DRUG AND/OR ALCOHOL ABUSE OUTCOME IN MENTAL ILLNESS:THE MEDIATING ROLE OF SOCIAL SUPPORT

Abstract

ABSTRACT Introduction: The present study examined whether social support served as a mediator (i.e., an apparent causal mechanism) for the relationship between mental illness (MI) and drug and/or alcohol abuse (SA). Objective: This study’s objective was to determine the role of social support as a potential mediator in the relationship between mental illness and drug and/or alcohol abuse. Methods: We utilized data from a randomized controlled trial (RCT), conducted on individuals at risk for HIV in Philadelphia jails by following 600 study participants coming out of jails. Out of these 600 individuals, data was selected for individuals with mental illness and drug and/or alcohol abuse. In the parent study, these individuals were identified by asking questions about their mental illness, and drug and/or alcohol abuse problems. They also answered Norbeck Social Support Questionnaire about social support. They provided demographic data on their age, race, gender, religion, marital status and education level, which was collected as part of a face-to-face demographic interview conducted during the baseline assessment. Results: Data was analyzed using negative binomial regression method to test for mediation effect. Results indicated that social support mediated the relationship between mental illness and drug abuse. We ran two generalized linear and one general linear regression models. In the first model, we looked at the total effect of mental illness on drug abuse (c); we found that the incidence rate for drug abuse (SA) would be expected to increase by a factor of 1.387, (IRR,1.387; CI,1.270-1.515; P =0.000) for every one-unit increase in mental illness (M1), while holding all other variables in the model constant. In the second model we adjusted our mediator, social support (SS) and looked at the direct effect of mental illness on drug abuse (c’). We interpreted that for every one-unit increase in mental illness, the incidence rate for drug abuse would be expected to increase by a factor of 2.717, (IRR,2.717; 95% CI,1.629 - 4.532; P = .000). For every one-unit increase in social support, the incidence rate for drug abuse is expected to decrease by a factor of .498, (IRR,.498; 95% CI,.443 - .560; P = .000). In the third model we found, for every one-unit increase in mental illness, there is an associated increase by 2.495 units in social support, P= .000. There was an inconsistent mediation in our model. Effect size for mediation by Percent mediation (Pm) method was found to be 0.64, it is the proportion of the effect that is mediated by our mediator social support. Sobel test showed the significance of mediation with a test statistic of 4.8282 at a significance level of 0.000. Conclusion: Our data supported an alternative theory of inconsistent mediation. We found that social support mediates the relationship between mental illness and drug abuse, where positive social support has a stimulator effect on mental illness and a suppressor effect on drug abuse. Mental illness may have direct unfavorable effects on the drug abuse (outcome) and positive social network has beneficial effects on this outcome. We conclude that positive support allows betterment of mental health of patients and prevents involvement in drug abuse. Further, there is a need to consider both the positive and negative effects of social support while keeping in mind these associations may differ among sociodemographic groups.