With regard to the availability of various drug information, policies, employee assistance program, and drug testing program at workplaces, this report relies on employees' own awareness of the existence of these policies or programs. Measurement error may emerge if employees' own subjective perceptions were not consistent with the actual existence of workplace policies and programs. For example, it is possible that some workers were not aware that their workplaces provided information, had written policies, provided access to EAPs, or had drug testing programs and thus reported that such were not provided despite facts to the contrary. However, the potential measurement error due to the worker's subjective evaluation may be attenuated because the local workplace, rather than the entire company, was targeted in the survey questionnaire and most employees are familiar with their local workplace settings. In addition, a written policy not recognized by employees might have no impact on the substance abuse behaviors. It was also unlikely for employees to incorrectly report that the workplace provided information, had written policies, provided access to EAPs, and had drug testing programs.

Although separate models are analyzed for workers' illicit drug use and heavy alcohol use in the multivariate analysis, the survey instruments regarding the workplace drug use policies and programs are on drug or alcohol use. Also, this analysis does not present information on factors such as job autonomy and job satisfaction that might be associated with workers' substance abuse behavior. The analysis is also constrained by the limited information available on the style and content of the drug-related policies inhibiting studying of the impact of variations in characteristics of the policies. In addition, without more detailed organizational information, it is not possible to identify whether the availability of drug-related information and policies suggests good faith effort and real commitment to drug-free workplaces.