The latent structure of illicit opioid dependence and comorbid psychiatric disorders

Download files
Access & Terms of Use
open access
Copyright: Shand, Fiona Laurel
Altmetric
Abstract
Opioid dependence is strongly associated with multiple substance dependence diagnoses and psychiatric disorders. Given that these diagnoses are used for clinical and research decision making, it is important in the lead up to DSM-V that both the diagnoses themselves and the relationships between them are based on empirically tested models. However, because large studies of opioid dependent people are rare, it has been difficult to examine the structure of the opioid dependence diagnosis and the underlying relationships between co-occurring disorders amongst this population. This thesis aims to identify the latent structure of opioid dependence and of comorbid DSM-IV diagnoses, and the associations of patterns of diagnoses with other clinically relevant outcomes: suicide attempts, non-fatal overdose, criminality, and treatment history. This was possible through studying a large sample of opioid dependent individuals (n=1,511) recruited from opioid replacement therapy clinics in urban, regional and rural centres in the greater Sydney area. Study 1 describes the key clinical features of the sample and sex differences in these: the lifetime prevalence of other substance dependence and mental illness was extremely high, as was the prevalence of personality disorders, multiple non-fatal opioid overdoses and suicide attempts, and imprisonment. Females were at greater risk for most internalising disorders whereas males were more likely to be diagnosed with externalising disorders. Study 2 brings together the characteristics identified in study 1 by examining the latent structure of the opioid dependence diagnosis and its relationship to other disorders and clinical variables. Several models were compared: a one factor and two factor model, latent class models, and factor mixture models. A 2 class, 1 factor model provided the best fit. Membership of the more severely dependent class was associated with higher risk for non-fatal overdose, other substance dependence, borderline personality disorder (BPD), more frequent opioid use, and antisocial personality disorder. Studies 3 and 4 examine the latent structure of externalising disorders, internalising disorders and a BPD screener. Study 3 confirmed that the comorbidity of externalising disorders is best conceptualised as a single severity spectrum. Greater externalising severity was also associated with BPD, criminal activity, internalising disorders, suicide attempts, and non-fatal overdose. In study 4, internalising disorders also formed a single severity spectrum. BPD was associated with both the externalising and internalising dimensions. The final study uses structural equation modelling to examine the relationships between the internalising and externalising spectrum, and four other important variables: suicide attempts, non-fatal opioid overdose, criminal activity, and treatment history. This thesis uses statistical techniques that have never been applied to an opioid dependent sample, and rarely to more prevalent disorders, in order to answer questions about opioid dependence and its associated problems in a more empirically valid way. In the case of the opioid dependence diagnosis, factor mixture modelling produced a different finding to previous studies using more limited statistical techniques. The model identified has some similarities to the proposed DSM-V substance use disorder diagnosis. This thesis also adds to a growing body of evidence that the comorbidity amongst 11 disorders can be explained by two dimensions. Finally, using structural equation modelling in the final study made it possible to test the strength of the relationships between these two dimensions – externalising and internalising – and a number of important psychosocial outcomes.
Persistent link to this record
Link to Publisher Version
Link to Open Access Version
Additional Link
Author(s)
Shand, Fiona Laurel
Supervisor(s)
Degenhardt, Louisa
Slade, Timothy
Creator(s)
Editor(s)
Translator(s)
Curator(s)
Designer(s)
Arranger(s)
Composer(s)
Recordist(s)
Conference Proceedings Editor(s)
Other Contributor(s)
Corporate/Industry Contributor(s)
Publication Year
2010
Resource Type
Thesis
Degree Type
PhD Doctorate
UNSW Faculty
Files
download whole.pdf 1.63 MB Adobe Portable Document Format
Related dataset(s)