6.1 Introduction

This chapter applies multivariate logistic regression models to assess the independent effects of workplace information and policies about alcohol and drug use, access to workplace employee assistance programs, and drug testing programs on workers= current illicit drug use and current heavy alcohol use. Independent effects are the effects of workplace characteristics after statistically controlling for other factors that could affect workers= drug use. Such other important factors to be controlled in this analysis are worker characteristics of age, gender, race/ethnicity, education level, and personal income, as well as establishment size of the workplace and the census region where the employee lives and probably works. In addition, patterns of relationship are examined for differences between 1994 and 1997.

Four major research questions motivate the multivariate analyses in this chapter:

1. Do sociodemographic characteristics of workers or attributes of workplaces impact outcomes while other demographic characteristics are held constant? For example, does establishment size of the workplace have significant effect on workers= substance abuse behavior even after workers= sociodemographic characteristics are statistically controlled?

2. Do the workplace drug information, policy, and Employee Assistance Programs (EAPs) maintain significant effects on workers= substance abuse after other important characteristics of workers and workplace drug-use test programs are controlled statistically?

3. Do various workplace drug use testing programs influence workers= self-reported substance abuse behavior after other important sociodemographic characteristics of workers and workplace drug information, policy, and EAPs are controlled statistically?

4. Do the characteristics of workers and workplaces show similar effects on workers= current illicit drug use and current heavy alcohol use?

We present the Aodds ratio (OR)³@, or the Arelative likelihood@ to depict the effect of the independent factors on the binary outcome variables in this chapter. The OR is the number by which we would multiply the odds of being a current illicit drug user (the probability, divided by one minus the probability), or a current heavy alcohol user, for each one-unit increase in the independent variable. As a multiplicative coefficient, an OR with a value greater than 1 suggests a Apositive@ effect, and a value between 0 and 1 implies a Anegative@ effect⁴.

4 Magnitudes of positive and negative effects can be compared by taking the inverse of the negative effect (or vice versa). For example, the magnitude of effect indicated by a positive OR of 2 is the same as the magnitude of the effect indicated by a negative OR of .5 (=2).

SAMHSA, an agency in the Department of Health and Human Services, is the Federal Government's lead agency for improving the quality and availability of substance abuse prevention, addiction treatment, and mental health services in the United States.