Developing a latent coping resources factor for recovery from substance use disorder

Abstract

Background

It is crucial for individuals with substance use disorders, especially those with a co-occurring mental health disorder, to access effective coping resources.

Aim: We quantify coping through four domains of individual resources (self-esteem, self-efficacy, perceived social support, and hope) to examine the extent to which individuals with varying psychiatric severity can access coping-related resources. Since sex is an additional, important consideration in treatment, we also explore both sexes’ access to coping-related resources.

Method: We generate a multilevel latent variable of coping resources in class structures, one for males and one for females, to measure (1) the extent individuals with varying psychiatric presentations (types, symptoms, severity) are able to access this latent resource and (2) to control for house level effects.

Results

Our variables of self-esteem, self-efficacy, perceived social support, and hope all coalesced into a latent variable, named coping resources. Furthermore, we find that psychiatric severity is negatively related to coping resources at the individual level but function differently for males and females at the house level.

Conclusions: Treatment guidelines should address the nuanced needs of individuals and consider individual differences, such as sex, that impact access to coping resources.

Keywords:

• Recovery homes
• substance use disorder
• coping resources
• psychiatric severity
• sex differences

Acknowledgments

The authors acknowledge the assistance from members of the Oxford House organization, and in particular Paul Molloy, Alex Snowden, Casey Longan, and Howard Wilkins.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Information about the data set associated with this paper can be found by contacting Dr. Leonard A. Jason at [email protected].

Notes

1 We could have dichotomized PSI scores to represent low and high PSI individuals, but this could produce confounds and Type II error (Royston et al. 2006).

Additional information

Funding

The present work was financially supported by the National Institute on Alcohol Abuse and Alcoholism [AA022763].