افزایش قد در اختلال شخصیت ضد اجتماعی و انواع آن

Abstract Research suggests that those with antisocial tendencies are larger than controls, but studies have not assessed this association in antisocial personality disorder (APD) or its hypothesized sub-types (i.e. adolescence-limited, late-onset). Height, weight, body mass index, bulk, and psychosocial adversity were assessed in 44 controls, nine adolescent-limited antisocials, 21 APDs, and 13 late-onset antisocials from the community. Adult antisocial individuals, regardless of age of onset, were significantly taller and had greater body bulk than controls. Although groups tended to differ on weight, they did not differ on body mass index. In addition, APDs and adolescent-limited individuals reported greater psychosocial adversity than the other groups. Adversity did not account for height or bulk differences. Results suggest prior findings on height and bulk may apply to APD and support differentiating adolescent-limited and life-course persistent subgroups.

. Introduction A somewhat unexpected, but intriguing finding in recent research is that larger body size is linked to antisocial behavior. For example, increased height at age 8–10 years is associated with violence at age 16–18 years (Farrington, 1989), and greater height and body bulk at age 3, independent of stimulation-seeking behavior and fearlessness, are each associated with aggression at age 11 (Raine et al., 1998). In addition, body mass at age 12 is uniquely related to aggression at age 13 (Tremblay et al., 1998) and, conversely, conduct disorder at age 14 is associated with greater body mass at age 22 (Pine et al., 1997). Body size has been thought to relate to antisocial behavior because larger people are able to intimidate others successfully (Raine et al., 1998), or because increased size may reflect neurobiological processes involved in antisocial behavior (Pine et al., 1997). However, no study has explored whether increased body size and antisocial behavior are related in adulthood. In addition, Raine et al. (1998) did not control for environmental stressors, yet Farrington (1989) found that adverse family conditions (e.g. economic deprivation, family criminality, poor parenting) were also associated with delinquency. Thus, it remains possible that family adversity, and not body size, per se, is accounting for the body size–antisocial behavior relationship. Another important gap in this preliminary literature is that studies have not explored whether body size is differentially related to hypothesized subgroups of antisocials. Two such subgroups involve what Moffitt (1993) has termed adolescent-limited and life-course persistent antisocial behavior. Although antisocial behavior first emerges in childhood or early adolescence for both groups, the former group engages in delinquent acts typically seen in adolescence that decline by late adolescence/early adulthood, whereas the latter group engages in stable, more severe antisocial behavior across the lifespan. Moffitt argues that life-course persistent antisocials, but not their adolescent-limited counterparts, have a biological basis to their behavior. A third subgroup recently identified by Hamalainen and Pulkkinen (1996) consists of late-onset antisocial individuals who first engage in antisocial acts in late adolescence or early adulthood. This later time of onset is regarded as developmentally unusual because the behavior first emerges when the delinquent behavior of their non-APD peers is typically diminishing. Despite their lack of early delinquency, these late-onset individuals exhibit lower educational attainment, more impulsive, hostile personality traits, cross-generational alcoholism and substance dependence problems, and severe antisocial behavior similar to life-course persistent antisocials (Elkins et al., 1997). It has been speculated that genetic or biological factors may play a role in the development of this antisocial typology; however, no biologically based research appears to have been conducted with this subgroup. The current, cross-sectional study makes an initial attempt to address several gaps in the literature. First, no study to date has looked at body size in relation to a formal diagnosis of antisocial personality disorder (APD), which may reflect more chronic, severe antisocial tendencies than a general measure of antisocial behavior. Thus, adolescent and adult antisocial behavior was defined using DSM-IV diagnoses for APD and conduct disorder. Using these criteria, the current study then addressed the following questions: Do life-course persistent individuals (i.e. APD) and ‘late-onset antisocials’ (i.e. APD without early conduct disorder) show height, weight, body mass, and bulk differences compared to controls? Do adolescent-limited individuals differ from controls? Finally, do body size differences remain after controlling for early psychosocial adversity?

. Results 3.1. Data screening and analysis plan 3.1.1. Data transformation Log transformations corrected for significant positive skewness and kurtosis of the weight, BMI, and psychosocial adversity variables. Height and bulk were normally distributed. 3.1.2. Covariates SES and age did not correlate with any of the four body size measures, r (85–87)=0.02–0.21, P>0.05, nor did whites and non-whites differ on the body size measures, t (85)=0.64–1.49, P>0.05. Psychosocial adversity inversely correlated with socioeconomic status, r (85)=−0.26 and P=0.02. However, it did not correlate with age, r (87)=−0.11, P>0.05, or differ between whites and non-whites, t (85)=0.04, P>0.05. 3.1.3. Data analysis plan Group differences on height, weight, BMI, and bulk were tested separately using one-way analysis of variance (ANOVA). Significant main effects of group (α=0.05) were followed up by three independent sample t-tests comparing controls to each antisocial subgroup. Psychosocial adversity was tested using analysis of covariance (ANCOVA) with SES as a co-variate. A significant main effect for group was followed up with three separate ANCOVAs that compared controls to each antisocial subgroup. Degrees of freedom were adjusted to correct for heterogeneous variances when warranted. All planned, pairwise comparisons were evaluated at α=0.017. Effect sizes adjusted for sample size (d, Cohen, 1998) were also calculated for all pairwise comparisons. Means and S.D.s for each dependent variable are presented in Table 2. Table 2. Descriptive information for body size measures and psychosocial adversity according to antisocial group Group Life-course persistent Late-onset Adolescent-limited Controls (n=21) (n=13) (n=9) (n=44) M S.D. M S.D. M S.D. M S.D. Body size Height 182.74 7.03 185.90 8.40 179.77 5.24 177.74 6.82 Weight 84.13 8.48 87.73 15.67 84.95 13.64 78.82 17.11 BMI 0.0025 0.0003 0.0025 0.0004 0.0026 0.0003 0.0025 0.0005 Bulk 104.74 12.63 111.65 18.13 101.46 14.62 94.70 15.67 Adversity 2.20 3.65 −0.28 3.25 1.77 3.27 −1.29 3.12 Note: height in cm and weight in kg; BMI=body mass index; and adversity=psychosocial adversity. Table options 3.2. Main analyses 3.2.1. Height Groups significantly differed on height, F3,83=5.60, P=0.002. Post hoc t-tests indicated that both the late-onset and life-course persistent antisocial individuals were taller than the controls, t (55)=3.59 and t (63)=2.73, P=0.001 and 0.008, respectively. Effect sizes were large for the late-onsets (d=1.12) and moderate-to-large for the life-course persistents (d=0.72). Adolescent-limited individuals did not differ from controls on height; t (51)=0.84, P=0.40, d=0.30. 3.2.2. Weight Although groups tended to differ on weight in the expected direction, the result did not reach statistical significance, F3,83=2.20, P=0.09. Effect sizes comparing the antisocial subgroups to controls were in the moderate range, d=0.57, 0.59, 0.43 for life-course persistents, late-onsets, and adolescent-limited, respectively. 3.2.3. BMI Groups did not differ on BMI, F3,83=0.41, P=0.75, with effect sizes generally in the small range, d=0.19, 0.13, and 0.33 for life-course persistents, late-onsets, and adolescent-limited, respectively. 3.2.4. Bulk Groups significantly differed on bulk, F3,83=4.96, P=0.003. Similar to the findings for height, both late-onsets and life-course persistents had larger bulk than controls, t (55)=3.31 and t (63)=2.56, P=0.002 and 0.01, respectively. Adolescent-limited individuals did not differ from controls on bulk, t (51)=1.19, P=0.24. Effect sizes were large for late-onsets (d=1.03), and moderate for life-course persistents (d=0.67) and adolescent-limited (d=0.43). 3.2.5. Psychosocial adversity After controlling for SES, the groups were found to differ on the amount of psychosocial adversity experienced as a child, F3,80=7.81, P<0.001. Co-variate adjusted, post-hoc analyses showed that both adolescent-limited and life-course persistent individuals had experienced greater psychosocial adversity than controls, F1,49=7.15, P=0.01, d=0.96 and F1,61=20.54, P<0.001, d=1.19, respectively. Adolescent-limiteds did not differ from controls, F1,52=0.85, P=0.36, d=0.30. In an exploratory analysis, late-onset antisocials were also found to have experienced less adversity than life-course persistent antisocials, F1,30=6.73, P=0.015, d=0.91. 3.2.6. Psychosocial adversity as a mediator/moderator To explore whether psychosocial factors mediated the relationship between height/bulk and antisocial typology, the adversity (and SES) measures were included as co-variates. The main effect of group remained significant for height and bulk, F3,79=4.88 and 4.56, P=0.004 and 0.005, respectively, indicating that body size directly related to antisocial subgroups and the findings were not attributable to adversity. In addition, adversity (dichotomized based upon a median split) was tested as a possible moderator of the antisocial group–body size relationship. The interaction term was not significant in either analysis, F3,76=1.58 and 0.33, P=0.20 and 0.80, respectively, indicating that the group differences in height did not depend on degree of adversity. 3.3. Exploratory analyses 3.3.1. Dropping the adolescent-limited group Due to the possibility that the small sample of adolescent-limited individuals may have skewed the results, the main analyses were re-run with this group excluded. The pattern of results remained the same for height, F2,75=7.99, P=0.001; weight, F2,75=3.02, P=0.06; BMI, F2,75=0.22, P=0.80; bulk, F2,75=7.35, P=0.001; and psychosocial adversity, F2,72=10.28, P<0.001. 3.3.2. Other psychiatric diagnostic groupings Participants were grouped according to whether or not they had a history of other psychiatric disorders (i.e. mood, anxiety, psychotic; cluster A, cluster B excluding APD, or cluster C personality disorders) and compared on height and bulk. None of the t-tests were significant for height, t (85)=0.05–1.65, P=0.96–0.10, or for bulk, t (85)=0.01–1.64, P=0.99–0.10, suggesting that the body size findings were specific to antisocial behavior.