Antisocial behavioral syndromes and DSM-IV drug use disorders in the United States: Results from the National Epidemiologic Survey on Alcohol and Related Conditions

Abstract

Background

Antisocial behavioral syndromes, including antisocial personality disorder (ASPD), syndromal adult antisocial behavior (AABS) without conduct disorder (CD) before age 15, and CD without progression to ASPD (“CD only”) are highly comorbid with drug use disorders (DUDs). Among patients in DUD treatment, antisocial syndromes are associated with greater severity and poorer outcomes. Comparative data concerning associations of antisocial syndromes with clinical characteristics of DUDs among general population adults have not previously been available. This study describes associations of antisocial syndromes with clinical characteristics of lifetime Diagnostic and Statistical Manual—Version IV DUDs in the general U.S. adult population.

Methods

This report is based on the 2001–2002 National Epidemiologic Survey on Alcohol and Related Conditions (n = 43,093, response rate = 81%). Respondents (n = 4068) with lifetime DUDs were classified according to whether they met criteria for ASPD, AABS, “CD only,” or no antisocial syndrome. Associations of antisocial syndromes with clinical characteristics of DUDs were examined using logistic regression.

Results

Antisocial syndromes were significantly associated with the phenomenology of DUDs, particularly ASPD with the most severe clinical presentations. Associations with AABS were similar to those with ASPD; those with “CD only” were weak, inconsistent, and not statistically significant. Patterns of associations differed little between men and women.

Conclusions

Both ASPD and AABS, but not “CD only,” appear to identify greater clinical severity of DUDs among adults in the general U.S. population.

Introduction

Antisocial personality disorder (ASPD) is highly comorbid with drug use disorders (DUDs) among adults in both DUD treatment settings and the general population. In the Epidemiologic Catchment Area survey, the prevalence of ASPD according to Diagnostic and Statistical Manual—Version III (DSM-III; American Psychiatric Association, 1980) criteria among adults in the U.S. general population with any lifetime DSM-III DUD was 17.2% (Anthony and Helzer, 1991). Comparable data have not been published from other major epidemiologic surveys conducted in the U.S. However, among National Comorbidity Survey respondents with lifetime cannabis dependence, the prevalence of ASPD was 21.4% (Agosti et al., 2002). Prevalences in DUD treatment samples range from 8.9% to 68.0%, depending in part upon the diagnostic criteria and assessment instruments utilized to identify the disorder (Broome et al., 1999, Brooner et al., 1992, Cacciola et al., 1996, Compton et al., 2000a, Grella et al., 2003, Hasin et al., 2006, Kidorf et al., 2004, King et al., 2001, Ladd and Petry, 2003, McKay et al., 2000, Messina et al., 2002, Ross et al., 1988, Westermeyer and Thuras, 2005).

Patients in drug treatment settings with ASPD demonstrate more severe DUDs than those without ASPD, including more extensive drug use histories and greater impairment across multiple domains of functioning at intake (Cacciola et al., 1996, Darke et al., 2004, Ladd and Petry, 2003, McKay et al., 2000, Westermeyer and Thuras, 2005). ASPD has been associated with poorer treatment outcomes, including worse psychosocial functioning, more legal problems, and more problematic substance use in many (Basu et al., 2004, Cacciola et al., 1996, Compton et al., 2003, Fridell et al., 2006, Grella et al., 2003, King et al., 2001, Leal et al., 1994) but not all (Brooner et al., 1998, Crits-Cristoph et al., 1999, McKay et al., 2000, Messina et al., 2002, Messina et al., 2003) studies. Among patients with DUDs, comorbid ASPD is associated with higher levels of drug and sexual HIV risk behaviors (Compton et al., 1995, Disney et al., 2006, Fals-Stewart et al., 2003, Kelley and Petry, 2000, Ladd and Petry, 2003, Woody et al., 1997). However, ASPD has been an inconsistent predictor of response to HIV risk reduction interventions (Compton et al., 1998, Compton et al., 2000b, McCusker et al., 1995, Woody et al., 2003).

DSM-III, Diagnostic and Statistical Manual—Version III-Revised (DSM-III-R; American Psychiatric Association, 1987), and Diagnostic and Statistical Manual—Version IV (DSM-IV; American Psychiatric Association, 2000) criteria for ASPD require both syndromal levels of antisocial behavior since age 15, and evidence of conduct disorder (CD) with onset before age 15. However, it has now been well-documented in both clinical (Black and Braun, 1998, Brooner et al., 1992, Cacciola et al., 1994, Cacciola et al., 1995, Cottler et al., 1995, Goldstein et al., 1998, Goldstein et al., 2001) and epidemiologic (Compton et al., 2005, Galbaud du Fort et al., 2002, Marmorstein, 2006, Tweed et al., 1994) samples that individuals with syndromal levels of antisocial behavior in adulthood frequently do not report symptomatic behaviors sufficient to meet criteria for CD with onset before age 15 (AABS, not a codable disorder in DSM-IV). Regardless of ascertainment source, individuals with AABS differ little from those with ASPD on adult antisocial behavior and psychiatric comorbidity (Black and Braun, 1998, Cottler et al., 1995, Goldstein et al., 1998, Langbehn and Cadoret, 2001, Tweed et al., 1994). In addictions treatment settings, clients with AABS also differ little from clients with ASPD on lifetime drug and alcohol histories and substance-related problems at intake (Cecero et al., 1999, Cacciola et al., 1995, Goldstein et al., 1998). Further, the limited available evidence suggests that treatment outcomes may not differ importantly between clients with DUDs who have comorbid ASPD and those with AABS (Cacciola et al., 1995, Cecero et al., 1999, Goldstein et al., 2001).

These findings raise increasing concern that categorizing individuals with AABS as “without ASPD” may obscure important differences in clinical characteristics and treatment outcomes of DUDs, between individuals with ASPD and individuals with no adult antisocial syndrome, and between those with ASPD and those with no lifetime history of antisociality (Black and Braun, 1998, Cottler et al., 1995). Further, the lack of a dedicated DSM diagnostic category for AABS leaves affected individuals’ antisociality to be recorded in clinical settings under the V-code of “adult antisocial behavior” (Langbehn and Cadoret, 2001, Tweed et al., 1994) or the diagnosis of “Personality Disorder Not Otherwise Specified” (Black and Braun, 1998). Both options raise nosologic concerns. The former may downplay the severity of the antisociality by implying that symptomatic behavior is not maladaptive, or characterized by the inflexibility, persistence, and impairment or distress that would define ASPD, while both may obscure the need for clinical attention to behaviors that would meet criteria for ASPD but for the requirement of CD with onset before age 15 (Black and Braun, 1998, Marmorstein, 2006). Recent research has raised similar issues regarding the diagnosis of attention-deficit/hyperactivity disorder (ADHD; Faraone et al., 2006a, Faraone et al., 2006b). Strong similarity in clinical and neuropsychological characteristics, as well as patterns of familial transmission, between “full” ADHD, in which all DSM-IV criteria are met including onset before age 7 years, and “late-onset” ADHD, in which all criteria are met except for age at onset, suggests that the age-at-onset requirement in ADHD may be too stringent, at least for diagnosing adults (Faraone et al., 2006a, Faraone et al., 2006b).

CD is a well-documented risk factor for substance use disorders (Marmorstein and Iacono, 2005, Nock et al., 2006, Robins and McEvoy, 1990, Robins and Price, 1991, Sung et al., 2004). However, associations of CD that does not progress to ASPD (hereinafter, “CD only”) with clinical characteristics and outcomes of DUDs have received little attention in adults. Nock et al. (2006) showed DSM-IV CD to be associated with persistence of substance use disorders (alcohol and drug combined) from the lifetime to the 12-month time frame, but found less elevation in risk of persistence with “remitted” than with “active” CD.

Comparative data concerning associations of DSM-IV ASPD, “CD only,” and AABS with clinical characteristics of DUDs have not previously been available from adult general population samples. There are also no comparative data concerning the sociodemographic and clinical correlates of these syndromes among general population adults with DUDs. Accordingly, this report has 2 purposes. First, we examine the prevalences of ASPD, AABS, and “CD only,” among adults diagnosed with DUDs in Wave 1, conducted in 2001–2002, of the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC; Grant et al., 2003b, Grant et al., 2004c). We compare the sociodemographic and clinical characteristics, including family histories and psychiatric comorbidity, of U.S. adults in the general population with DUDs who are affected with each syndrome and with no antisocial syndrome. We then examine associations of ASPD, AABS, and “CD only” with clinical characteristics of DUDs, including ages at drug use milestones, treatment seeking, lifetime symptom counts, number of lifetime episodes, duration of longest or only episode of DUD, and patterns of drug consumption during respondents’ periods of heaviest lifetime drug use.

The NESARC is the first major psychiatric epidemiology survey to employ DSM-IV criteria. With a nationally representative sample of 43,093 respondents, including 4068 who met lifetime criteria for DUDs, the NESARC allows precise estimates of sociodemographic correlates, family histories of drug problems and antisociality, lifetime psychiatric comorbidity, and clinical characteristics of DUDs by antisocial syndrome. In addition, the large number of cases of DUDs both with and without antisocial syndromes allows examination of whether patterns of associations between antisocial syndromes and DUD phenomenology vary by sex.