Substance-related Consequences, Dependence and Substance Use Disorder: Promising new model taking clinical and statistical points of view into consideration

Friday, 25 November, 2022 - 13:20 to 14:50

Abstract

According to the conceptualization proposed by the DSM-IV and ICD-11, problems related to the use of psychoactive substances have two dimensions corresponding to (1) dependence and (2) the consequences that result from use. These two dimensions, although distinct, are closely related under a two-dimensional conceptual model. However, the release of the DSM-5 proposes a different design and merges them; Called substance use disorder (SUD), the two previous concepts are now grouped under a continuum. Would it be possible that a two-level bidimensional model where the two dimensions would be overseen by a supra-concept, since it could be a single phenomenon measured via two dimensions? Objective: This presentation will compare the empirical basis of the already existing conceptual model of the SUD with the new two-level bidimensional model

Setting: Research interviews in clinical settings and in the general population; Participants: 1006 Quebecers divided into three groups: users of alcohol (n = 893), cannabis (n = 411) and other drugs (n = 255); Measurements: Brief Dependence Scale (BDS-7) and Brief Consequences Scale (BCS-7); Analysis: Confirmatory factorial analyses compare a first-order model to a model where both dimensions are evaluated simultaneously under an umbrella concept (second-order model).

The adjustment indicators for the two models proposed are statistically satisfactory, although the two-level bidimensional model shows a better fit for all indicators.

Current studies confirm that a single diagnosis merging consequences and manifestations of dependence into a single disorder is a statistically sound combination. However, results illustrate that a single diagnosis based on two strongly correlated but conceptually distinct dimensions best represents the reality of SUD. The two-level bidimensional model is statistically robust and better adapted to clinical needs.

Speakers

Type

Tracks

Part of session