Are the symptoms of cannabis use disorder best accounted for by dimensional, categorical, or factor mixture models? A comparison of male and female young adults

Dimensionality and differential functioning of DSM-5 cannabis use disorder criteria in an online sample of adults with frequent cannabis use

The DSM-5 criteria for cannabis use disorder (CUD) combine DSM-IV dependence and abuse criteria (without legal problems) and new withdrawal and craving criteria. Information on dimensionality, internal reliability, and differential functioning of the DSM-5 CUD criteria is lacking. Additionally, dimensionality of the DSM-5 withdrawal items is unknown. This study examined the psychometric properties of the DSM-5 CUD criteria among adults who used cannabis in the past 7 days (N = 5,119). Adults with frequent cannabis use were recruited from the US general population through social media and filled in a web-based survey about demographics and cannabis use behaviors. Factor analysis was used to assess dimensionality, and item response theory analysis models were used to explore relationships between the criteria and the underlying latent trait (CUD), and whether each criterion and the criteria set functioned differently by demographic and clinical characteristics: sex, age, state-level cannabis laws, reasons for cannabis use, and frequency of use. The DSM-5 CUD criteria showed unidimensionality and provided information about the CUD latent trait across the severity spectrum. The cannabis withdrawal items indicated one underlying latent factor. While some CUD criteria functioned differently in specific subgroups, the criteria set as a whole functioned similarly across subgroups. In this online sample of adults with frequent cannabis use, evidence supports the reliability, validity, and utility of the DSM-5 CUD diagnostic criteria set, which can be used for determining a major risk of cannabis use, i.e., CUD, to inform cannabis policies and public health messaging, and for developing intervention strategies.

The genetic aetiology of cannabis use: from twin models to genome-wide association studies and beyond

Cannabis is among the most widely consumed psychoactive substances worldwide. Individual differences in cannabis use phenotypes can partly be explained by genetic differences. Technical and methodological advances have increased our understanding of the genetic aetiology of cannabis use. This narrative review discusses the genetic literature on cannabis use, covering twin, linkage, and candidate-gene studies, and the more recent genome-wide association studies (GWASs), as well as the interplay between genetic and environmental factors. Not only do we focus on the insights that these methods have provided on the genetic aetiology of cannabis use, but also on how they have helped to clarify the relationship between cannabis use and co-occurring traits, such as the use of other substances and mental health disorders. Twin studies have shown that cannabis use is moderately heritable, with higher heritability estimates for more severe phases of use. Linkage and candidate-gene studies have been largely unsuccessful, while GWASs so far only explain a small portion of the heritability. Dozens of genetic variants predictive of cannabis use have been identified, located in genes such as CADM2, FOXP2, and CHRNA2. Studies that applied multivariate methods (twin models, genetic correlation analysis, polygenic score analysis, genomic structural equation modelling, Mendelian randomisation) indicate that there is considerable genetic overlap between cannabis use and other traits (especially other substances and externalising disorders) and some evidence for causal relationships (most convincingly for schizophrenia). We end our review by discussing implications of these findings and suggestions for future work.

Cannabis use in college: Genetic predispositions, peers, and activity participation

BACKGROUND

Among adult college students in the US, cannabis use is common and associated with considerable negative consequences to health, cognition, and academic functioning, underscoring the importance of identifying risk and protective factors. Cannabis use is influenced by genetic factors, but genetic risk is not determinative. Accordingly, it is critical to identify environments that reduce risk among those who are at elevated genetic risk. This study examined the impact of polygenic scores for cannabis initiation, various forms of social activity participation, and peer deviance on recent cannabis use. Our aim was to test whether these environments moderate genetic risk for cannabis use.

METHODS

Data came from a longitudinal sample of undergraduate college students of European American (EA; NEA = 750) and African American (AA; NAA = 405) ancestry. Generalized estimating equations with a logit link function were used to examine main effects and two-way interactions.

RESULTS

Engagement with church activities was associated with lower probability of cannabis use. Peer deviance was associated with higher probability of cannabis use. Engagement with community activities moderated the influence of the polygenic risk score in the EA sample, such that PRS was associated with recent cannabis use among those who never engaged in community activities. This effect did not replicate in AAs, which may have been due to the portability of PRS based on EA discovery samples.

CONCLUSIONS

Results suggest that community activities may limit the influence of genetic risk, as associations between PRS and cannabis use were only observed among individuals who never engaged in community activities.

Re-evaluation of the KMSK scales, rapid dimensional measures of self-exposure to specific drugs: Gender-specific features

BACKGROUND

The Kreek-McHugh-Schluger-Kellogg (KMSK) scales provide a rapid assessment of maximal self-exposure to specific drugs and can be used as a dimensional instrument. This study provides a re-evaluation of the KMSK scales for cannabis, alcohol, cocaine, and heroin in a relatively large multi-ethnic cohort, and also the first systematic comparison of gender-specific profiles of drug exposure with this scale.

METHODS

This was an observational study of n = 1,133 consecutively ascertained adult volunteers. The main instruments used were the SCID-I interview (DSM-IV criteria) and KMSK scales for cannabis, alcohol, cocaine, and heroin.

RESULTS

Participants were 852 volunteers (297 female) with specific DSM-IV abuse or dependence diagnoses, and 281 volunteers without any drug diagnoses (154 female). Receiver operating characteristic (ROC) curves were calculated for concurrent validity of KMSK scores with the respective DSM-IV dependence diagnoses. The areas under the ROC curves for men and women combined were 99.5% for heroin, 97% for cocaine, 93% for alcohol, and 85% for cannabis. Newly determined optimal KMSK "cutpoint" scores were identical for men and women for cocaine and heroin dependence diagnoses, but were higher in men than in women, for cannabis and alcohol dependence diagnoses.

CONCLUSIONS

This study confirms the scales' effectiveness in performing rapid dimensional analyses for cannabis, alcohol, cocaine, and heroin exposure, in a cohort larger than previously reported, with "cutpoints" changed from initial determinations, based on this larger sample. The KMSK scales also detected gender differences in self-exposure to alcohol and cannabis that are associated with the respective dependence diagnoses.

Associations between personality disorders and cannabis use and cannabis use disorder: a population-based twin study

BACKGROUND AND AIMS

Individual differences in DSM-IV personality disorders (PDs) are associated with increased prevalence of substance use disorders. Our aims were to determine which combination of PDs trait scores best predict cannabis use (CU) and cannabis use disorder (CUD), and to estimate the size and significance of genetic and environmental risks in PD traits shared with CU and CUD.

DESIGN

Linear mixed-effects models were used to identify PD traits for inclusion in twin analyses to explore the genetic and environmental associations between the traits and cannabis use.

SETTING

Cross-sectional data were obtained from Norwegian adult twins in a face-to-face interview in 1999-2004 as part of a population-based study of mental health.

PARTICIPANTS

Subjects were 1419 twins (μage  = 28.2 years, range = 19-36) from the Norwegian Institute of Public Health Twin Panel with complete PD and cannabis data.

MEASUREMENTS

PD traits were assessed using DSM-IV criteria. Life-time CU and CUD were based on DSM-IV abuse and dependence criteria, including withdrawal and craving.

FINDINGS

After adjusting for age and sex, antisocial [β = 0.23, 95% confidence interval (CI) = 0.19-0.28] and borderline PDs (β = 0.20, 95% CI = 0.14-0.26) were associated strongly with CU. Antisocial (β = 0.26, 95% CI = 0.21-0.31) and borderline PDs (β = 0.12, 95% CI = 0.06-0.18) were also linked strongly to CUD. Genetic risks in antisocial and borderline PD traits explained 32-60% of the total variance in CU and CUD. Dependent and avoidant PDs explained 11 and 16% of the total variance in CU and CUD, respectively.

CONCLUSIONS

Individual differences in the liability to cannabis use and cannabis use disorder appear to be linked to genetic risks correlated with antisocial and borderline personality disorder traits.

Insights from individuals successfully recovered from cannabis use disorder: natural versus treatment-assisted recoveries and abstinent versus moderation outcomes

BACKGROUND

Increasing understanding of the pathways and processes of recovery from cannabis use disorder may help in designing effective and attractive interventions to promote recovery. We report insights from individuals who had successfully recovered from cannabis use disorder with a variety of pathways. Recovered individuals describe their perceptions of why they developed the problem, why they were successful in recovering, and the advice they would offer to individuals with similar problems.

METHODS

Media announcements were used to recruit 119 volunteers who met lifetime but not past year criteria for cannabis use disorder. Participants were asked open-ended questions which were content analyzed and compared between individuals who whose recoveries were treatment-assisted (45%) versus natural (55%) and between individuals who were abstinent (57%) versus those who continued non-problematic consumption (43%).

RESULTS

Participants most frequently described their problems as having developed due to the use of cannabis to cope, because of environmental and social influences, and enjoyment of the positive effects. Success in recovery was attributed to focusing on reasons for change, goal commitment to change, and conquering denial/self-deception. Treatment-assisted participants were more likely to perceive that they overcame their cannabis problem due to treatment/self-help and conquering underlying issues, whereas naturally recovered participants were more likely to describe focusing on reasons for change, will power, and lost enjoyment/lifestyle change. Treatment-assisted participants were more likely to recommend seeking help/social support and naturally recovered participants were more likely to endorse reflecting on reasons for change, engaging in hobbies/distracting activities, and stimulus control/avoidance/change social environment. The majority recommended professional treatment (79.1%) and self-help materials (76.9%), and a little over half (53.2%) would also recommend natural recovery.

CONCLUSIONS

These insights from people with lived experience further support previous research that treatment-assisted and natural recoveries are for the most part similar with respect to the recovery process. However, participants, whether or not they had had treatment involvement, recommended the use of treatment and self-help materials to sharpen their focus on the reasons to change and to enhance their commitment to change. At the same time, they saw value in the efforts of individuals to recovery without help.

Characterizing severe and enduring anorexia nervosa: An empirical approach

Targeted approaches for the treatment of severe and enduring anorexia nervosa (SE-AN) have been recommended, but there is no consensus definition of SE-AN to inform research and clinical practice. This study aimed to take initial steps toward developing an empirically based definition of SE-AN by characterizing associations among putative indicators of severity and chronicity in eating disorders. Patients with AN (N = 355) completed interviews and questionnaires at treatment admission and discharge; height and weight were assessed to calculate body mass index (BMI). Structural equation mixture modeling was used to test whether associations among potential indicators of SE-AN (illness duration, treatment history, BMI, binge eating, purging, quality-of-life) formed distinct subgroups, a single group with one or more dimensions, or a combination of subgroups and dimensions. A three-factor (dimensional), two-profile (categorical) mixture model provided the best fit to the data. Factor 1 included eating disorder behaviors; Factor 2 comprised quality-of-life domains; Factor 3 was characterized by illness duration, number of hospitalizations, and admission BMI. Profiles differed on eating disorder behaviors and quality-of-life, but not on indicators of chronicity or BMI. Factor scores, but not profile membership, predicted outcome at discharge from treatment. Data suggest that patients with AN can be classified on the basis of eating disorder behaviors and quality-of-life, but there was no evidence for a chronic subgroup of AN. Rather, indices of chronicity varied dimensionally within each class. Given that current definitions of SE-AN rely on illness duration, these findings have implications for research and clinical practice.

Clearing the haze: the complexities and challenges of research on state marijuana laws

As states increasingly liberalize marijuana laws, high-quality research is needed that will inform the public and policymakers about the health and societal impact of these laws. However, there are many challenges to studying marijuana policy, including the heterogeneity of the drug and its use, the variability in the laws and their implementation from state to state, the need to capture a wide variety of relevant outcomes, and the poorly understood influence of marijuana commercialization. Furthermore, current instruments generally fail to distinguish between types of users and lack accurate and detailed measures of use. This review provides a background on marijuana laws in the United States and an overview of existing policy research, discusses methodological considerations when planning analysis of state marijuana laws, and highlights specific topics needing further development and investigation.

Item Response Theory Analysis to Assess Dimensionality of Substance Use Disorder Abuse and Dependence Symptoms

BACKGROUND

Item response theory (IRT) based studies conducted on diverse samples showed a single dominant factor for DSM-III-R and DSM-IV substance use disorder (SUD) abuse and dependence symptoms of alcohol, cannabis, sedative, cocaine, stimulants, and opiates use disorders. IRT provides the opportunity, within a person-centered framework, to accurately gauge each person's severity of disorder that, in turn, informs required intensiveness of treatment.

OBJECTIVES

The aim of this study was to determine whether the SUD symptoms indicate a unidimensional trait or instead need to be conceptualized and quantified as a multidimensional scale.

METHODS

The sample was composed of families of adult SUD+ men (n=349), and SUD+ women (n=173), who qualified for DSM-III-R diagnosis of substance use disorder (abuse or dependence) and families of adult men and women who did not qualify for a SUD diagnosis (SUD- men: n=190, SUD- women: n=133). An expanded version of the Structured Clinical Interview for DSM-III-R (SCID) was administered to characterize lifetime and current substance use disorders. Item response theory methodology was used to assess the dimensionality of DSM-III-R SUD abuse and dependence symptoms.

RESULTS

A bi-factor model provided the optimal representation of the factor structure of SUD symptoms in males and females. SUD symptoms are scalable as indicators of a single common factor, corresponding to general (non-drug-specific, common) liability to addiction, combined with drug-specific liabilities.

CONCLUSIONS

IRT methodology used to quantify the continuous general liability to addiction (GLA) latent trait in individuals having SUD symptoms was found effective for accurately measuring SUD severity in men and women. This may be helpful for person-centered medicine approaches to effectively address intensity of treatment.

Cannabis problem experiences among users of the tobacco-cannabis combination known as blunts

BACKGROUND

In most of the world, cannabis smokers mix loose tobacco inside a joint, pipe, spliff, or cone. More recently, a 'blunt' formulation combines these two drugs by inserting cannabis into a hollowed-out cigar. Epidemiological research linking simultaneous use of these two drugs and the development of cannabis use disorders (CUD) remains unclear. This study estimates associations linking blunt smoking with levels and subtypes of cannabis problems.

METHODS

Cross-sectional data on 27,767 past-year cannabis users were analyzed from the US National Survey on Drug Use and Health (NSDUH) conducted from 2009 to 2012. Ten self-reported items of DSM-IV CUD features elicited a single latent trait of cannabis problem (CP) severity, which was then regressed on past-year blunt smoking and past-month blunt frequency measures within the context of a conceptual model. Differential item functioning (DIF) analysis evaluated potential bias in CP feature response by blunt smoking history.

RESULTS

Past-year blunt smoking was associated with higher CP severity compared to cannabis users who did not smoke blunts. Days of blunt smoking in the past month also predicted higher CP severity than less frequent blunt use. Those smoking blunts experienced more subjectively felt tolerance and having spent more time obtaining or using cannabis, but were less likely to experience other problems, even at the same level of CP severity.

CONCLUSIONS

These findings suggest smoking blunts might promote the development of problematic cannabis use. Responses to cannabis problems differed by history of blunt smoking, possibly implicating an influence of tobacco on measurement of cannabis use disorders.

Comparing factor, class, and mixture models of cannabis initiation and DSM cannabis use disorder criteria, including craving, in the Brisbane longitudinal twin study

Accumulating evidence suggests that the Diagnostic and Statistical Manual of Mental Disorders (DSM) diagnostic criteria for cannabis abuse and dependence are best represented by a single underlying factor. However, it remains possible that models with additional factors, or latent class models or hybrid models, may better explain the data. Using structured interviews, 626 adult male and female twins provided complete data on symptoms of cannabis abuse and dependence, plus a craving criterion. We compared latent factor analysis, latent class analysis, and factor mixture modeling using normal theory marginal maximum likelihood for ordinal data. Our aim was to derive a parsimonious, best-fitting cannabis use disorder (CUD) phenotype based on DSM-IV criteria and determine whether DSM-5 craving loads onto a general factor. When compared with latent class and mixture models, factor models provided a better fit to the data. When conditioned on initiation and cannabis use, the association between criteria for abuse, dependence, withdrawal, and craving were best explained by two correlated latent factors for males and females: a general risk factor to CUD and a factor capturing the symptoms of social and occupational impairment as a consequence of frequent use. Secondary analyses revealed a modest increase in the prevalence of DSM-5 CUD compared with DSM-IV cannabis abuse or dependence. It is concluded that, in addition to a general factor with loadings on cannabis use and symptoms of abuse, dependence, withdrawal, and craving, a second clinically relevant factor defined by features of social and occupational impairment was also found for frequent cannabis use.

Towards the characterization and validation of alcohol use disorder subtypes: integrating consumption and symptom data

BACKGROUND

There is evidence that measures of alcohol consumption, dependence and abuse are valid indicators of qualitatively different subtypes of alcohol involvement yet also fall along a continuum. The present study attempts to resolve the extent to which variations in alcohol involvement reflect a difference in kind versus a difference in degree.

METHOD

Data were taken from the 2001-2002 National Epidemiologic Survey of Alcohol and Related Conditions. The sample (51% male; 72% white/non-Hispanic) included respondents reporting past 12-month drinking at both waves (wave 1: n = 33644; wave 2: n = 25186). We compared factor mixture models (FMMs), a hybrid of common factor analysis (FA) and latent class analysis (LCA), against FA and LCA models using past 12-month alcohol use disorder (AUD) criteria and five indicators of alcohol consumption reflecting frequency and heaviness of drinking.

RESULTS

Model comparison revealed that the best-fitting model at wave 1 was a one-factor four-class FMM, with classes primarily varying across dependence and consumption indices. The model was replicated using wave 2 data, and validated against AUD and dependence diagnoses. Class stability from waves 1 to 2 was moderate, with greatest agreement for the infrequent drinking class. Within-class associations in the underlying latent factor also revealed modest agreement over time.

CONCLUSIONS

There is evidence that alcohol involvement can be considered both categorical and continuous, with responses reduced to four patterns that quantitatively vary along a single dimension. Nosologists may consider hybrid approaches involving groups that vary in pattern of consumption and dependence symptomatology as well as variation of severity within group.