Physiology & Behavior 98 (2009) 543–546 Contents lists available at ScienceDirect Physiology & Behavior j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p h b Early trauma and adult obesity: Is psychological dysfunction the mediating mechanism? Alberto D'Argenio a, Cristina Mazzi a, Luca Pecchioli a, Giorgio Di Lorenzo b, Alberto Siracusano b, Alfonso Troisi b,⁎ a b Obesity Center, INI Istituto Neurotraumatologico Italiano, Grottaferrata, Rome, Italy Department of Neurosciences, University of Rome Tor Vergata, Rome, Italy a r t i c l e i n f o Article history: Received 8 April 2009 Received in revised form 26 August 2009 Accepted 27 August 2009 Keywords: Early trauma Obesity Psychopathology Anxious attachment Eating disorder Bariatric surgery a b s t r a c t Several studies have shown that physical and/or sexual abuse during childhood may lead to the development of obesity later in life. Despite these consistent findings, the mechanism for the increased risk of obesity following developmental trauma is unknown. It has been suggested that psychological dysfunction, including the presence of disordered eating behavior, may account for the added risk of adult obesity. To test this hypothesis, we analyzed the prevalence and severity of different types of early traumatic life events, assessed the presence of coexisting psychiatric disorders and measured adult attachment style in a sample of 200 subjects including nonobese healthy volunteers and obese participants undergoing a psychiatric assessment to determine suitability for bariatric surgery. Participants who scored higher on a scale measuring the severity of traumatic events experienced during the first 15 years of their lives were more likely to be obese at the time of testing. The exclusion of the participants who experienced physical and/or sexual abuse did not change the results of statistical analysis. Severity of early trauma remained a significant predictor of adult obesity when the influence of psychiatric diagnosis and anxious attachment was taken into account. These findings suggest that: (1) not only sexual or physical abuse but also less severe forms of early-life stress are linked to the development of obesity later in life; and (2) psychological dysfunction is not the only mechanism mediating the elevated risk of obesity in persons exposed to early-life trauma. © 2009 Elsevier Inc. All rights reserved. 1. Introduction Over the last decade the science of developmental origins of health and disease has advanced rapidly. It is now clear that the etiology of many chronic diseases, such as coronary heart disease, type 2 diabetes and osteoporosis, lays not only in genetic predisposition or in adult lifestyle but also in the ways in which early-life events could affect subsequent biology and behavior [1]. Evidence that obesity might have a developmental origin came originally from a series of studies which showed that prenatal exposure to over- or undernutrition, rapid growth in early infancy, an early adiposity rebound in childhood, and early pubertal development are all associated with an increased risk of adult obesity [reviewed by 2,3]. In parallel with these findings focusing on early nutritional environment and physiological factors, several studies have shown that stressful emotional experiences during childhood may also lead to the development of obesity later in life [4–8]. ⁎ Corresponding author. Department of Neurosciences, School of Medicine, University of Rome Tor Vergata, Rome, Italy. E-mail address: alfonso.troisi@uniroma2.it (A. Troisi). 0031-9384/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.physbeh.2009.08.010 For some time, research on childhood adversity and adult obesity has focused on single types of early trauma, particularly sexual or physical abuse. However, recent research findings across diverse areas have highlighted the importance of considering varied developmental traumatic experiences when studying the long-term effects of psychological trauma early in life [9]. Accordingly, the first aim of this study was to ascertain whether not only sexual or physical abuse but also less severe forms of early-life stress (e.g., separation from parents, parents' marital problems, psychiatric illness of mother or father) are linked to the development of obesity later in life. Exposure to psychological trauma early in life is associated with multifaceted sequelae that may increase risk for obesity through different mechanisms. A recent review suggested that disordered eating behavior and/or psychological reactions to a traumatic experience may account for the added risk of adult obesity [10]. Childhood adversity is associated with increased risk for a variety of mental health problems which are characterized by intense negative emotional experience [11]. An individual's inability to cope with stress and negative emotions may be associated not only with disordered eating behavior leading to obesity but also with a reduced capacity to implement nutrition and physical activity plans suggested for weight loss or prevention of weight gain [5]. 544 A. D'Argenio et al. / Physiology & Behavior 98 (2009) 543–546 In the light of these previous findings, the second aim of this study was to ascertain whether the association between early trauma and adult obesity is explained by psychological dysfunction. Unlike previous studies, we expanded the definition of psychological dysfunction to include the assessment of anxious attachment, a dysfunctional personality trait that may, or may not, be associated with the presence of a psychiatric disorder. We hypothesized that, even when the diagnostic threshold for a psychiatric condition has not been reached, anxious attachment may represent a psychological factor mediating the association between early trauma and adult obesity. Our hypothesis was based on two related sets of data that have been repeatedly confirmed by studies of clinical and non-clinical populations. First, individuals who experience different types of early traumas are more likely to develop an anxious attachment orientation, thereby fostering vulnerability to stress and an exaggerated emotional reactivity [12,13]. Second, attachment anxiety may contribute to eating disorder symptoms [14] and lead to impulsive food intake and emotional eating [15–17]. 2. Method 2.1. Participants Three groups of participants were enrolled into the sample. The first group (CON, N = 50) was a convenience sample and included non-obese healthy volunteers with no current or past DSM-IV-TR diagnosis of psychiatric disorder. The second group (OB, N = 65) included obese participants with no current or past DSM-IV-TR diagnosis of psychiatric disorder. The third group (OBPSY, N = 85) included obese participants with a current DSM-IV-TR diagnosis of psychiatric disorder. Healthy volunteers were recruited among students in the medical school, paramedic staff members, and conscripts of the Italian army. Inclusion in the CON group required the absence of current or past psychiatric disorders, as confirmed by diagnostic interview, and a BMI comprised between 18 and 28. The decision to include some overweight control subjects (N = 15, BMI > 25) aimed at enrolling individuals with a BMI reflecting that of the Italian non-obese general population [18]. Overweight control subjects did not differ significantly from the rest of healthy volunteers on the measures of early trauma and anxious attachment. Obese participants (BMI > 30) were a consecutively evaluated series of individuals undergoing a psychiatric assessment to determine readiness and suitability for bariatric surgery. Their lifetime prevalence of psychiatric disorders corresponded to that reported for other samples of bariatric surgery patients [19]. Obese participants with severe medical disorders (e.g., cardiac disease, thyroid disease, decompensated diabetes), neurological disorders, mental retardation, or psychotic disorders were excluded from the sample. The University Intramural Ethical Committee approved all procedures and protocols, and written informed consent was obtained before enrollment. All the participants were blind to the aims of the study. 2.2. Assessment and measures Diagnostic assessment was made by experienced clinical psychiatrists using the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-CV) [20] and the Schedule for Interviewing DSM-IV Personality Disorders-IV (SIDP-IV) [21]. The diagnostic composition of the OBPSY group was as follows: eating disorders of the bulimic spectrum (binge eating disorder or bulimia nervosa), 41%; major depressive disorder, 29%; anxiety disorders, 24%; bipolar disorder, 5%; cluster B personality disorders, 1%. The occurrence of traumatic life events between 0 and 15 years of age was assessed by interviewing participants with an interview (Early Traumatic Life Events, ETLE) developed by Bandelow et al. [22] to measure exposure to a variety of adverse early experiences including: (1) separation from the natural mother, due to death of mother, long absence of the mother due to illness (>100 days), adoption, upbringing in a foster home, upbringing by other family kin or unrelated persons, or long absence due to divorce or separation of parents; (2) separation from the natural father, due to death of the father, long absence of the father due to illness (>100 days), adoption, upbringing in a foster home, upbringing by other family members or unrelated persons, long absence of the father due to war service or war imprisonment, absence due to separation or divorce of parents, or absence due to imprisonment; (3) separation from parents due to illness of the proband (>100 days); (4) severe physical handicap of the subject during childhood; (5) severe physical handicap of sibling; (6) severe parents' marital problems; (7) alcohol addiction of one or both parents; (8) severe psychiatric illness of mother or father (other than alcohol dependence); (9) violence in the family, including physical abuse of the subject; (10) sexual molestation or abuse. We analyzed the ETLE data in two different ways. First, we calculated the prevalence rates of traumatic life events in the three groups of participants by using a dichotomous approach (i.e., the presence/absence of at least one of the traumatic life events listed by the ETLE scale). Then, to determine whether a combination of multiple early traumatic life events was associated with obesity, we calculated a dimensional ETLE score on a 0- to 10-point scale, with each of the traumatic life events receiving one point [22]. To measure adult attachment style, we used the Italian version [23] of the Relationship Questionnaire (RQ) [24]. The RQ is a single item measure made up of four short paragraphs, each describing a prototypical attachment pattern as it applies in close adult peer relationships. Participants are asked to rate their degree of correspondence to each prototype on a 7-point scale. The RQ was designed to obtain continuous ratings of each of the four attachment patterns (i.e., secure, preoccupied, fearful, and dismissing). The four attachment patterns are defined in terms of two dimensions: anxiety (i.e., a strong need for care and attention from attachment figures coupled with a pervasive uncertainty about the willingness of attachment figures to respond to such needs) and avoidance (i.e., discomfort with psychological intimacy and the desire to maintain psychological independence). The anxious attachment score is computed by summing the preoccupied and fearful scales. We used the RQ because it is relatively brief, has been implemented in multiple studies, and because it is the only measure, among popular measures of attachment, to demonstrate independence from self-deceptive biases [25]. In general, reliability estimates for the RQ continuous ratings (r's of about 0.50) are comparable to those of brief self-report measures [26]. 2.3. Statistical analysis Binary logistic regression was used to identify the significant predictors of adult obesity. The method of regression was Forward Conditional. Effects are reported as odds ratios (exp(β)) with 95% confidence intervals (CI) and P-values against the hypothesis of no association. Analysis was performed on a personal computer using SPSS for Windows, version 16.0 (SPSS, Chicago, IL). 3. Results Table 1 reports descriptive data for the three groups of participants. The prevalence rates calculated as the occurrence of at least one of the events listed by the ETLE scale were 24% in the CON group, 38.5% in the OB group, and 43.5% in the OBPSY group. These differences did not reach statistical significance (chi-square = 5.24, df = 2, P = 0.07). Table 2 reports the occurrence of the different types of traumatic life events during the first 15 years of life in the three groups of participants. Based on their self-reports, very few (8%) of the 545 A. D'Argenio et al. / Physiology & Behavior 98 (2009) 543–546 Table 1 Descriptive data (mean ± SD) for the three groups of participants. Age (years) Gender (M:F) BMI Early trauma (no. of participants with at least one traumatic event) Early trauma (dimensional score) Anxious attachment Table 3 Predictors of adult obesity: results of binary logistic regression. CON OB OBPSY Variable B (SE) 32.58 ± 11.21 26: 24 23.38 ± 2.85 12 (24%) 40.38 ± 12.06 23: 42 41.33 ± 6.80 25 (38.5%) 39.12 ± 11.52 21: 64 38.27 ± 6.69 37 (43.5%) Early trauma 0.50 0.44 0.44 0.48 0.21 0.30 ± 0.65 4.82 ± 2.85 1.02 ± 1.74 5.35 ± 2.87 0.92 ± 1.35 7.94 ± 2.87 Anxious attachment obese participants experienced physical and/or sexual abuse. Among obese participants, the traumatic events with the highest prevalences were separation from one or both parents and severe parents' marital problems. To ascertain whether adult obesity was related to early trauma, we conducted a binary logistic regression with obesity as the outcome variable (coded as: non-obese = 1, obese = 2) and the severity of early trauma (as measured by the ETLE score) as the predictor variable. Gender (coded as: women = 1, men = 2) and age were entered into the model as independent variables to control for their possible confounding effects on obesity (Table 3). Overall, the model correctly classified 78% of participants. Severity of early trauma emerged as a significant predictor of adult obesity. We repeated the analysis by excluding those participants who had been physically and/or sexually abused to ascertain whether these types of severe maltreatments were instrumental in determining the significant relationship between early trauma and obesity. The results did not change and early trauma remained a significant predictor of adult obesity even though the relationship between the two variables became weaker. Overall, the model correctly classified 75.5% of participants. Subsequent analyses aimed at ascertaining whether early trauma contributed significantly to the prediction of adult obesity when the influence of psychiatric diagnosis and anxious attachment was taken into account. To assess the impact of psychiatric diagnosis, we repeated the logistic regression on a sub-sample (N = 115) that did not include obese participants with a psychiatric diagnosis. If psychiatric diagnosis was instrumental in determining the significant relationship between early trauma and obesity, then such a relationship should disappear in a sample not including participants with a psychiatric diagnosis. This was not the case. Early trauma remained a significant predictor of adult obesity and the model correctly classified 72.2% of participants. Whereas the inclusion in the control group of non-obese participants required, by definition, the absence of a psychiatric diagnosis, a high level of self-reported attachment anxiety was not an exclusion criterion. This allowed us to use the entire sample to conduct the logistic regression with the scale measuring anxious attachment as a predictor. Anxious attachment made a substantial contribution to the prediction of obesity. However, early trauma remained a significant and independent predictor of obesity. Overall, the model correctly classified 77.5% of participants. Table 2 Prevalence (no. of participants and percentage) of different types of self-reported traumatic events during the first 15 years of life in the three groups of participants. Separation from the mother Separation from the father Separation due to subject's illness Physical handicap of the subject Physical handicap of sibling Parents' marital conflict Parental alcohol addiction Parental psychiatric illness Physical abuse Sexual abuse CON OB OBPSY 0 (0%) 5 (10%) 0 (0%) 1 (2%) 2 (4%) 1 (2%) 0 (0%) 1 (2%) 0 (0%) 0 (0%) 6 (9.23%) 12 (18.46%) 0 (0%) 0 (0%) 2 (3.08%) 7 (10.78%) 2 (3.08%) 1 (1.54%) 5 (7.69%) 2 (3.08%) 10 (11.76%) 24 (28.23%) 0 (0%) 1 (1.18%) 2 (2.35%) 13 (15.29%) 1 (1.18%) 3 (3.53%) 1 (1.18%) 4 (4.70%) (0.21)⁎ (0.22)⁎⁎ (0.21)⁎⁎⁎ (0.22)⁎⁎⁎⁎ (0.07)⁎⁎⁎⁎ P exp(β) 95% CI for exp(β) 0.021 0.050 0.038 0.034 0.002 1.65 1.55 1.56 1.61 1.23 1.08–2.52 1.00–2.40 1.03–2.36 1.04–2.50 1.08–1.41 ⁎ Entire sample (N = 200); Nagelkerke's R2: 0.20; model: χ2 = 29.64, df = 3, P = 0.000. ⁎⁎ Sub-sample of participants who did not report physical and/or sexual abuse (N = 188); Nagelkerke's R2: 0.18; model: χ2 = 24.14, df = 3, P = 0.000. ⁎⁎⁎ Sub-sample of participants without a psychiatric diagnosis (N = 115); Nagelkerke's R2: 0.20; model: χ2 = 18.62, df = 2, P = 0.000. ⁎⁎⁎⁎ Entire sample (N = 200) when anxious attachment was included into the model; Nagelkerke's R2: 0.27; model: χ2 = 40.20, df = 4, P = 0.000. 4. Discussion Results from the present study confirm that exposure to early-life trauma is associated with an increased risk of adult obesity. Specifically, participants who scored higher on a scale measuring the severity of traumatic events experienced during the first 15 years of their lives were more likely to be obese at the time of testing. These associations were observed in a sample of obese participants with and without psychiatric disorders compared with a control group of non-obese healthy volunteers after controlling for the effects of age and gender. Our findings are in line with those of a number of previous studies showing that exposure to traumatic events during childhood is associated with an elevated risk of adult obesity [4–8]. An original contribution of the present study is the demonstration that not only physical or sexual abuse but also less severe early-life stressors are associated with an increased risk of adult obesity. Based on their selfreports, few (8%) of the obese participants experienced physical and/or sexual abuse whereas most of them (56%) experienced some other types of apparently less severe early trauma such as separation from one or both parents or marital conflict between parents. The exclusion of the participants who experienced physical and/or sexual abuse did not change the results of statistical analysis. With the notable exception of the study by Thomas et al. [9], this is the first study to show that not only sexual or physical abuse but also less severe forms of early-life stress are linked to the development of obesity later in life. Despite the consistent findings of the studies that have analyzed the association between early trauma and adult obesity, the mechanism for the increased risk of obesity following developmental trauma is unknown. A recent review suggested that disordered eating behavior and/or psychological reactions to a traumatic experience may account for the added risk of adult obesity [10]. In partial accord with such a hypothesis, we found that, of the 85 obese participants with a current diagnosis of psychiatric disorder, 41% had an eating disorder of the bulimic spectrum (binge eating disorder or bulimia nervosa) and 53% had an affective disorder (depression or anxiety). These are the psychological conditions that more often lead to overeating either directly (as in the case of binge eating disorder and bulimia) or indirectly (as in the case of emotional eating associated with depression and anxiety) [27,28]. However, the results of the present study show that the presence of a diagnosable psychiatric disorder is not a necessary condition to explain the association between early trauma and adult obesity. In fact, when we excluded obese participants with a psychiatric diagnosis from the sample, early trauma remained a significant predictor of adult obesity. In line with this finding, Gunstad et al. [29] found a significant association between exposure to early-life stressors and adult obesity in a large sample (N = 696) of persons without psychiatric disorders. Psychiatric diagnosis is not the only manifestation of psychological dysfunction. Subclinical psychological symptoms and dysfunctional personality traits may be present in individuals who do not reach the 546 A. D'Argenio et al. / Physiology & Behavior 98 (2009) 543–546 diagnostic threshold for a psychiatric condition. Based on this notion, we analyzed the role of anxious attachment as a potential mediator of the relationship between early trauma and adult obesity. In accord with our hypothesis, we found a strong association between anxious attachment and obesity. This finding was expected because insecure attachment is a frequent (although not an invariable) outcome of early adverse experiences and because disordered eating behavior is often associated with anxious attachment [15–17]. However, in the binary logistic model including anxious attachment among the independent variables, early trauma remained a significant and independent predictor of adult obesity. When considered in combination with the nonsignificant impact of psychiatric diagnosis, this finding suggests that psychological dysfunction is not the only mechanism mediating the elevated risk of obesity in persons exposed to early-life trauma. If psychological dysfunction does not fully explain the association between early trauma and adult obesity, which alternative mechanisms might be involved? Preliminary data point to epigenetic mechanisms and long-term physiological alterations affecting food intake and energy balance [30,31]. This hypothesis is in line with the results of recent studies conducted on animal models [reviewed by 32]. In male mice, neonatal separation from the mother leads to the development of adult obesity, with metabolic and hormonal alterations that are similar to those found in experimental models of diabetes mellitus [33]. In human subjects, there is considerable evidence that early-life stress leads to chronic hyperactivity of the hypothalamic–pituitary–adrenal–cortical (HPA) axis [34,35] which in turn can cause accumulation of depot fat in visceral adipose tissues [36]. From an evolutionary perspective, the finding that early trauma can direct gene expression affecting food intake and energy balance is interpreted as an adaptive mechanism for transmitting a message concerning the quality of the future environment the individual will have to live in [37]. The results of this study should be evaluated in the context of several methodological limitations. Our findings pertain to extremely obese bariatric surgery candidates evaluated at a psychiatric center. The findings may not be generalizable to obese patients who seek different (nonsurgical) forms of treatment or to non-treatment-seeking community populations. We also note the well-known potential limitations inherent in self-report and retrospective assessments. In this regard, it is worth noting that the self-reported prevalence of physical and sexual abuse among the obese participants of this study was much lower than that found by previous surveys [4,6]. The interview method used in this study may have led to a possible underreporting of physical and sexual abuse [38]. Finally, the cross-sectional nature of the data also precludes any speculation about causality. It is possible that obesity in childhood and adolescence would, per se, influence the risk of being exposed to specific stressors, thereby leading to spurious associations with adult obesity. Prospective studies are needed to clarify whether exposure to early-life stressors increases the risk of adult obesity or whether obese children are at higher risk for exposure to these stressors. In conclusion, the results of the present study show that: (1) not only sexual or physical abuse but also less severe forms of early-life stress are linked to the development of obesity later in life, and (2) psychological dysfunction is not the only mechanism mediating the elevated risk of obesity in persons exposed to early-life trauma. References [1] Gluckman PD, Hanson MA, Cooper C, Thornburg KL. Effect of in utero and early-life conditions on adult health and disease. N Engl J Med 2008;359:61–73. [2] Adair LS. Child and adolescent obesity: epidemiology and developmental perspectives. Physiol Behav 2008;94:8–16. [3] Cottrell EC, Ozanne SE. Early life programming of obesity and metabolic disease. Physiol Behav 2008;94:17–28. [4] Aaron DJ, Hughes TL. Association of childhood sexual abuse with obesity in a community sample of lesbians. Obesity 2007;15:1023–8. [5] Alvarez J, Pavao J, Baumrind N, Kimerling R. The relationship between child abuse and adult obesity among California women. Am J Prev Med 2007;33:28–33. [6] Grilo CM, Masheb RM, Brody M, Toth C, Burke-Martindale CH, Rothschild BS. Childhood maltreatment in extremely obese male and female bariatric surgery candidates. Obes Res 2005;13:123–30. [7] Noll JG, Zeller MH, Trickett PK, Putnam FW. Obesity risk for female victims of childhood sexual abuse: a prospective study. Pediatrics 2007;120:e61–7. [8] Williamson DF, Thompson TJ, Anda RF, Dietz WH, Felitti V. Body weight and obesity in adults and self-reported abuse in childhood. Int J Obes Relat Metab Disord 2002;26: 1075–82. [9] Thomas C, Hyppönen E, Power C. Obesity and type 2 diabetes risk in midadult life: the role of childhood adversity. Pediatrics 2008;121:e1240–9. [10] Gustafson TB, Sarwer DB. Childhood sexual abuse and obesity. Obes Rev 2004;5: 129–35. [11] Anda RF, Felitti VJ, Bremner JD, Walker JD, Whitfield C, Perry BD, et al. The enduring effects of abuse and related adverse experiences in childhood. A convergence of evidence from neurobiology and epidemiology. Eur Arch Psychiatry Clin Neurosci 2006;256:174–86. [12] Waters E, Merrik S, Treboux D, Crowell J, Albersheim L. Attachment security in infancy and early adulthood: a twenty-year longitudinal study. Child Dev 2000;71:684–9. [13] Maunder RG, Hunter JJ. Attachment relationships as determinants of physical health. J Am Acad Psychoanal Dyn Psychiatry 2008;36:11–32. [14] Tasca GA, Szadkowski L, Illing V, Trinneer A, Grenon R, Demidenko N, et al. Adult attachment, depression, and eating disorder symptoms: the mediating role of affect regulation strategies. Pers Individ Differ 2009;47:662–7. [15] Troisi A, Di Lorenzo G, Alcini S, Nanni RC, Di Pasquale C, Siracusano A. Body dissatisfaction in women with eating disorders: relationship to early separation anxiety and insecure attachment. Psychosom Med 2006;68:449–53. [16] Troisi A, Massaroni P, Cuzzolaro M. Early separation anxiety and adult attachment style in women with eating disorders. Br J Clin Psychol 2005;44:89–97. [17] Ward A, Ramsay R, Turnbull S, Benedettini M, Treasure J. Attachment patterns in eating disorders: past in the present. Int J Eat Disord 2000;28:370–6. [18] Donfrancesco C, Lo Noce C, Brignoli O, Riccardi G, Ciccarelli P, Dima F, et al. Italian network for obesity and cardiovascular disease surveillance: a pilot project. BMC Fam Pract 2008;29(9):53. [19] Marcus MD, Kalarchian MA, Courcoulas AP. Psychiatric evaluation and follow-up of bariatric surgery patients. Am J Psychiatr 2009;166:285–91. [20] First MB, Spitzer RL, Gibbon M, Williams JBW. Structured clinical interview for DSM-IV axis I disorders — clinician version (SCID-CV). Washington, DC: American Psychiatric Press; 1997. [21] Pfohl B, Blum N, Zimmerman M. Structured interview for DSM-IV personality disorders. Washington, DC: American Psychiatric Press; 1997. [22] Bandelow B, Krause J, Wedekind D, Broocks A, Hajak G, Rüther E. Early traumatic life events, parental attitudes, family history, and birth risk factors in patients with borderline personality disorder and healthy controls. Psychiatry Res 2005;134:169–79. [23] Troisi A, D'Argenio A, Peracchio F, Petti P. Insecure attachment and alexithymia in young men with mood symptoms. J Nerv Ment Dis 2001;189:311–6. [24] Bartholomew K, Horowitz LM. Attachment style among young adults: a test of a four-category model. J Pers Soc Psychol 1991;61:226–44. [25] Leak GK, Parsons CJ. The susceptibility of three attachment style measures to socially desirable responding. Soc Behav Personal 2001;29:21–30. [26] Scharfe E, Bartholomew K. Reliability and stability of adult attachment patterns. Pers Relationships 1994;1:23–43. [27] Dallman MF, Pecoraro N, Akana SF, La Fleur SE, Gomez F, Houshyar H, et al, Manalo S. Chronic stress and obesity: a new view of “comfort food”. Proc Natl Acad Sci USA 2003;100:11696–701. [28] Goossens L, Braet C, Van Vlierberghe L, Mels S. Loss of control over eating in overweight youngsters: the role of anxiety, depression and emotional eating. Eur Eat Disord Rev 2009;17:68–78. [29] Gunstad J, Paul RH, Spitznagel MB, Cohen RA, Williams LM, Kohn M, et al. Exposure to early life trauma is associated with adult obesity. Psychiatry Res 2006;142:31–7. [30] Sinclair KD, Lea RG, Rees WD, Young LE. The developmental origins of health and disease: current theories and epigenetic mechanisms. Soc Reprod Fertil Suppl 2007;64: 425–43. [31] Stöger R. The thrifty epigenotype: an acquired and heritable predisposition for obesity and diabetes? Bioessays 2008;30:156–66. [32] Coccurello R, D'Amato FR, Moles A. Chronic social stress, hedonism and vulnerability to obesity: lessons from rodents. Neurosci Biobehav Rev 2009;33:537–50. [33] Loizzo A, Loizzo S, Galietta G, Caiola S, Spampinato S, Campana G, et al. Overweight and metabolic and hormonal parameter disruption are induced in adult male mice by manipulations during lactation period. Pediatr Res 2006;59:111–5. [34] Nemeroff CB. Early-life adversity, CRF dysregulation, and vulnerability to mood and anxiety disorders. Psychopharmacol Bull 2004;38(Suppl 1):14–20. [35] Teicher MH, Andersen SL, Polcari A, Anderson CM, Navalta CP, Kim DM. The neurobiological consequences of early stress and childhood maltreatment. Neurosci Biobehav Rev 2003;27:33–44. [36] Björntorp P. Do stress reactions cause abdominal obesity and comorbidities? Obes Rev 2001;2:73–86. [37] Junien C, Nathanielsz P. Report on the IASO stock conference 2006: early and lifelong environmental epigenomic programming of metabolic syndrome, obesity and type II diabetes. Obes Rev 2007;8:487–502. [38] Dill DL, Chu JA, Grob MC, Eisen SV. The reliability of abuse history reports: a comparison of two inquiry formats. Compr Psychiatry 1991;32:166–9.