatic experience that has been frequently implicated as a potential trigger is sexual abuse. A number of studies suggest that childhood sexual abuse is a general risk factor for bulimia nervosa (Perkins & Luster, 1999; Romans, Gendall, Martin, et al, 2001; Wonderlich, Crosby, Mitchel, et al, 2001; Wonderlich, Brewerton, Jocic et al, 1997; Wonderlich & Donaldson, 1996). As much as 25% of bulimic behaviors have been attributed to childhood sexual abuse (Thompson et al., 2001; Wonderlich, Wilsnack, Wilsnack, et al, 1996).

Very few studies have utilized childhood or adolescent populations to examine both bulimia and sexual abuse issues. Furthermore, few studies have examined vomiting as a dietary practice and how it is complicated by alcohol abuse (Loxton & Dawe, 2001; Rafiroiu, Sargent, Parra-Medina, Drane, & Valois, 2003). The purpose of this study was to examine vomiting behaviors and alcohol abuse among a group of rural southern adolescent females. Of particular interest were the associations between violence measures, vomiting and alcohol abuse status.

METHOD

INSTRUMENT

This study utilized data from the 1998 Alabama Adolescent Survey (AAS). The AAS was previously administered in 1988, 1990, and 1993 (Adcock & Nagy, 1994; Nagy & Adcock, 1991, 1992). In its current form, the 109-item AAS has core questions that have test-retest reliability exceeding .80 (Nagy & Adcock, 1994).

SAMPLING

A voluntary sample of adolescents was drawn from 16 school districts in Alabama during 1998; half of the school districts had participated in previous surveys. Flyers notifying superintendents of the survey had been distributed at a state-wide superintendents' retreat and resulted in eight new superintendents requesting to take part in the 1998 survey.

PARTICIPANTS

The University of Alabama Institutional Review Board approved instrumentation and recruitment procedures. Participants included 9th and 10th grade students (N = 2272) present on days when data were collected. In smaller rural districts, all students participated. In semi-rural and urban districts, classes were selected at random from the participating schools. Surveys were completed in settings determined by school administrators. After listening to instructions about anonymity and voluntary participation, students completed the inventory and then deposited their completed answer sheets in a large enclosed collection box. A review of Alabama Health Department data on morbidity and mortality rates and teenage pregnancy and STDs has shown that the participating sample did not differ from adolescents in counties in the southern and central regions of Alabama. Although 2,272 students completed the survey, the current study only includes responses from the females classifying themselves as Caucasian or African American (n = 1170).

MEASURES

Group Measures: Responses to items on vomiting and binge-drinking behaviors were used to separate respondents into four groups. The vomiting measure asked students to indicate how many times they had thrown up on purpose to lose weight during the past month. A yes response at any frequency categorized a student as practicing vomiting behaviors. The binge-drinking measure asked students whether they had consumed five or more alcoholic drinks at one sitting during the past two weeks. A yes response on this measure categorized the student as alcohol abusing. Using these two measures four groups were established: No Vomiting and No Binge- drinking (NV-NBd), Vomiting Only (V), Binge-drinking Only (Bd), and both Vomiting and Binge-drinking (VBd).

Violence Measures: Seven items examined issues of assault and coercion. Coercion items asked:

* How many times over the past 12 months did someone take something from you by using force or by threatening to hurt you?

* How many times did someone threaten to hurt you but not actually hurt you?

* How many times did a girl attack you in a way that you were left with a bruise or a scar?

* How many times did a boy attack you in a way that you were left with a bruise or a scar?

* Have you ever been forced to have sex or forced to do sexual things that you really didn't want to do?

* Have you ever been forced to have intercourse (sex)?

One measure addressed assault and consisted of an item asking "How many times did you take something from someone by using force or by threatening to hurt them?" The violence measures also were categorized into yes or no responses.

ANALYSIS

Data from the survey were entered into the Statistical Package for the Social Sciences (SPSS Version 12.0.1). Univariate analyses, including the use of frequencies, were conducted for each demographic variable and responses to the dependent and independent measures. Chi-square tests were used to test for differences between groups on each measure, using a p value of <.05 to determine statistically significant differences.

RESULTS

A total of 1277 adolescent females took part in the AAS. Only responses from Caucasian (56%) and African American students (44%) were included in the current analyses due to few students from other ethnic groups. The final sample consisted of 1,170 young women. On average, the young women were 15.5 years of age with a median age of 15 years. Within this group, 53% lived with both their mother and father, 28% lived with one parent only, 4% lived with their grandparents, and 15% reported other living arrangements. Demographic information for this sample is presented in Table 1.

When asked about vomiting behavior, 12% of the sample reported making themselves vomit to lose weight during the past month. In addition, 29% reported having at least one 5-drink occasion in the past month. A review of the four groups revealed that the No Vomiting and No Binge-drinking-group (NV-NBd) comprised the largest number of students (n = 743) followed by the Binge-drinking Only group (Bd; n = 285), the Vomit Only Group (V;n = 83), and the both Vomiting and Binge-drinking-group (VBd; n = 59).

With respect to the violence measures, young women most frequently reported being threatened, but not actually hurt (44%), followed by being attacked by a boy (24%). The young women in this sample were least likely to report being forced to have sex (8%). Table 2 shows the proportion of young women indicating that they had experienced each violence measure at least once by vomiting and binge-drinking groups.

The four groups were then compared on each of the violence measures using Chi-square analysis. Two-by-two post hoc comparisons (p=.05) identified specific groups that were different on each violence measure. Table 3 identifies the direction of significant differences between the groups.

The analysis utilized seven independent measures compared across six groups, resulting in forty-two comparisons. Nineteen significant group differences resulted across the seven violence measures. A review of the group comparisons focused on establishing patterns among the various groups. Given the existing literature, it was hypothesized that group differences would be greatest between the NV-NBd group and the other three groups (V, Bd, VBd), with the VBd group portraying the most negative profile.

A review of the six columns in Table 3 provides support for this hypothesis. Thirteen of the nineteen significant comparisons included the NV-NBd group. A review of percentages for this group in Table 2 confirms lower rates for this group on each of the violence measures. Ten significant comparisons involved the VBd group and all pointed in the anticipated direction of VBd being the group at highest risk of experiencing violence.

An examination of established patterns was most striking for the sexual abuse measures. There were significant differences across the majority of group comparisons. Again, the lowest risk of sexual abuse was for the NV-NBd group and the highest risk was for the VBd group. Individuals in the V group had the second highest risk for sexual abuse revealing a statistically significant relationship between sexual violence and vomiting to lose weight, regardless of an individual's binge-drinking status.

DISCUSSION

The purpose of this investigation was to explore possible relationships between violence measures, vomiting with the intention of losing weight, and alcohol abuse status in rural southern adolescent females. Vomiting group members were defined as those individuals who reported having vomited at least once in the past month for the purpose of losing weight. Binge-drinkers, or alcohol abusers, were those students who reported drinking five or more alcoholic drinks at one sitting during the past two weeks.

Definite patterns arose giving an overall picture of each group's risk. Significant differences existed between the Vomiting Only group when compared with the No-Vomiting No-Binge-drinking group. Although neither of these groups abused alcohol, the Vomiting Only group was more likely to report encountering violence than the group participating in neither risky behavior. The most similar groups, the Vomiting Only and Binge-Drinking Only groups, showed only two significant differences. These differences revealed that the females who vomited only were more likely than the binge-drinking only females to have been threatened and to have been forced to have sex. Clearly, there is a relationship between vomiting to lose weight and experiencing violence that is independent of alcohol abuse.

The two groups who reported vomiting showed significant differences only when examining violent sexual activity, with those abusing alcohol and vomiting at higher risk of forced sexual activity. The only differences seen between the two drinking groups, the Binge-drinking Only group and the Vomiting and Binge-drinking group, occurred when analyzing the forced sexual activity measures. Those who abused alcohol and vomited experienced forced sexual activity at higher rates.

Students participating in both drinking and vomiting behaviors were consistently the most at risk for participating in or experiencing violence. Correspondingly, the females who do not binge-drink and do not display vomiting behaviors were the least likely to be involved in or be victims of violence. These findings suggest that there is a relationship between the incidence of violence in this sample with both vomiting behavior and high alcohol consumption. Most importantly, binge drinking further complicates the risk for violence in individuals who vomit to lose weight.

Although the results of this study offer new information about the relationships between these variables, these findings should be viewed with caution. First, causality of eating disorders and alcohol abuse cannot be inferred from these results. Those individuals placed into the vomiting categories were done so based on their answers to a self-report questionnaire and not based on the diagnosis of an eating disorder by a health professional. As mentioned, the data is composed of self-report answers that may underestimate the actual occurrence of vomiting behaviors, binge-drinking, and violent events. Another issue is that participants who did not respond to the survey honestly were removed from the considered sample. It is possible that these individuals had a higher risk of experiencing sexual abuse. Additionally, each student in the sample was a volunteer from a rural southern school system. Therefore, the results were not from a randomly selected sample and may not apply universally.

Findings from this study support previous research linking substance abuse and traumatic experiences with disordered eating (Thompson, Wonderlich, Crosby, et al, 2001; Williams & Ricciardelli, 2003). Most importantly, this research looks at several measures of substance abuse, vomiting, and violence at the same time within a nonclinical sample. The results support the hypotheses that all these factors can be occurring at the same time. Vomiting to lose weight and violence are related independent of the effect of alcohol. Individuals with both the vomiting and substance abusing traits are at the highest risk for violence. This risk profile is particularly strong when considering sexual violence.

IMPLICATIONS

School counselors should be aware that these behaviors (vomiting to lose weight and binge drinking) and violent events are clustered, and may occur alone or in combination with other risk factors. This knowledge can assist professionals in identifying individuals suffering from violence and direct them in meeting the needs of those individuals.

This is of particular importance when creating a school-based intervention or education program. School health education professionals should collaborate with other educators and health workers, such as pediatricians and mental health professionals, to create inclusive programs addressing eating disorders, substance abuse, and violence collectively. These programs should focus on healthy ways to relieve stress, cope with the experiences of life, and have a sense of control over one's life without vomiting or abusing alcohol. Additionally, persons who binge-drink and vomit to lose weight share characteristics of a risk taking personality. Therefore, programs should address healthy ways to have fun and take positive risks in life.

An awareness of the co-occurrence of the factors in this study can help prevent the perpetuation of the cycle of eating disorders, substance abuse, and violence. Future research should examine the patterns of substance abuse and violence with adolescent females who skip meals more often than vomit to lose weight. Also, this study could be replicated examining the abuse of substances other than alcohol to explore whether these substances are equivocally related to eating disorders and violence, or if these other substances have a more substantial influence on vomiting behavior and the presence of violence in the population.

REFERENCES

Ackerman, R.J., & Gondolf, E.W. (1991). Adult children of alcoholics: The effects of background and treatment on ACOA symptoms. International Journal of the Addictions, 26, 1159-1172.

Adcock, A.G., Jacobs, D.P., & Nagy, S. (1999). Health risk behaviors among Alabama adolescents, 1988-1998. ASAHPERD Journal, 21(1).

Berkowitz, A., & Perkins, H. (1988). Personality characteristics of children of alcoholics. Journal of Consulting and Clinical Psychology, 56, 206-209.

Bulik, C.M. (1987). Drug and alcohol abuse by bulimic women and their families. American Journal of Psychiatry, 144, 1604-1606.

Cooley, E., & Toray, T. (2001). Disordered eating in college freshman women: A prospective study. Journal of American College Health, 49(5), 229-235.

Eating Disorders Awareness and Prevention, Inc. (2001). Eating disorders and their precursors. Retrieved on August 31, 2000, from http://www.eadp.org/edinfo/stats2.html

Grau, E., & Orter, G. (1999). Personality traits and alcohol consumption in a sample of non-alcoholic women. Personality and Individual Differences, 27, 1057-1066.

Hoek, H.W. (1995). The distribution of eating disorders. In K.D. Brownell and C.G. Fairburn (Eds.), Eating disorders and obesity: A comprehensive handbook (pp. 207-211). New York: Guilford Press.

K-Seidenfeld, M., & Rickert, V.I. (2001). Impact of anorexia, bulimia and obesity on the gynecologic health of adolescents. American Family Physician, 64(3), 445-450.

Loxton, N.J., & Dawe, S. (2001). Alcohol abuse and dysfunctional eating in adolescent girls: The influence of individual differences in sensitivity to reward and punishment. International Journal of Eating Disorders, 29, 455-462.

Lyubomirsky, S., Casper, R.C., & Sousa, L. (2001). What triggers abnormal eating in bulimic and nonbulimic women? The role of dissociative experiences, negative affect, and psychopathology. Psychology of Women Quarterly, 25, 223-232.

Nagy, S., & Adcock, A.G. (1991). The 1990 Alabama Adolescent Health Survey. ASAHPERD Journal, 14, 18-20.

Nagy, S., & Adcock, A.G. (1992). Summary report II, the Alabama Adolescent Health Survey: Health knowledge and behaviors. ERIC Resources in Education, 338, 608.

Nagy, S., Adcock, A.G., & Nagy, M.C. (1994). A comparison of risky health behaviors of sexually active, sexually abused and abstaining adolescents. Pediatrics, 93(4), 570-575.

Perkins, D.F., & Luster, T. (1999). The relationship between sexual abuse and purging: Findings from community-wide surveys of female adolescents. Child Abuse and Neglect, 23(4), 371-382.

Rafirouiu, A.C., Sargent, R.G., Parra-Medina, D., Drane, W.J., & Valois, R.F. (2003). Covariations of adolescent weight-control, health-risk and health-promoting behaviors. American Journal of Health Behavior, 27(1), 3-14.

Romans, S.E., Gendall, K.A., Martin, J.L., & Mullen, Paul E. (2001). Child sexual abuse and later disordered eating: A New Zealand epidemiological study. International Journal of Eating Disorders, 9, 380-392.

Specker, S., Westermeyer, J., & Thuras, P. (2000). Course and severity of substance abusing patients with comorbid eating disorders. Substance Abuse, 21(3), 137-147.

Stout, M.L., & Mintz, L.B. (1996). Differences among nonclinical college women with alcoholic mothers, alcoholic fathers, and nonalcoholic parents. Journal of Counseling Psychology, 43(4), 466-472.

Thompson, K., Wonderlich, S.A., Crosby R., et al. (2001). Sexual violence and weight control techniques among adolescent girls. International Journal of Eating Disorders, 29, 166-176.

Vitousek, K., & Manke, F. (1994). Personality variables and disorders in anorexia nervosa and bulimia nervosa. Journal of Abnormal Psychology, 103, 137-147.

Watts, D.W., & Ellis, A.M. (1992). Drug abuse and eating disorders: Prevention implications. Journal of Drug Education, 22(3), 223-240.

Williams, R.J., & Ricciardelli, L.A. (2003). Negative perceptions about self-control and identification with gender-role stereotypes related to binge eating, problem drinking, and to comorbidity among adolescents. Journal of Adolescent Health, 32, 66-72.

Wonderlich, S.A., Brewerton, T.D., Jocic, Z., et al. (1997). Relationship of childhood sexual abuse and eating disorders. Journal of the American Academy of Child and Adolescent Psychiatry, 36(8), 1107-1115.

Wonderlich, S., Crosby, R., Mitchel, J., Thompson, K., Redlin, J., Demuth, G. et al. (2001). Pathways mediating sexual abuse and eating disturbance in children. International Journal of Eating Disorders, 29, 270-279.

Wonderlich, S., & Donaldson, M.A. (1996). Eating disturbance and incest. Journal of Interpersonal Violence, 11(2), 195-207.

Wonderlich, S.A., Wilsnack, R.W., Wilsnack, S.C., & Harris, T.R. (1996). Childhood sexual abuse and bulimic behavior in a nationally representative sample. American Journal of Public Health, 86(8, Pt.1), 1082-1086.

CHES AREAS

Responsibility I

Competency A: Obtain health related data about social and cultural environments, growth and development factors, needs, and interests.

Sub-competencies 1: Select valid sources of information about health needs and interests.

Sub-competencies 2: Utilize computerized sources of health-related information.

Sub-competencies 3: Employ or develop appropriate data-gathering instruments.

Sub-competencies 4: Apply survey techniques to acquire health data.

Sub-competencies 5: Conduct health-related needs assessment in communities.

Competency B: Distinguish between behaviors that foster and those that hinder well-being.

Sub-competencies 1: Investigate physical, social, emotional, and intellectual factors influencing health behavior.

Sub-competencies 2: Identify behaviors that tend to promote or compromise health.

Sub-competencies 3: Recognize the role of learning and affective experiences in shaping patterns of health behavior.

Sub-competencies 4: Analyze social, cultural, economic, and political factors that influence health.

Competency C: Infer needs for health education on the basis of obtained data.

Sub-competencies 1: Analyze needs assessment data.

Sub-competencies 2: Determine priority areas of need for health education.

Lea G. Yerby, MS, CHES is affiliated with the Department of Health Science at The Unviersity of Alabama. M. Christine Nagy, PhD and Stephen Nagy, PhD are affiliated with Western Kentucky University. Address all correspondence to Lea G. Yerby, MS, Doctoral Student, The University of Alabama, Department of Health Science, 209 East Annex, Box 870311, Tuscaloosa, AL 35487-0311; PHONE: [phone]; FAX: [phone]; E-MAIL: [e-mail].

Table 1. Demographic Information of Sample (N = 1,170)

  Characteristic      %

Age
  13 or younger       9
  14                 41
  15                 41
  16                  9
  17 or older
Ethnicity
  Caucasian          56
  African American   44
Living Arrangement
  Mother & Father    53
  Mother Only        25
  Father Only         3
  Mother & Others     9
  Father & Others     2
  Grandparents        4
  Other               4

Table 2. Proportion of Responses on Each Violence Measure by Group.

                                      V        VBd      Bd      NV-NBd
                                    n = 83   n = 59   n = 285   n = 743
         Measure                      %        %        %         %

Something taken from you              21       32       22        17
Threatened, but not actually hurt     58       53       45        41
Attacked by a girl                    11       20       12         8
Attacked by a boy                     31       37       30        20
You threaten someone                  18       25       21        13
Forced to do sexual things            25       48       29        20
Forced to have sex                    12       22       11         5

* Note: V = Vomit Only; VBd = Binge-Drink and Vomit; Bd = Binge-Drink
Only; NVNBd = No Vomit, No Binge-Drink

Table 3. Direction of Data

Column                           1           2            3

         Measure              I vs II     I vs III     I vs IV

Something taken from you
Threatened, but not
  actually hurt                          II > III *   I > IV **
Attacked by a girl
Attacked by a boy                                     I > IV *
You threaten someone
Forced to do sexual things   I < II **
Forced to have sex           I < II **   I > III *    I > IV *

Column                          4             5             6

        Measure             II vs III     II vs IV      III vs IV

Something taken from you                 II > IV **
Threatened, but not
  actually hurt
Attacked by a girl                       II > IV **
Attacked by a boy                        II > IV **    III > IV ***
You threaten someone                     II > IV **    III > IV ***
Forced to do sexual
  things                   II > III **   II > IV ***   III > IV **
Forced to have sex         II > III *    II > IV ***   III > IV ***

Note: * = p < .05, ** = p [less than or equal to] .01,
*** = p [less than or equal to] .001

I = Vomiting Only, II = Vomiting and Binge-drinking, III =
Binge-drinking Only, N = No Vomiting and No Binge-drinking.