تجزیه و تحلیل کلاس پنهان خوردن و علائم رفتاری تکانشی در زنان تایوانی مبتلا به بولیمیا

Objective The implications of impulsivity in its relationship with binge-eating or purging behaviors remain unclear. This study examined the patterns of eating behaviors and co-morbid impulsive behaviors in individuals with bulimia nervosa n optimally homogeneous classes using latent class analysis (LCA). Methods All participants (n=180) were asked to complete a series of self-reported inventories of impulsive behaviors and other psychological measures. Information regarding the lifetime presence of symptoms of eating disorder was assessed by clinical interviews. LCA was conducted using eating disorder symptoms, impulsive behaviors, and the number of purging methods. Results Three latent classes of bulimic women were identified. These were women who exhibited relatively higher rates of purging, symptoms of impulsive behavior, and multiple purging methods (17.8%), women who used no more than one purging method with a low occurrence of impulsive behavior (41.7%), and women who showed higher rates of purging behaviors and the use of multiple purging methods with a low rate of impulsive behavior (41.7%). The impulsive sub-group had comparable severity of eating-related measures, frequency of binge-eating, and higher levels of general psychopathology than that of the other two sub-groups. Conclusion This study provides empirical support for the existence of an impulsive subgroup with distinctive features among a non-Western group of BN patients. This study also suggests that mechanisms other than impulse dysregulation may exist for the development of binge-eating and purging behaviors in bulimia nervosa patients, or the mechanisms contributing to binge-eating and impulsive behaviors may be different.

In addition to their binge-eating and purging behaviors, women with bulimia nervosa (BN) often have various kinds and intensities of impulsive behavior [1]. The prevalence of impulsive behaviors or impulse control disorders is high in women with BN, with rates of 3 to 50% for substance use and/or dependence [2], 10 to 50% for alcohol use disorders [2] and [3], 15 to 30% for self-injurious behavior or suicide attempts [4], 20% for sexual promiscuity [5], 28 to 47% for shoplifting, and 18% for compulsive buying [6]. Patients with multi-impulsivity, as defined by Fichter et al. as the presence of at least three impulsive behaviors, have been postulated to be a distinct subgroup of BN patients [5] and [7]. Although correlation studies have demonstrated that measures of impulsivity are positively correlated with the severity of disordered eating [8], [9], [10] and [11], some studies have failed to show that patients with multi-impulsivity have more severe binge eating or other eating pathologies [5], [12], [13], [14] and [15]. The relationships between impulsivity and binge-eating or purging behaviors remain unclear. Bulimia nervosa has been classified as being of either the purging or non-purging type since the publication of the Diagnostic and Statistical Manual of Mental Disorders Fourth Edition (DSM-IV) in 1994 [16]; however, the use of purging or non-purging as compensatory behaviors to distinguish between subtypes in terms of clinical relevance, severity, or co-morbidity was recently questioned [17]. The characteristics of individuals with different compensatory behaviors, such as exercise, vomiting, laxative use, and diet pill use, have been studied in eating disorder (ED) patients [18], [19], [20], [21] and [22]. Notably, laxative abuse is most frequently associated with greater general psychopathology, self-injurious behavior, suicide attempts, history of depression and anxiety, borderline personality features, and greater eating symptomatology [22]. Patients with laxative abuse have a higher level of impulsivity than those with other disordered eating behaviors [14], [23] and [24] with a few exceptions [5] and [25]. Some published studies have suggested that the number of different types of purging behaviors provides important information about the severity of ED, self-injurious behavior, or general psychopathology [26], [27] and [28]. Moreover, one study indicated that the use of multiple purging methods instead of any specific purging method was associated with greater severity of disordered eating attitudes [29]. Both the type and the number of different purging methods should be considered in studying the relationships between ED symptoms and comorbid psychopathology. Several studies have identified subgroups of individuals resembling those with BN that are characterized by binge-eating and purging symptoms in both clinical and community populations by using latent class analysis (LCA) [30], [31], [32], [33], [34], [35], [36] and [37] or latent profile analysis (LPA) [38], [39], [40], [41], [42], [43], [44] and [45]. Latent class analysis seeks to identify the smallest number of identifiable latent classes (i.e., categories of a latent factor) that account for the association between three or more observed dichotomous and polychotomous variables [46] and [47]. It is an empirical method for examining the inter-relationships among observed categorical variables and characterizing the underlying set of mutually exclusive latent classes that account for the associations among the observed categorical variables. Conditioning on the latent classes, probabilities of observed variables are assumed independent of each other. Similarly, LPA is used to determine latent classes from observed continuous variables. Of those studies conducted with participants with bulimic symptoms, four have identified coherent groupings of patients on the basis of bulimic and purging symptoms [33], [34], [35] and [42], one on the basis of six impulsive behaviors [48], and three on the basis of psychological and interview measures of co-morbid psychopathology [37], [38] and [45]. Among them, only one study replicated the previous multi-impulsivity bulimic classification empirically [48]; nevertheless, the issue of how behavioral symptoms of eating and impulsivity co-exist within individuals was seldom addressed in these studies. Bulimia nervosa has been found to have a prevalence in Taiwan as comparable to that in Western countries [49], but literature regarding empirical data about BN among non-Western populations is still scarce. Little is known about the clinical manifestations and co-occurring impulsive behaviors among patients with BN in non-Western countries. Only two studies have investigated the multi-impulsivity that was found among Japanese BN patients [13] and [25]. In this study, we sought to determine whether an impulsive subgroup of BN patients could be identified through the use of LCA, and to assess the patterns of eating behaviors co-aggregating with impulsive behaviors among a group of Taiwanese women with BN. In addition, latent classes were compared to determine whether empirically derived classes showed specific associations with psychological measures of eating pathology and comorbid psychopathology.