Depression
Eight recent studies focused on the association between eating disorders and depression (Carter, Bewell, Blackmore, & Woodside, 2006; Fornari et al., 1992; Garcia-Alba, 2004; Johnson, Cohen, Kasen, & Brook, 2006; Monteleone et al., 2005; Speranza et al., 2003a; Speranza, Corcos, Atger, Pterniti, & Jeammet, 2003b; Speranza et al., 2005). Both Garcia-Alba and Monteleone et al. suggested that the chemical alterations in the body observed among depressed and eating disorder patients may be associated with malnutrition. Garcia-Alba found that depression occurred more frequently among anorexics than controls (36% versus 20%), suggesting that depression was associated with alterations in neurotransmitters associated with starvation and significant weight loss.
Monteleone et al. (2005) found that brain-derived neurotrophic factor, thought to have anti-depressant effects, was reduced in underweight anorexics and bulimics, but not in overweight binge-eating disorder patients. Among eating disorder patients 37% of anorexics, 45.8% of bulimics and 16.6% of patients with binge-eating disorder had a comorbid depressive disorder. Since not all patients were depressed,researchers proposed that malnourishment lowered the levels of the brain-derived neurotrophic factor.
Speranza et al. (2003b) found thatpatients using “purely active weight control strategies” had fewer depressive symptoms (i.e., those who exercised had lower levels than those who binged and/or induced vomiting). Anorexic and bulimic subtypes scored higher than controls on the Beck Depression Inventory and the Depressive Experiences Questionnaire (Speranza et al., 2003a; 2003b), with bulimics or binge/purge anorexics scoring highest (Speranza et al., 2005). Researchers suggested that depression among bulimics might relate more to an inability to identify and describe feelings than to depression. Fornari at al. (1992) found that 75% of anorexics had at least one diagnosis of depression during their lifetime, and bulimic-anorexics showed the most depressive symptoms and major depressive disorder comorbidity. Comorbid depressive disorders occurred among 28.57% of bulimics, 54.17% of anorexics, and 72.22% of bulimic-anorexics. The most significant difference was between bulimics and bulimic-anorexics (p<.017).
Johnson et al. (2006) examined personality disorders as a risk factor for eating and weight problems in adulthood. Although depressive personality disorder is not an official psychiatric diagnosis, this study demonstrated a significant correlation between depressive personality disorder and increased risk for recurrent dietary restriction (p<.05) and recurrent binge eating during adulthood (p<.05). Carter et al. (2006) found that eating disorder participants reporting childhood sexual abuse had higher levels of depression (p=.008), anxiety (p=.000), low self-esteem (p=.007), and interpersonal problems (p=.003).
Five studies reported evidence suggesting a relationship between childhood sexual abuse and eating disorders (Carter et al., 2006; Deep, Lilenfeld, Plotincov, Pollice, & Kaye, 1999; Murray & Waller, 2002; Romans, Gendall, Martin, & Mullen, 2001; van Gerko, Hughes, Hamill, & Waller, 2005), with higher rates of childhood sexual abuse among eating disordered participants than controls. Deep et al. found childhood sexual abuse to be higher among participants with an eating disorder at 23% among anorexics (p<.03) and 37% among bulimics without substance dependence (p<.01). Bulimics with a comorbid substance abuse disorder demonstrated a significantly higher rate of childhood sexual abuse [65% compared to 7% of controls (p<.0001)] (Deep et al.). Among all sub-types, the sexual abuse preceded the eating disorder in a majority of cases.
Romans et al. (2001) studied clients primarily reporting sexual abuse to discover whether they might also have an eating disorder. Higher rates of anorexia and bulimia were found among the 254 women who experienced sexual abuse before the age of 16 compared to the 223 who had not. The group reporting childhood sexual abuse included 84.2% of the anorexics and 73.1% of the bulimics. Childhood sexual abuse clients reported being abused 10 or more times and experienced more intrusive forms of abuse, meaning attempted or completed intercourse (Romans et al.). Congruent with these findings, 48.1% of anorexics in Carter et al.’s (2006) study reported abuse, with 84% abusedon more than one occasion. Van Gerko et al. (2005) suggested that a history of childhood sexual abuse, rather than being a direct cause, may play a mediating role in linking other associated factors to the development of eating disorders, but did not identify these links.
Studies by Carter et al. (2006); Deep et al. (1999), Murray and Waller (2002); and van Gerko et al. (2005) supported the hypothesis that childhood sexual abuse is linked to bulimic behaviors and to disturbed body image. While Deep et al. found significantly higher rates of sexual abuse among all subtypes than controls (anorexics p<.01; bulimics p<.03), rates of childhood sexual abuse among bulimics with a comorbid substance abuse were significantly higher than all other participating groups (control group p<.0001; anorexics p<.01; bulimics without a comorbid substance abuse p<.03).
Van Gerko et al. (2005) found that 33.8% of women with purging behaviors reported childhood sexual abuse (p=.004), compared to 17.3% of those not purging. A higher level of binging, vomiting, laxative and diuretic abuse and more reports of greater concern about body shape were observed among participants reporting childhood sexual abuse. Carter et al. (2006) also found a difference in those who purged. Overall, the relationship between childhood sexual abuse and anorexia was significant (p=.018). Among purging anorexics, 65% reported sexual abuse, compared to 31% of restrictors (Carter et al.). Restrictors are those who do not regularly engage in binge/purge behaviors during their current episode of anorexia (Chitty et al., 2004).
Murray and Waller (2002) found a significant link between childhood sexual abuse and bulimic behavior test scores (BULIT) (p<.01) and Internalized Shame Scale scores (p<.05). Also, Internalized Shame Scale scores significantly predicted BULIT scores (p<.001), with significant links between any abuse by a family member and bulimic symptomology (p<.02) and between abuse by a family member and Internalized Shame Scale scores (p<.02). Researchers proposed that the internalized shame might arise from repeated sexual abuse viewed by victims as shameful to themselves as well as to the perpetrator.
Romans et al. (2001) concluded that high paternal over-control increased the risk of developing an eating disorder in women who had experienced childhood sexual abuse. Early menarche was also associated with bulimia in women experiencing childhood sexual abuse [68.4% experienced menarche before age 12 (p =0.001)].
Seven studies in this review examined anxiety disorders among participants with eating disorders (Carter et al., 2006; Fornari et al., 1992; Godart, Flament, Lecrubier, & Leammet, 2000; Godart et al., 2003; Johnson et al., 2006; Speranza et al., 2001; Thornton & Russell, 1997). Anorexics and bulimics were more likely than controls to have a comorbid anxiety disorder both currently or at some point during their lifetime (Godart et al., 2000; 2003; Johnson et al.). Congruently, Carter et al. found that eating disorder participants reporting childhood sexual abuse had higher levels of anxiety (p=.000) and more severe obsessive-compulsive symptoms (p=.002). According to Godart et al. (2000) 83% of anorexics and 71% of bulimics had at least one other lifetime anxiety disorder.
Other anxiety disorders often preceded the eating disorder (Godart et al., 2000; Godart et al., 2003; Speranza et al., 2001; Thornton & Russell, 1997). Godart et al. (2000) found that 75% of anorexics and 88% of bulimics with a comorbid anxiety disorder had another anxiety disorder at least one year prior to the onset of the eating disorder. Godart et al. (2003) found childhood separation anxiety was significantly greater (p<.01) among women with eating disorders, occurring in half of anorexics and more than half of bulimics (Godart et al., 2000).
Obsessive-compulsive disorder (OCD) was a commonly addressed anxiety disorder in eating disordered patients. Anorexics and bulimics scored higher than controls on diagnostic tests for this disorder; however, five studies demonstrated that anorexics were more likely to present with OCD (Fornari et al., 1992; Godart et al., 2000; 2003; Speranza et al., 2001; Thornton & Russell, 1997). Godart et al. reported a prevalence rate of 21% among anorexics, compared to no cases among bulimics in their 2003 study. Among 68 inpatients, Thornton and Russell reported comorbid OCD in 37% of anorexics, compared to only 3% of bulimics, with the diagnosis of OCD often preceding the diagnosis of the eating disorder (Thornton & Russell). Speranza et. al also found that OCD preceded the eating disorder in 65% of cases, occurred simultaneously in 17% of cases, and followed the eating disorder in the remaining 18 % of cases.
Seven studies examined the trait of perfectionism in persons with eating disorders (Bastiani, Rao, Weltzin, & Kaye, 1995; Bulik et al., 2003; Castro, Gila, Lahortiga, Saura, & Tro, 2004; Halmi et al., 2000; Romans et al., 2001; Sassaroli & Ruggiero, 2005; Waller & Hartley, 1994). Participants with eating disorders--particularly anorexics—scored higher than controls on multidimensional scales to measure perfectionism (Bastiani et al.; Bulik et al.; Castro et al.; Halmi et al.). Bulik et al. examined the association between perfectionism and psychiatric disorders using the Multidimensional Perfectionism Scale to determine which aspects of perfectionism were associated with ED. The strongest association with eating disorders was “concern over mistakes” (p<.01 for anorexics and p<.0001for bulimics). Higher scores on this subscale was also identified as a predictor of eating disorders, but was associated with lower rates of alcohol abuse and dependence (p<.05). Bulik et al. also found higher scores on the sub-scale “doubts about actions” to be associated with eating disorders and other anxiety disorders (p<.001 in anorexics; p<.0001 in bulimics).
Bastiani et al. (1995) found that the trait of perfectionism in anorexics continued after restoration of a healthy weight and suggested it as a contributing factor in resistance to treatment and in relapse. Both Bastiani et al. and Castro et al. (2004) suggested that the perfectionism stems from self-oriented expectations. Bastiani et al. found significantly higher scores among anorexics than controls on the Self-oriented perfectionism sub-scale of the Hewitt Multidimensional Perfectionism Scale (Hewitt MPS) (p≤.01 when comparing underweight anorexics to controls; p≤.1when comparing weight-restored anorexics to controls). There was no statistically significant difference between anorexics and controls on the Other-Oriented perfectionism sub-scale. Congruently, Castro et al. (2004) found that anorexics scored significantly higher on two sub-scales of the Child and Adolescents Perfectionism Scale. These sub-scales are “Self-oriented perfectionism” (p<.001) and “Perfectionistic self-presentation” (p<.001). However, there was no statistically significant difference between the scores of anorexics and control participants on the “Socially-prescribed perfectionism” sub-scale (p=.292).
In contrast, Waller and Hartley (1994) suggested a link to the expectations of others in their study of the association between patterns of family functioning and the psychopathology of eating disorders. They found that women with eating disorders reported that their parents had high expectations and were unusually disapproving, with this perception of inevitable disapproval a potential maintaining factor in eating disorders. Romans et al. (2001) found that high paternal over-control increased the risk of developing an eating disorder in women who had experienced childhood sexual abuse, and low maternal care was specifically associated with anorexia.
Sassaroli and Ruggiero (2005) studied the role of stress in eating disorders, in which perfectionism emerged as the primary personality predictor of eating disorders in contrast to other predictors or associated factors. They suggested that this occurs because perfectionism is a deeply rooted characteristic associated with eating disorders, present even when stress was not identified as being a mediating factor.
In five studies, researchers examined the relationship between early pubertal onset and eating disorders (Graber, Lewinsohn, Seeley, & Brooks-Gunn, 1997; Hayward et al., 1997; Heebink, Sunday, & Halmi, 1995; Romans et al., 2001; Stice, Presnell, & Bearman, 2001). Graber et al. found that early-maturing girls had higher lifetime rates of all psychiatric disorders, including eating disorders, compared to girls who matured at a more average rate (52.9% versus 37.8%). No specific cause-effect link was noted.
Heebink et al. (1995) found that girls with maturity fears may be at risk for development of eating disorders to avoid physiological changes associated with development during adolescence. Heebink et al. found onset of anorexia nervosa before age 14 to be associated with maturity fears. After controlling for depression, researchers found maturity fears to be significant in patients with primary amenorrhea versus those with secondary amenorrhea (p =.003). Results also revealed significantly higher levels of maturity fears when the onset occurred in adolescence rather than adulthood, proposing that these fears influenced adolescents to diet to avoid normal physiologic changes. Congruently, Hayward et al. (1997) suggested that early pubertal onset may introduce a higher risk for negative body image and weight concerns, and that the concerns over body image and weight may precede or possibly predict the eating disorder. Romans et al. (2001) also found an association between early menarche and bulimia in women who have experienced childhood sexual abuse [68.4% experienced menarche before the age of 12 (p =0.001)].
In contrast to these studies, Stice et al. (2001) did not find a direct significant relationship between early menarche and eating disorders; however, an indirect link was suggested. A strong relationship was found among early onset puberty and emotional disturbance and between early onset puberty and substance abuse. Early-maturing girls were at three times the risk for developing depression or substance abuse, and both a diagnosis of depression and substance abuse showed a significant comorbidity with eating disorders (p <.0001 and p<.05 respectively).
© 2007 UNIVERSITY OF CENTRAL FLORIDA UNDERGRADUATE RESEARCH JOURNAL -COMPUTER CENTER II ROOM 209- ORLANDO, FL 32816-0086 - (407) 823-3125
Office of Undergraduate Research
Contact the webmaster with website questions.
View PDF