Acknowledgments

This report was prepared by the Division of Population Surveys, Office of Applied Studies, SAMHSA, and by RTI International, a trade name of Research Triangle Institute, Research Triangle Park, North Carolina. Work by RTI was performed under Contract No. 283-03-9028. Sharon Larson at SAMHSA was responsible for the content, analysis, and writing of the report. At RTI, Joe Eyerman coauthored the report and was the task leader for its production. Misty Foster at RTI coauthored the report and was lead statistician, and Mary Ellen Marsden reviewed the report. Joseph C. Gfroerer at SAMHSA coauthored and reviewed the report. The Division of Workplace Programs, Center for Substance Abuse Prevention, SAMHSA, provided suggestions on the content and reviewed the final draft of the report. Other contributors at RTI listed alphabetically include Walter Boyle, Andrew Clarke, Teresa Davis, David Heller, Erica Hirsch, and Lisa Packer. At RTI, Claudia Clark edited the report; Diane G. Eckard prepared the graphics; Brenda K. Porter formatted the tables; Joyce Clay-Brooks formatted and word processed the report; and Pamela Couch Prevatt, Teresa G. Bass, Cassandra Carter, and Shari B. Lambert prepared its press and Web versions. Final report production was provided by Beatrice Rouse, Coleen Sanderson, and Jane Feldmann at SAMHSA.

Public Domain Notice

All material appearing in this report is in the public domain and may be reproduced or copied without permission from the Substance Abuse and Mental Health Services Administration. However, this publication may not be reproduced or distributed for a fee without specific, written authorization of the Office of Communications, SAMHSA, U.S. Department of Health and Human Services. Citation of the source is appreciated. Suggested citation:

Larson, S. L., Eyerman, J., Foster, M. S., & Gfroerer, J. C. (2007). Worker Substance Use and Workplace Policies and Programs (DHHS Publication No. SMA 07-4273, Analytic Series A-29). Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies.

Obtaining Copies of Publications from SAMHSA's Office of Applied Studies

Web Access:  http://samhsa.gov/data/

Originating Office:  SAMHSA, Office of Applied Studies

June 2007

 

Worker Substance Use and Workplace Policies and Programs

Highlights

This report presents findings on substance use among workers and on workplace drug policy and programs from the 2002, 2003, and 2004 National Surveys on Drug Use and Health (NSDUHs). NSDUH is an annual survey of the civilian, noninstitutionalized population of the United States aged 12 years or older. The survey is sponsored by the Substance Abuse and Mental Health Administration (SAMHSA) to provide data on substance use and related issues among the U.S. population.

NSDUH collects information on employment status, type of business, specific occupations and industries, and information on drug-testing policies and programs from U.S. workers. This report analyzes the worker information in conjunction with the substance use data collected in the survey to investigate substance use among full-time employed U.S. workers aged 18 to 64 during the period 2002 to 2004.

Substance Use among Workers, by Demographic and Geographic Characteristics

• The prevalences of substance use behaviors and substance use disorders are higher among unemployed persons than among full-time workers, part-time workers, and those with other employment status. However, because full-time workers constitute about two thirds of the population aged 18 to 64 (or 114.7 million persons), most substance users and most of those with substance use disorders are employed full time.
• The prevalence of past month illicit drug use among full-time workers aged 18 to 64 was estimated to be 8.2 percent in 2002, 2003, and 2004. Nearly one out of five (19.0 percent) workers aged 18 to 25 used illicit drugs during the past month. This was a higher percentage than among the 26-to-34 (10.3 percent), 35-to-49 (7.0 percent), and 50-to-64 (2.6 percent) age groups.
• Males were more likely than females to report past month illicit drug use (9.7 vs. 6.2 percent). Males accounted for about two thirds (6.4 million) of the workers who reported past month illicit drug use.
• Workers with a college education had a lower prevalence of past month illicit drug use compared with those without a college education. The prevalence of past month use of illicit drugs was lower among those with higher levels of education than those with less education (college graduate [5.7 percent] vs. less than high school [11.2 percent]).
• The prevalence of past month illicit drug use was lower among workers with higher family incomes than among workers with lower family incomes. An estimated 13.2 percent of workers who reported family income that was less than $20,000 had used illicit drugs during the past month. In contrast, 6.0 percent of workers who reported income in the highest category––$75,000 or more––had used illicit drugs during the past month.
• An estimated 8.8 percent, or 10.1 million, of full-time workers reported past month heavy alcohol use. Among younger workers (18 to 25 years old), 16.3 percent reported past month heavy alcohol use compared with 10.4 percent of 26- to 34-year-olds, 8.1 percent of 35- to 49-year-olds, and 4.7 percent of 50- to 64-year-olds.

Substance Use among Workers, by Occupation, Industry, and Establishment Size

• Of the major occupational groups, food service workers (17.4 percent) and construction workers (15.1 percent) exhibited a higher prevalence of past month illicit drug use than other occupational groups. Those working in education, training, and library occupations (4.1 percent), community and social services occupations (4.0 percent), and protective service occupations (3.4 percent) had the lowest prevalence of past month illicit drug use among the major occupational groups.
• The major occupational groups with the highest prevalence of past month heavy alcohol use were construction and extraction occupations (17.8 percent) and installation, maintenance, and repair occupations (14.7 percent). Community and social services occupations (2.8 percent) had the lowest prevalence of past month heavy alcohol use of the major occupations.
• The major industry groups with the highest prevalence of past month illicit drug use were accommodations and food services (16.9 percent) and construction (13.7 percent). Public administration (4.1 percent), educational services (4.0 percent), and utilities (3.8 percent) had the lowest prevalence of past month illicit drug use.
• The industry groups with the highest prevalence of past month heavy alcohol use were construction (15.9 percent); arts, entertainment, and recreation (13.6 percent); and mining (13.3 percent) industries. However, health care and social assistance (4.3 percent) and educational services (4.0 percent) had the lowest prevalence of past month heavy alcohol use compared with the other major industries.
• Prevalence of past month illicit drug use was lower as establishment size increased. The prevalence among workers in establishments with 25 to 99 employees was 8.2 percent, compared with 6.7 percent among workers in establishments with 100 to 499 employees and 5.7 percent among workers in establishments with 500 or more employees. A similar pattern was found for past month heavy alcohol use.

Workplace Information Policies and Programs concerning Drug and Alcohol Use

• Among the nearly 115 million full-time workers aged 18 to 64 years in the United States, 47.7 million (43.8 percent) reported that they had access to educational information about drug and alcohol use in the workplace, 87.0 million (78.7 percent) reported that they were aware of a written policy about drug and alcohol use in the workplace, and 60.9 million (58.4 percent) reported that their employer offered an employee assistance program (EAP).
• The youngest adult workers were least likely to report access to educational information about drug and alcohol use in the workplace. Among 18- to 25-year-old workers, 33.2 percent reported that they had educational information available. This was significantly lower than among workers aged 26 to 34 years (39.6 percent), 35 to 49 years (46.3 percent), and 50 to 64 years (48.9 percent). Young adult workers between the ages of 18 and 25 were significantly less likely to report EAPs available in the workplace compared with all other age groups (39.7 vs. 56.4 to 62.6 percent).
• Nearly 3 million (32.1 percent) full-time workers between the ages of 18 and 64 who had used an illicit drug in the past month reported that they worked for an employer who offered educational information about alcohol and drug use. An EAP was reported available to 3.9 million (45.4 percent) workers who were past month users of an illicit drug, while 6.5 million (71.0 percent) reported working for employers who had a written policy about drug and alcohol use.
• Generally, past month illicit drug users were less likely to report working for employers who offered workplace drug or alcohol programs or policies, compared with those who did not use an illicit drug in the past month. An estimated 45.4 percent of past month illicit drug users reported that there was an EAP at their place of employment compared with 59.6 percent of workers who had not used an illicit drug in the past month.

Workplace Testing

• Among the Nation's full-time workers, 42.9 percent reported that tests for illicit drug or alcohol use occurred at their place of employment during the hiring process, or "prehire" testing. This equates to more than 47 million adults who worked in settings where testing for illicit drug or alcohol use occurred during the hiring process.
• The youngest and oldest adult workers (18 to 25 and 50 to 64 years) were less likely to report working for employers with prehire drug testing than workers aged 26 to 49 years. An estimated 40.8 percent of 50- to 64-year-olds and 39.0 percent of 18- to 25-year-olds reported prehire testing, compared with 44.3 percent of workers aged 26 to 34 years and 44.7 percent of workers aged 35 to 49 years.
• For each age group, past month illicit drug users were less likely than nonusers to report working for employers who conducted prehire drug or alcohol tests (29.4 vs. 41.3 percent of 18- to 25-year-olds, 32.0 vs. 45.8 percent of 26- to 34-year-olds, 34.2 vs. 45.5 percent of 35- to 49-year-olds, and 31.3 vs. 41.0 percent of 50- to 64-year-olds).
• A total of 29.6 percent, or 32 million, of full-time workers in the United States reported random drug testing in their current employment setting during the study period. The youngest workers (18 to 25 years) were less likely than all other age groups to report working for an employer who conducted random drug testing (27.3 vs. 29.6 percent of 26- to 34-year-olds, 30.6 percent of 35- to 49-year-olds, and 29.1 percent of 50- to 64-year-olds).
• Past month illicit drug users were less likely to report working for employers who conducted random drug or alcohol tests than were nondrug users. For 18- to 25-year-olds, 19.7 percent of illicit drug users reported that they worked in a random-testing environment compared with 29.1 percent of nonusers. The relationship was consistent for all age groups: 20.0 versus 30.8 percent of 26- to 34-year-olds, 22.6 versus 31.2 percent of 35- to 49-year-olds, and 20.4 versus 29.3 percent of 50- to 64-year-olds.

Workplace Behaviors and Attitudes toward Drug Testing

• Among full-time workers who reported past month illicit drug use, 12.3 percent reported working for three or more employers in the past year, compared with 5.1 percent of workers without past month drug use. They also were more likely to report missing 2 or more workdays in the past month due to illness or injury when compared with workers without past month use (16.4 vs. 11.0 percent). Finally, 16.3 percent of workers who used illicit drugs in the past month reported skipping 1 or more days of work in the past month (vs. 8.2 percent of workers who did not use an illicit drug during the past month).
• Among full-time workers in the United States, 52.5 million (46.0 percent) workers indicated that they would be more likely to work for an employer who tests before hiring, and an additional 56.2 million (49.1 percent) workers reported that prehire testing would not influence their decision to work for an employer. Only 5.6 million (4.9 percent) workers indicated that they would be less likely to work for an employer who conducts prehire drug testing.
• More than half of U.S. workers reported that it would make no difference to them if an employer tests employees randomly after hire for drug or alcohol use. An estimated 45.5 million (39.8 percent) workers reported that they would be more likely to work for such an employer, while 10.0 million (8.7 percent) workers reported that they would be less likely to work for an employer who tests randomly for drug or alcohol use. An estimated 58.8 million (51.4 percent) workers indicated that random testing would not influence their decision to work for an employer.
• An estimated 29.1 percent of workers with past month illicit drug use reported that they would be less likely to work for employers who conduct drug testing randomly, while only 6.9 percent of workers who did not report past month illicit drug use selected this response category. This relationship was consistent in the multivariate models while controlling for age, gender, race/ethnicity, educational attainment, family income, region, and county type (metropolitan statistical area).

Worker Substance Use and Workplace Policies and Programs

1. Introduction

This report presents findings on substance use among workers and workplace drug policy and programs from the 2002, 2003, and 2004 National Surveys on Drug Use and Health (NSDUHs). NSDUH is an annual survey of the civilian, noninstitutionalized population of the United States aged 12 years or older. It is the primary source of statistical information on the use of illegal drugs by the U.S. population. The purpose of this report is to describe the nature of illicit drug and alcohol use in the adult working population and the prevalence of workplace programs designed to reduce drug and alcohol use. In addition, this report provides an assessment of the association of these programs with the prevalence of worker substance use.

In 1994, the Substance Abuse and Mental Health Services Administration (SAMHSA), in cooperation with the U.S. Department of Labor and the U.S. Small Business Administration, developed and implemented a module in NSDUH designed to gather specific information on employment status, type of business, and specific occupations and industries among persons aged 15 or older. In addition, this module also was designed to collect worker reports of drug-testing policies and programs. SAMHSA continues to include a workplace module in NSDUH. The workplace data collected in this module can be used in conjunction with drug use data collected in the survey to investigate drug use among U.S. workers.

Previous NSDUH reports related to employment status and workplace drug policies have focused on workers aged 18 to 49 years old (Hoffmann, Brittingham, & Larison, 1996; Hoffmann, Larison, & Sanderson, 1997; Office of Applied Studies [OAS], 1999; Zhang, Huang, & Brittingham, 1999). The current report expands the age range to include full-time workers aged 18 to 64 years old. In 1994, 44 percent of full-time workers aged 18 to 49 in the United States reported a drug- and/or alcohol-testing program in their place of employment. This increased to 49 percent by 1997. Although there was an increase in drug testing in the workplace, current illicit drug use by full-time workers remained essentially unchanged, with a rate of 7.6 percent reporting drug use in 1994 and 7.7 percent reporting use in 1997 (Zhang et al., 1999).

The analysis presented in this report demonstrates that worker substance use is a serious problem, with an estimated 9.4 million full-time workers aged 18 to 64 reporting illicit drug use in the past month. About 43.8 percent of full-time workers reported access to educational information about drug and alcohol use through work, 58.4 percent reported access to an employee assistance program (EAP), and 78.7 percent reported access to a written workplace policy about drug and alcohol use. In general, past month illicit drug users were less likely to work for employers who provided these programs. Finally, testing programs were fairly prevalent, with 48.8 percent of full-time workers reporting that their employer conducted testing for drug use. Multivariate analysis suggests that illicit drug users are less likely to work for employers who have a drug-testing program.

1.1. Summary of NSDUH

NSDUH is the primary source of statistical information on the use of illegal drugs by the U.S. civilian, noninstitutionalized population aged 12 or older. Conducted by the Federal Government since 1971, the survey collects data by administering questionnaires to a representative sample of the population through face-to-face interviews at their places of residence. The survey, which has been repeated annually since 1990, is sponsored by the Substance Abuse and Mental Health Services Administration (SAMHSA) of the U.S. Department of Health and Human Services, and is planned and managed by SAMHSA's Office of Applied Studies (OAS). Data collection is conducted under contract with RTI International, Research Triangle Park, North Carolina.¹ This section briefly describes the survey methodology; a more complete description is provided in Appendices A and B.

Prior to 2002, the survey was called the National Household Survey on Drug Abuse (NHSDA). Because of improvements to the survey in 2002, the 2002 data constitute a new baseline for tracking trends in substance use and other measures. For this reason, findings in this report are based on data only from the 2002, 2003, and 2004 NSDUHs. Estimates from these 3 survey years should not be compared with estimates from the 2001 or earlier versions of the survey. A discussion of survey methodology and results from the 2002 NSDUH are presented in OAS (2003). A more detailed discussion of the impact of changes in NSDUH methods on the survey results can be found in Appendix C in the 2004 national findings report (OAS, 2005).

NSDUH collects information from residents of households, noninstitutional group quarters (e.g., shelters, rooming houses, dormitories), and civilians living on military bases. The survey does not include homeless persons who do not use shelters, military personnel on active duty, and residents of institutional group quarters, such as jails and hospitals.

Since 1999, the NSDUH interview has been carried out using computer-assisted interviewing (CAI). Most of the questions are administered with audio computer-assisted self-interviewing (ACASI). ACASI is designed to provide the respondent with a highly private and confidential means of responding to questions to increase the level of honest reporting of illicit drug use and other sensitive behaviors. Less sensitive items are administered by interviewers using computer-assisted personal interviewing (CAPI). Overall, approximately 61 percent of the time that respondents spend answering questions is for items administered by ACASI.

The 2002, 2003, and 2004 NSDUHs employed a 50-State sample design with an independent, multistage area probability sample for each of the 50 States and the District of Columbia. The eight States with the largest population (California, Florida, Illinois, Michigan, New York, Ohio, Pennsylvania, and Texas), which together account for 48 percent of the total U.S. population aged 12 or older, were designated as large sample States. For these States, the design provided an annual sample sufficient to support direct State estimates. For the remaining 42 States and the District of Columbia, smaller, but adequate, samples were selected to support State estimates using small area estimation (SAE) techniques. The NSDUH design also oversampled youths and young adults, so that each State's sample was approximately equally distributed among three major age groups: 12 to 17 years, 18 to 25 years, and 26 years or older.

Each year's survey was conducted from January through December of that calendar year (e.g., January through December 2004 for the 2004 NSDUH). Sampled dwelling units were screened to identify eligible residents aged 12 or older. Up to two persons per dwelling unit were selected to be interviewed. In each year, respondents were given an incentive payment of $30 for completing the interview.

The weighted response rate for household screening was 90.8 percent between 2002 and 2004. The weighted response rate for interviewing among persons aged 18 to 64 was 77.9 percent. Sample sizes for persons aged 18 to 64 were 42,215 in 2002, 42,708 in 2003, and 43,053 in 2004, for a total of 127,976 completed interviews in this age group across the 3 years.

A number of key measures of substance use and dependence or abuse are reported from the NSDUH data. A complete listing and explanation of the key definitions used in NSDUH can be found in Appendix C. Occupational and industry classifications are provided in Appendix D. NSDUH defines "full-time" as respondents who usually work 35 or more hours per week and who worked in the past week or had a job despite not working in the past week. Illicit drugs include marijuana/hashish, cocaine (including crack), heroin, hallucinogens, inhalants, or prescription-type psychotherapeutics used nonmedically. "Heavy" alcohol use is defined as drinking five or more drinks on the same occasion on 5 or more days in the past 30 days. Analyses focus primarily on past month use, which also is referred to as "current use." NSDUH includes a series of questions to assess the prevalence of substance use disorders (i.e., dependence on or abuse of a substance) in the past year. These questions are used to classify persons as dependent or abusing specific substances based on criteria specified in the Diagnostic and Statistical Manual of Mental Disorders, 4^th edition (DSM-IV; American Psychiatric Association [APA], 1994). The questions on dependence ask about health and emotional problems associated with substance use, unsuccessful attempts to cut down on use, tolerance, withdrawal, reducing other activities to use substances, spending a lot of time engaging in activities related to substance use, or using the substance in greater quantities or for a longer time than intended. The questions on abuse ask about problems at work, home, and school; problems with family or friends; physical danger; and trouble with the law due to substance use. Dependence is considered to be a more severe substance use problem than abuse. Although individuals may meet the criteria specified for both dependence and abuse, persons are classified with abuse of a particular substance only if they are not classified as dependent on that substance. Measures of dependence on or abuse of alcohol and dependence on or abuse of illicit drugs are used in this report.

1.2. Format of Report and Explanation of Tables

Tables, text, and figures present prevalence measures for the population in terms of both the number of persons and the percentage of the population. Estimates presented in this report are based on averages for the 2002, 2003, and 2004 surveys. Combining data from these 3 survey years increases the sample size to support estimates among more detailed demographic and geographic domains, including the various occupational and industry characteristics described in this report.

Statistical tests have been conducted for all statements appearing in the text of the report that compare estimates between subgroups of the population. Unless explicitly stated that a difference is not statistically significant, all statements that describe differences are significant at the .05 level. Statistically significant differences are described using terms such as "higher" or "lower." Statements that use terms such as "similar," "no difference," or "same" to describe the relationship between estimates denote that a difference is not statistically significant. In addition, a set of estimates for population subgroups may be presented without a statement of comparison, in which case a statistically significant difference between these estimates is not implied and testing was not conducted.

All estimates presented in this report have met the criteria for statistical reliability (see Section B.2.2 of Appendix B). Estimates that do not meet these criteria are suppressed and do not appear in tables, figures, or text. Also, subgroups with suppressed estimates are not included in statistical tests of comparisons. For example, a statement that "whites had the highest prevalence" means that the rate among whites was higher than the rate among all racial/ethnic subgroups for which estimates were reliable, but not necessarily higher than the rate among a subgroup for which the estimate was suppressed.

Data are presented for racial/ethnic groups, based on current standards for collecting and reporting race and ethnicity data (Office of Management and Budget [OMB], 1997). Because respondents were allowed to choose more than one racial group, a "two or more races" category is presented that includes persons who reported more than one category among the seven basic groups listed in the survey question (white, black or African American, American Indian or Alaska Native, Native Hawaiian, Other Pacific Islander, Asian, Other). It should be noted that, except for the "Hispanic or Latino" group, the racial/ethnic groups discussed in this report include only non-Hispanics. The category "Hispanic or Latino" includes Hispanics of any race. Also, more detailed categories describing specific subgroups were obtained from survey respondents if they reported either Asian race or Hispanic ethnicity. Data on Native Hawaiians and Other Pacific Islanders are combined in this report.

Data also are presented for four U.S. geographic regions. These regions and divisions within these regions, defined by the U.S. Census Bureau, consist of the following groups of States:

Northeast Region - New England Division: Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, Vermont; Middle Atlantic Division: New Jersey, New York, Pennsylvania.

Midwest Region - East North Central Division: Illinois, Indiana, Michigan, Ohio, Wisconsin; West North Central Division: Iowa, Kansas, Minnesota, Missouri, Nebraska, North Dakota, South Dakota.

South Region - South Atlantic Division: Delaware, District of Columbia, Florida, Georgia, Maryland, North Carolina, South Carolina, Virginia, West Virginia; East South Central Division: Alabama, Kentucky, Mississippi, Tennessee; West South Central Division: Arkansas, Louisiana, Oklahoma, Texas.

West Region - Mountain Division: Arizona, Colorado, Idaho, Montana, Nevada, New Mexico, Utah, Wyoming; Pacific Division: Alaska, California, Hawaii, Oregon, Washington.

Geographic comparisons also are made based on urban influence county type, which reflects different levels of population size, urbanization, and access to larger communities based on county-level Urban Influence Codes (UIC) created by the Economic Research Service (ERS) of the U.S. Department of Agriculture. The codes group metropolitan and nonmetropolitan counties according to the official county-level metro status issued by the OMB in June 2003 (OMB, 2003). Each county is either inside or outside a metropolitan statistical area (MSA), as defined by the OMB. The definitions of urban influence county type are different than county-type definitions used in other NSDUH reports. The definitions for this report are based on UIC, whereas the 2004 national findings (OAS, 2005) definitions are based on Rural-Urban Continuum Codes (RUCC).

Large MSAs have a population of 1 million or more. Small MSAs have a population of fewer than 1 million. Nonmetropolitan areas are areas outside MSAs that have been categorized into four groups based on UIC. The first group consists of micropolitan statistical areas (MiSAs), which include a county with an urban cluster of at least 10,000 persons or more and any additional counties where commuting to the central county is 25 percent or higher, or where 25 percent of the employment in an outlying county is made up of commuters from the central county. The remaining three groups of nonmetropolitan areas consist of noncore counties and are divided based on their adjacency to larger areas and whether or not they have their "own town" of at least 2,500 residents. The "noncore adjacent with town" group includes those areas that are adjacent to a large MSA, adjacent to a small MSA and have their own town, or adjacent to a MiSA and have their own town. Noncore areas that have no town of their own but are adjacent to a small MSA or MiSA compose the "noncore adjacent, no town" group. The "noncore rural, not adjacent" group consists of counties that are not adjacent to any MSA or MiSA and have no town of their own.

1.3. Content and Organization of This Report

Subsequent chapters contain detailed information about several issues related to worker substance use and workplace drug and alcohol policies and programs. Chapter 2 provides estimates of substance use among full-time workers in the United States by demographic and geographic characteristics, including age, gender, race/ethnicity, education, family income, and place of residence. This provides some insight about the magnitude of worker substance use across different settings and population subgroups. Chapter 3 examines the characteristics of employers by providing estimates of the prevalence of substance use behaviors and substance use disorders by occupation, industry, and establishment size. Chapter 4 explores worker reports of drug information programs and policies in their employment settings. Chapter 5 provides detailed estimates about the prevalence of drug testing in the workplace. Chapter 6 discusses workplace behaviors and attitudes toward drug testing. In addition to the selected data tables included in these chapters and in Appendix E of this report, supplemental tables of estimates and their associated standard errors are available on the Internet at http://www.oas.samhsa.gov/work.htm. Standard error tables are presented for the tables in Appendix E as well as for the supplemental Web tables. The supplemental tables of estimates, which are denoted by "S" after the table number, are numbered to correspond to chapters and to consecutively follow Appendix E tables (e.g., Table 2.6S follows Table  2.5 in Appendix E and contains data related to analyses presented in Chapter 2). The standard error tables, which are denoted by "SE" after the table number, are numbered to correspond to the tables in Appendix E and to the supplemental Web tables (e.g., Table  2.1SE is the standard error table for Table  2.1 in Appendix E). Appendix F contains the reference list for this report.

Worker Substance Use and Workplace Policies and Programs

2. Substance Use among Workers, by Demographic and Geographic Characteristics

This chapter summarizes the substance use patterns of the population of full-time workers aged 18 to 64 in the United States between 2002 and 2004. The chapter also examines the differences in substance use for different demographic and geographic groups within that population. (See Table s 2.1 through 2.5 in Appendix E.) Demographic and geographic characteristics examined include age, gender, race/ethnicity, education, family income, and place of residence. Understanding which subpopulations in the workforce have the greatest prevalence of substance use behaviors and substance use disorders may allow policy makers to target workplace programs to specific settings and subpopulations.

The prevalences of substance use behaviors and substance use disorders are higher among unemployed persons than other employment statuses (Figure 2.1). However, because full-time workers constitute about two thirds of the population aged 18 to 64 (or 114.7 million persons), most substance users and most of those with substance use disorders are employed full time (Figure 2.2). Specifically, among those aged 18 to 64, 57.5 percent of past month illicit drug users, 58.0 percent of past month marijuana users, 67.3 percent of heavy alcohol users, 52.8 percent of those with dependence on or abuse of illicit drugs, and 65.1 percent of those with alcohol dependence or abuse were employed full time from 2002 to 2004 (Table  2.1).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 2.1 Past Month Substance Use and Past Year Substance Dependence or Abuse among Persons Aged 18 to 64, by Employment Status: 2002-2004 Combined

Below is a pie graph, click here for the text describing this graph.

Figure 2.2 Percentage Distribution of Persons Aged 18 to 64, by Employment Status: 2002-2004 Combined

2.1. Illicit Drug Use among Full-Time Workers

• The prevalence of past month illicit drug use among adult full-time workers was 8.2 percent (Figure 2.3 and Table s 2.2 and 2.3).
• Nearly one out of five (19.0 percent) workers aged 18 to 25 used illicit drugs during the past month. This was a higher percentage than among the 26-to-34 (10.3 percent), 35-to-49 (7.0 percent), and 50-to-64 (2.6 percent) age groups (Figure 2.3 and Table  2.2).
• Males were more likely than females to report past month illicit drug use (9.7 vs. 6.2 percent). Males accounted for about two thirds (6.4 million) of the workers who reported past month illicit drug use (Figure 2.4 and Table  2.2).
• The prevalence of past month illicit drug use for white adults was 8.8 percent, higher than the prevalence for Asian (2.2 percent) or Hispanic (6.7 percent) adults, and lower than that reported for adults who reported two or more races (13.5 percent). The prevalence of past month illicit drug use by Asians was lower than that reported by all other racial/ethnic groups reported here (Figure 2.5 and Table  2.2).
• Workers with a college education had a lower prevalence of current illicit drug use compared with those without a college education. The prevalence of past month use of illicit drugs was lower among those with higher levels of education than those with less education (college graduate [5.7 percent] vs. less than high school [11.2 percent]) (Figure 2.6 and Table  2.3).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 2.3 Past Month Substance Use and Past Year Substance Dependence or Abuse among Full-Time Workers, by Age Group: 2002-2004 Combined

Below is a vertical bar graph, click here for the text describing this graph.

Figure 2.4 Past Month Substance Use and Past Year Substance Dependence or Abuse among Full-Time Workers Aged 18 to 64, by Gender: 2002-2004 Combined

Below is a vertical bar graph, click here for the text describing this graph.

Figure 2.5 Past Month Substance Use among Full-Time Workers Aged 18 to 64, by Hispanic or Latino Origin and Race: 2002-2004 Combined

Below is a vertical bar graph, click here for the text describing this graph.

Figure 2.6 Past Month Substance Use and Past Year Substance Dependence or Abuse among Full-Time Workers Aged 18 to 64, by Education: 2002-2004 Combined

• The prevalence of current illicit drug use was lower among workers with higher family incomes than among workers with lower family incomes. An estimated 13.2 percent of workers who reported family income that was less than $20,000 had used illicit drugs during the past month. In contrast, 6.0 percent of workers who reported income in the highest category––$75,000 or more––had used illicit drugs during the past month (Figure 2.7 and Table  2.3).
• Residents of noncore counties had a lower prevalence of current illicit drug use (4.5 to 6.2 percent) compared with residents of micropolitan statistical area (7.1 percent), small metropolitan statistical area (MSA; 8.8 percent), and large MSA (8.3 percent) counties (Table  2.3).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 2.7 Past Month Substance Use and Past Year Substance Dependence or Abuse among Full-Time Workers Aged 18 to 64, by Family Income: 2002-2004 Combined

2.2. Past Month Marijuana Use among Full-Time Workers

• An estimated 6.4 percent, or 7.3 million, of full-time workers reported use of marijuana during the past month (Table s 2.2 and 2.3).
• Adults aged 26 to 34 were only about half as likely as 18- to 25-year-olds to be past month marijuana users (8.0 vs. 15.9 percent). Past month use of marijuana was lower with increasing age (Table  2.2).
• The prevalence of past month marijuana use was higher for males than females (7.9 vs. 4.3 percent, respectively) (Table  2.2).
• An estimated 11.0 percent of workers reporting two or more races used marijuana during the past month. This was higher than among non-Hispanic white adults (6.9 percent). Fewer Hispanic adults (4.6 percent) reported past month marijuana use than non-Hispanic white adults who reported two or more races (Table  2.2).
• Higher educational attainment and higher family income were associated with a lower prevalence of current marijuana use (Table  2.3).

2.3. Past Month Heavy Alcohol Use among Full-Time Workers

• An estimated 8.8 percent, or 10.1 million, of full-time workers reported past month heavy alcohol use (Figure 2.3 and Table s 2.2 and 2.3).
• Past month heavy alcohol use was related to age. Among younger workers (18 to 25 years old), 16.3 percent reported past month heavy alcohol use compared with 10.4 percent of 26- to 34-year-olds, 8.1 percent of 35- to 49-year-olds, and 4.7 percent of 50- to 64-year-olds (Figure 2.3 and Table  2.2).
• Males were three times as likely as females to be past month heavy alcohol users (12.3 vs. 4.1 percent) (Figure 2.4 and Table  2.2).
• An estimated 10.1 percent of white adults reported heavy alcohol use in the past month. This was higher than the percentage among black adults (5.4 percent), Asian adults (2.9 percent), Hispanic adults (6.9 percent), and adults reporting two or more races (7.5 percent) (Figure 2.5 and Table  2.2).
• Residents of noncore rural counties had a lower prevalence of past month heavy alcohol use (7.5 percent) compared with residents of micropolitan statistical area (9.2 percent), small MSA (9.8 percent), and large MSA (8.1 percent) counties (Table  2.3).
• Workers with a college education had a lower prevalence of past month heavy alcohol use compared with those without a college education. Past month heavy alcohol use was lower among those with higher levels of education than those with less education (college graduate [6.7 percent] vs. less than high school [10.8 percent]) (Figure 2.6 and Table  2.3).
• Workers in the Midwest had the highest prevalence of past year heavy alcohol use (10.6 vs. 8.4 percent in the Northeast, 8.5 percent in the South, and 7.8 percent in the West) (Table  2.3).

2.4. Past Year Illicit Drug Dependence or Abuse among Full-Time Workers

• Approximately 3 million full-time workers (2.6 percent) aged 18 to 64 met the criteria for past year illicit drug dependence or abuse (Figure 2.3 and Table  2.4).
• Approximately 7.5 percent of 18- to 25-year-old workers had past year illicit drug dependence or abuse. This was higher than among all other age groups studied (26- to 34-year-olds [3.3 percent], 35- to 49-year-olds [1.9 percent], and 50- to 64-year-olds [0.7 percent]) (Figure 2.3 and Table  2.4).
• Males were nearly twice as likely as females to meet the criteria for past year illicit drug dependence or abuse (3.3 vs. 1.8 percent) (Figure 2.4 and Table  2.4).
• Hispanics (3.2 percent) had a higher prevalence of past year illicit drug dependence or abuse than non-Hispanics (2.6 percent) (Figure 2.8 and Table  2.4).
• Within non-Hispanic subgroups, Asians had the lowest prevalence of past year illicit drug dependence or abuse (1.1 percent). This was lower than non-Hispanic white adults (2.5 percent), black (2.9 percent) adults, American Indian or Alaska Native (4.5 percent) adults, and adults reporting two or more races (4.3 percent) (Figure 2.8 and Table  2.4).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 2.8 Past Year Substance Dependence or Abuse among Full-Time Workers Aged 18 to 64, by Hispanic or Latino Origin and Race: 2002-2004 Combined

• Educational attainment was linked to past year illicit drug dependence or abuse. An estimated 1.2 percent of college graduates met the criteria for dependence or abuse compared with 5.1 percent of those with less than high school education, 2.9 percent of high school graduates, and 2.9 percent of workers with some college (Figure 2.6 and Table  2.5).
• Among higher income groups ($75,000 or more and $50,000-74,999), 1.7 and 2.0 percent had past year illicit drug dependence or abuse, respectively. These two groups had a lower prevalence than among lower income groups (3.0 to 5.6 percent) (Figure 2.7 and Table  2.5).

2.5. Past Year Alcohol Dependence or Abuse among Full-Time Workers

• Approximately 10.6 million full-time workers aged 18 to 64 (9.2 percent) met the criteria for past year alcohol dependence or abuse (Figure 2.3 and Table s 2.4 and 2.5).
• The prevalence of past year alcohol dependence was highest among those aged 18 to 25 (18.4 percent) compared with those aged 26 to 34 (12.3 percent), 35 to 49 (7.8 percent), and 50 to 64 (4.0 percent) (Figure 2.3 and Table  2.4).
• Males were about twice as likely as females to have past year alcohol dependence or abuse (11.8 vs. 5.7 percent) (Figure 2.4 and Table  2.4).
• The prevalence of past year alcohol dependence or abuse was lower among non-Hispanic Asian and non-Hispanic black adults than among other non-Hispanic groups (4.6 and 7.3 percent, respectively). Hispanic and non-Hispanic adults did not differ in the prevalence of alcohol disorders (10.0 vs. 9.1 percent) (Figure 2.8 and Table  2.4).
• An estimated 11.9 percent of those with less than a high school diploma reported past year alcohol dependence or abuse compared with 9.4 percent of high school graduates, 9.7 percent of those with some college, and 7.5 percent of those who graduated from college (Figure 2.6 and Table  2.5).
• Workers in the Midwest had the highest prevalence of past year alcohol dependence or abuse (10.5 vs. 8.3 percent in the Northeast, 8.5 percent in the South, and 9.9 percent in the West) (Table  2.5).
• The prevalence of alcohol dependence or abuse was higher among workers with family incomes that were less than $20,000 (13.3 percent) compared with workers with family incomes that were higher than $75,000 (7.6 percent) (Figure 2.7 and Table  2.5).

2.6. Summary

The results presented in this chapter demonstrate a consistent relationship between several demographic and geographic characteristics and substance use and dependence or abuse. Age was consistently associated with substance use behaviors and substance use disorders. Young adults aged 18 to 25 were more likely than other age groups to report past month illicit drug or heavy alcohol use or to meet the criteria for substance dependence or abuse during the past year. Adult males who worked full time had a higher prevalence on all substance use and dependence or abuse measures compared with females who worked full time. Lower educational attainment and lower family income also were consistently associated with a higher prevalence of substance use behaviors and substance use disorders. There were some race/ethnicity-related differences, and Asians generally had a lower prevalence of use and dependence on or abuse of illicit drugs and alcohol. Findings were less consistent across measures of substance use behaviors and substance use disorders for county type. Past month illicit drug use was highest in the Northeast, while past year alcohol dependence or abuse was highest in the Midwest.

Worker Substance Use and Workplace Policies and Programs

3. Substance Use among Workers, by Occupation, Industry, and Establishment Size

This chapter reports the prevalence of substance use among full-time workers within occupations, industries, and across different sizes of workplace establishments. Prevalence is reported for measures of past month illicit drug use, past month heavy alcohol use, past month marijuana use, and past year dependence on or abuse of illicit drugs or alcohol. Major occupational groups are identified using the standard occupational classifications (SOCs) of the U.S. Department of Labor (see Appendices C and D) and selected broader and more detailed occupational groups. Industry groupings are identified using the North American Industry Classification System (NAICS) developed by the U.S. Census Bureau. Data are presented for NAICS sectors and selected subsectors and industry groups. The chapter also examines the relationship between demographic correlates of substance use within industry and occupational categories in order to assess if the observed rates of substance use are a function of the characteristics of the workers in those categories. Finally, this chapter examines the relationship between substance use among workers and establishment size, which ranges from small (fewer than 10 employees) to large (500 or more employees).

Previous research has demonstrated that substance use and dependence or abuse may vary by workplace environment. In 1996, the Substance Abuse and Mental Health Services Administration (SAMHSA) released a report detailing the prevalence of illicit drug use among workers in specific occupation and industry categories (Hoffmann, Brittingham, & Larison, 1996). Occupations involved in construction and food preparation were found to have the highest prevalence of illicit drug use and heavy alcohol use. Illicit drug use was lowest among public safety occupations, including police, teachers, child care workers, and data clerks. Heavy alcohol use was lowest among data clerks, personnel specialists, and secretaries. Industries associated with the highest prevalence of illicit drug use included eating and drinking establishments, certain retail sales categories, and the entertainment industry. The computer and data processing industry had the highest rates of heavy alcohol use. The lowest rates of illicit drug use were found among workers in child care, professional medical offices, and schools.

The 1996 SAMHSA report also demonstrated that the relationship between substance use, age, gender, and marital status within industries and occupations was consistent with the pattern for all full-time workers. This suggests that substance use patterns across industries and occupations may partly be a function of the demographic characteristics of the workers. That is, demographic groups with relatively higher prevalence of substance use may be more likely to be employed in certain industries or occupations for reasons unrelated to substance use. For example, young workers tend to work in food preparation occupations because these positions require less experience, education, and training than other occupations. To account for these variations in the demographic makeup of different occupation and industry categories, it is useful to consider the data shown in Table s 3.1 and 3.2 in Appendix E when comparing substance use rates across occupation and industry groups. Table s 3.1 and 3.2 show the average age and percentage of male workers in each group.

Substance use, dependence, or abuse also may vary by the size of the work establishment. Establishment size becomes an important factor if either substance use or the prevalence of reduction programs is disproportionately distributed to large or small employment settings. Previous findings suggest that workplace drug testing is less likely to occur in small workplace settings (Zhang, Huang, & Brittingham, 1999; Hartwell, Steele, French, & Rodman, 1996; Hartwell, Steele, & Rodman, 1998).

3.1. Standard Occupational and Industry Classifications

The occupations are coded into groups using the 2000 Standard Occupational Classification (SOC) released by the U.S. Department of Labor, Bureau of Labor Statistics (2000), which categorizes all occupations into 21 major groups. Within these major groups are 96 minor groups, 449 broad occupations, and 821 detailed occupations. Occupations with similar skills or work activities are grouped at each of the four levels of hierarchy to facilitate comparisons. The NAICS, which replaced the Standard Industry Classification (SIC), categorizes all industries into 19 major groups and is used to classify industries in this report. Industries are organized within the NAICS by the processes used to produce goods or services. This report focuses on these major groups, but a more detailed analysis of industry and occupational classifications can be found in Appendix D. Table s 3.1 and 3.2 show the population totals for industry and occupation categories.

3.2. Illicit Drug Use among Full-Time Workers, by Occupation

It is notable, particularly in light of demographic characteristics presented earlier in this report, that among the five occupations with the highest prevalence of current illicit drug use, three were overwhelmingly dominated by male workers. An estimated 97.4 percent of construction workers were male, while 96.2 percent of installation maintenance and repair and 87.2 percent of transportation and material-moving occupations were male. Moreover, the full-time workers in this "top five" list were among the youngest workers in the population (Table  3.1).

Figure 3.1 lists the 21 major occupational groups ordered highest to lowest in percentage of full-time workers within the occupation who reported past month use of illicit drugs.

• Looking at the major occupational groups for 2002 through 2004, food service workers (17.4 percent) and construction workers (15.1 percent) exhibited a higher prevalence of past month illicit drug use than other occupational groups (Figure 3.1 and Table  3.3).
• Those working in education, training, and library occupations (4.1 percent), community and social services occupations (4.0 percent), and protective service occupations (3.4 percent) had the lowest prevalence of past month illicit drug use among the major occupational groups (Figure 3.1 and Table  3.3).

Below is a bar graph, click here for the text describing this graph.

Figure 3.1 Past Month Illicit Drug Use among Full-Time Workers Aged 18 to 64, by Major Occupational Categories: 2002-2004 Combined

3.3. Heavy Alcohol Use among Full-Time Workers, by Occupation

• The major occupational groups with the highest prevalence of past month heavy alcohol use were construction and extraction occupations (17.8 percent) and installation, maintenance, and repair occupations (14.7 percent) (Figure 3.2 and Table  3.3).
• Community and social services occupations (2.8 percent) had the lowest prevalence of past month heavy alcohol use of the major occupations (Figure 3.2 and Table  3.3).

Below is a bar graph, click here for the text describing this graph.

Figure 3.2 Past Month Heavy Alcohol Use among Full-Time Workers Aged 18 to 64, by Major Occupational Categories: 2002-2004 Combined

3.4. Dependence and Abuse among Full-Time Workers, by Occupation

• The major occupational groups with the highest prevalence of illicit drug dependence or abuse in the past year were food preparation and serving related occupations (6.5 percent) and construction and extraction occupations (6.2 percent) (Table  3.4).
• Construction and extraction occupations (16.9 percent) and food preparation and serving related occupations (14.7 percent) had the highest prevalence of alcohol dependence or abuse in the past year in the major occupational groups (Table  3.4).

3.5. Illicit Drug Use among Full-Time Workers, by Industry

• The major industry groups with the highest prevalence of past month illicit drug use were accommodations and food services (16.9 percent) and construction (13.7 percent). Public administration (4.1 percent), educational services (4.0 percent), and utilities (3.8 percent) had the lowest prevalence of past month illicit drug use (Figure 3.3 and Table  3.5).

Below is a bar graph, click here for the text describing this graph.

Figure 3.3 Past Month Illicit Drug Use among Full-Time Workers Aged 18 to 64, by Industry Categories: 2002-2004 Combined

3.6. Heavy Alcohol Use among Full-Time Workers, by Industry

• The industry groups with the highest prevalence of past month heavy alcohol use were construction (15.9 percent); arts, entertainment, and recreation (13.6 percent); and mining (13.3 percent) industries. However, health care and social assistance (4.3 percent) and educational services (4.0 percent) had the lowest prevalence of past month heavy alcohol use compared with other major industries (Figure 3.4 and Table  3.5).

Below is a bar graph, click here for the text describing this graph.

Figure 3.4 Past Month Heavy Alcohol Use among Full-Time Workers Aged 18 to 64, by Industry Categories: 2002-2004 Combined

3.7. Dependence and Abuse among Full-Time Workers, by Industry

• The major industry groups with the highest prevalence of illicit drug dependence or abuse in the past year were accommodations and food services (6.0 percent) and construction (5.1 percent) (Table  3.6).
• Construction (15.6 percent) and accommodations and food services (15.4 percent) had the highest prevalence of alcohol dependence or abuse in the past year in the major industry groups (Table  3.6).

3.8. Past Month Illicit Drug Use among Full-Time Workers, by Establishment Size

• Among the 9.4 million full-time workers who reported past month illicit drug use, 2.9 million were employed in establishments with fewer than 10 employees. The prevalence of past month illicit drug use for workers in establishments with fewer than 10 employees was 9.9 percent (Table  3.7).
• An estimated 1.9 million workers who reported past month illicit drug use worked in settings with 10 to 24 employees. The past month illicit drug use prevalence among workers in these establishments was 9.7 percent, which was similar to the prevalence estimate (9.9 percent) for smaller establishments (fewer than 10 employees) (Table  3.7).
• Prevalence of past month illicit drug use was lower as establishment size increased. The prevalence among workers in establishments with 25 to 99 employees was 8.2 percent, compared with 6.7 percent among workers in establishments with 100 to 499 employees and 5.7 percent among workers in establishments with 500 or more employees (Table  3.7).

3.9. Past Month Marijuana Use among Full-Time Workers, by Establishment Size

• Among full-time workers in establishments with fewer than 10 employees, 7.9 percent had used marijuana in the past month. Similarly, 7.8 percent of workers in slightly larger establishments with 10 to 24 employees and 6.4 percent of workers in establishments with 25 to 99 employees reported past month marijuana use (Table  3.7).
• Prevalence of past month marijuana use was lower among workers in larger establishments of 100 to 499 employees and 500 or more employees (4.9 and 4.1 percent, respectively) than among workers in smaller establishments (Table  3.7).

3.10. Past Month Heavy Alcohol Use among Full-Time Workers, by Establishment Size

• The prevalence of past month heavy alcohol use among full-time workers in the smallest establishments (fewer than 10 employees) was 10.1 percent. This was higher than the prevalence reported for similar workers in the largest establishments (6.8 percent in establishments with 500 or more employees) (Table  3.7).

3.11. Past Year Illicit Drug Dependence or Abuse among Full-Time Workers, by Establishment Size

• An estimated 3.0 million full-time workers in the United States met the criteria for past year illicit drug dependence or abuse. More than half of these workers (1.6 million) were employed in small establishments with fewer than 25 employees (Table  3.8).
• Among workers in small establishments with fewer than 10 employees, 3.1 percent reported symptoms consistent with past year illicit drug dependence or abuse. This was higher than the prevalence of illicit drug dependence or abuse among workers in large establishments with 500 or more employees (1.6 percent). There was a steady decline in prevalence with increases in establishment size (Table  3.8).

3.12. Past Year Alcohol Dependence or Abuse among Full-Time Workers, by Establishment Size

• Full-time workers in the largest establishments (500 or more employees) reported the lowest prevalence of past year alcohol dependence or abuse (7.5 vs. 9.0 to 9.9 percent in smaller establishments) (Table  3.8).

3.13. Summary

The prevalence of substance use, dependence, and abuse varied across occupations and industries. For example, highest rates of illicit drug use were found among food preparation, construction and extraction, and arts occupations and among food services, construction, and arts industries. These findings were likely influenced by characteristics of the workers in occupations and industries. Certain occupations and industries have higher percentages of males and/or younger workers, characteristics associated with higher rates of substance use. Indeed, the five occupations with the highest prevalence of illicit drug use were dominated by male workers. Cultural shifts that increase the rate of females working in positions previously dominated by males should be monitored to assess the impact of these changes on substance use prevalence within occupational and industrial groups.

Worker Substance Use and Workplace Policies and Programs

4. Workplace Information Policies and Programs concerning Drug and Alcohol Use

This chapter examines the prevalence of workplace policies and programs designed to reduce substance use and the relationship between worker substance use and those workplace efforts. The three types of workplace policies and programs examined are (1) the availability of educational materials about the effects of substance use, (2) the use of a written policy about substance use, and (3) the presence of an employee assistance program (EAP). Employers may offer any, all, or none of these programs and policies to employees, and these programs may be available only to full-time employees in some workplaces.

There is little prior evidence to suggest that these efforts alone are effective deterrents, unless they are combined with drug-testing programs. Some research has demonstrated that EAPs provide workers a venue for seeking help, but only when the employees are aware of the services, have a generally positive attitude toward the EAP, and believe that job security is not threatened as a result of seeking assistance with a substance use problem (Delaney, Grube, & Ames, 1998; Reynolds & Lehman, 2003).

4.1. Demographic Characteristics of Workers Reporting Workplace Drug Policies and Programs

• Among the nearly 115 million full-time workers aged 18 to 64 years old in the United States, 47.7 million (43.8 percent) reported that they had access to educational information about drug and alcohol use in the workplace, 87.0 million (78.7 percent) reported that they were aware of a written policy about drug and alcohol use in the workplace, and 60.9 million (58.4 percent) reported that their employer offered an EAP (Table s 4.1 and 4.2 in Appendix E).
• The youngest adult workers were least likely to report access to educational information about drug and alcohol use in the workplace. Among 18- to 25-year-old workers, 33.2 percent reported that they had educational information available. This was significantly lower than among workers aged 26 to 34 years (39.6 percent), 35 to 49 years (46.3 percent), and 50 to 64 years (48.9 percent) (Figure 4.1 and Table  4.1).
• Young adult workers between the ages of 18 and 25 were significantly less likely to report EAPs available in the workplace compared with all other age groups (39.7 vs. 56.4 to 62.6 percent) (Figure 4.1 and Table  4.1).
• Females were slightly more likely to report working for an employer who provided a written drug and alcohol use policy in the workplace (81.8 vs. 76.4 percent) and who had an EAP available (61.0 vs. 56.5 percent) in comparison with males (Figure 4.2 and Table  4.1).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.1 Workplace Provides Educational Information, Prepares a Written Policy, or Maintains an Employee Assistance Program concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Age Group: 2002-2004 Combined

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.2 Workplace Provides Educational Information, Prepares a Written Policy, or Maintains an Employee Assistance Program concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Gender: 2002-2004 Combined

• Non-Hispanic black (56.3 percent), American Indian or Alaska Native (54.1 percent), and Native Hawaiian or Other Pacific Islander (54.0 percent) adults were more likely to report that they had access to educational information in the workplace than non-Hispanic white (42.7 percent) and Hispanic adults (38.4 percent) (Figure 4.3 and Table  4.1).
• Workers who were non-Hispanic Asian (41.6 percent) or who were Hispanic (38.4 percent) had the lowest prevalence of all racial/ethnic groups to report working for employers who had a written policy about drug and alcohol use (Figure 4.3 and Table  4.1).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.3 Workplace Provides Educational Information, Prepares a Written Policy, or Maintains an Employee Assistance Program concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Hispanic or Latino Origin and Race: 2002-2004 Combined

• Educational attainment was consistently associated with higher access to information resources. Those with higher levels of education were more likely than those with a high school education or less to report access to drug and alcohol information (Figure 4.4).
• The availability of an EAP also was associated with education. An estimated 32.1 percent of adults with less than a high school diploma reported an EAP available in the workplace, while 55.4 percent of high school graduates, 61.7 percent of adults with some college, and 69.3 percent of college graduates reported access to an EAP (Table  4.2).
• For all levels of education, workers with family income less than $20,000 were less likely than those with higher family incomes to report availability of educational information about drug and alcohol use, availability of an EAP, and the presence of a written policy about drug and alcohol use through the employment setting (Figures 4.4 through 4.6).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.4 Workplace Provides Educational Information concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Education and Family Income: 2002-2004 Combined

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.5 Workplace Prepares a Written Policy concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Education and Family Income: 2002-2004 Combined

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.6 Workplace Maintains an Employee Assistance Program concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Education and Family Income: 2002-2004 Combined

• Workers who lived in the most rural settings (noncore adjacent, no town and noncore rural, not adjacent) were less likely than workers in other settings to report working for employers with a written policy about drug and alcohol use (Table  4.2).
• About 47 percent of workers who resided in rural settings reported that they had an EAP at their place of employment (47.2 percent for noncore rural, not adjacent and 47.5 percent for noncore adjacent, no town), while more than 59 percent of metropolitan-residing workers (59.3 percent in large metropolitan statistical areas [MSAs] and 59.6 percent in small MSAs) reported that they had this benefit (Table  4.2).

4.2. Working for Employers with Drug and Alcohol Policies and Programs, by Current Illicit Drug Use

• Nearly 3 million (32.1 percent) full-time workers between the ages of 18 and 64 who had used an illicit drug in the past month reported that they worked for an employer who offered educational information about alcohol and drug use. An EAP was reported available to 3.9 million (45.4 percent) workers who were past month users of an illicit drug, while 6.5 million (71.0 percent) reported working for employers who had a written policy about drug and alcohol use (Figure 4.7 and Table s 4.3 and 4.4).
• Generally, past month illicit drug users were less likely to report working for employers who offered workplace drug or alcohol programs or policies, compared with those who did not use an illicit drug in the past month. An estimated 45.4 percent of current illicit drug users reported access to an EAP at their place of employment compared with 59.6 percent of workers who had not used an illicit drug in the past month. This finding occurred across demographic comparisons with few exceptions (Figure 4.7 and Table s 4.3 and 4.4).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.7 Workplace Provides Educational Information, Prepares a Written Policy, or Maintains an Employee Assistance Program concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Past Month Illicit Drug Use: 2002-2004 Combined

• Among 18- to 25-year-old workers, 28.2 percent of current illicit drug users reported that they worked for an employer who provided educational information about drug and alcohol use, while 34.4 percent of same-age nondrug users reported that they worked for employers with educational programs. This finding was similar across other age group comparisons: 26 to 34 (31.3 vs. 40.6 percent), 35 to 49 (35.7 vs. 47.1 percent), and 50 to 64 (33.2 vs. 49.3 percent) (Figure 4.8 and Table  4.3).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.8 Workplace Provides Educational Information concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Age Group and Past Month Illicit Drug Use: 2002-2004 Combined

• A reported written policy regarding drug and alcohol use in the employment setting was unrelated to current drug use among workers aged 50 to 64. An estimated 76.3 percent of current drug users in this age group reported that they worked for employers with a written policy, while 77.1 percent of workers in this cohort who had not used drugs in the past month reported that they were aware of a written policy at work (Figure 4.9 and Table  4.3).
• Among all age groups, past month drug users were less likely than nondrug users to report access to an EAP (34.3, 44.4, 51.9, and 58.2 vs. 41.0, 57.9, 63.2, and 62.7 percent, respectively) (Figure 4.10).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.9 Workplace Prepares a Written Policy concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Age Group and Past Month Illicit Drug Use: 2002-2004 Combined

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.10 Workplace Maintains an Employee Assistance Program concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Age Group and Past Month Illicit Drug Use: 2002-2004 Combined

• Males were more likely than females to have used illicit drugs in the past month, but among past month drug users, females were more likely than males to report working for employers who offered EAPs or who had a written drug and alcohol policy (EAP: 49.4 vs. 43.5 percent; written policy: 76.3 vs. 68.5 percent) (Table  4.3).
• Among white adults, black adults, and adults who reported two or more races, past month drug users were more likely than nondrug users to report access to an EAP (47.0, 43.1, and 31.2 vs. 61.9, 67.7, and 66.8 percent, respectively). However, reporting access to an EAP did not differ among Hispanic drug users and nondrug users (37.1 vs. 40.7 percent). This may be related to the generally low availability of EAPs as a benefit for the Hispanic population (Figure 4.11 and Table  4.3).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.11 Workplace Maintains an Employee Assistance Program concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Hispanic or Latino Origin and Race and Past Month Illicit Drug Use: 2002-2004 Combined

Note: Due to low precision, estimates for American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, and Asian full-time workers with past month illicit drug use are not shown. Estimates for Native Hawaiian or Other Pacific Islander full-time workers with no past month illicit drug use also are not shown.

• Among workers with less than a high school diploma, nondrug users were more likely than past month drug users to report that their employers provided educational information regarding alcohol and drug use (34.8 and 29.5 percent, respectively). The differences for reporting written policies (66.2 percent of past month users vs. 69.4 percent of nonusers) and an EAP (28.5 percent of past month users vs. 32.5 percent of nonusers) were not found to be significant among workers with less than a high school diploma (Table  4.4).
• There was some regional variation among past month drug users who reported working for an employer with substance abuse policies and programs. Workers who resided in the South and who used an illicit drug in the past month were less likely to report working for an employer with an EAP than similar workers from other regions (39.5 vs. 48.0 to 48.5 percent). There were no differences among the other regions (Table  4.4).

4.3. Working for Employers with Educational Programs, Written Policies, or EAPs, by Current Heavy Alcohol Use

• The relationship between current heavy alcohol use and working for an employer with drug and alcohol programs was generally weaker than the relationship between current illicit drug use and the availability of employer-sponsored programs (Table s 4.5 and 4.8).
• More than 3.6 million full-time workers in the United States who were past month heavy alcohol users reported that they were provided educational information about drug and alcohol use by their employer. An estimated 7.2 million workers who were past month heavy alcohol users reported that they worked for employers with written policies about drug and alcohol use, and 4.7 million reported that they had access to an EAP at their place of employment (Table s 4.5 and 4.7).
• An estimated 71.5 percent of 18- to 25-year-old workers who were heavy alcohol users during the past month reported that they worked in employment settings with a written policy about alcohol and drug use. This was slightly less than the 77.9 percent of same-age workers who were not current heavy alcohol users but reported that they had a written policy in the workplace (Table  4.6).
• Heavy alcohol use was not significantly associated with the reported availability of an EAP among 26- to 34-year-olds. However, among 18- to 25-year-olds, past month heavy alcohol use was associated with a lower likelihood of working in a setting that offered an EAP, compared with those who had no past month heavy alcohol use in the same age group (38.1 vs. 40.1 percent). This was true in the 35-49 and 50-to-64 age categories as well (56.5 vs. 62.9 percent and 53.4 vs. 63.1 percent, respectively) (Table  4.6).
• Non-Hispanic white current heavy alcohol users were less likely to report working in a setting with an EAP available than their non-heavy alcohol user counterparts (52.4 vs. 61.5 percent). However, among Hispanic workers, 41.4 percent of past month heavy alcohol users reported access to an EAP, and 40.4 percent of non-heavy alcohol users reported access to an EAP. Recall that Hispanic workers were generally less likely than other racial/ethnic groups to report access to an EAP (Figures 4.12 through 4.14 and Table  4.6).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.12 Workplace Provides Educational Information concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Hispanic or Latino Origin and Race and Past Month Heavy Alcohol Use: 2002-2004 Combined

Note: Due to low precision, estimates for American Indian or Alaska Native and Native Hawaiian or Other Pacific Islander full-time workers with past month heavy alcohol use are not shown.

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.13 Workplace Prepares a Written Policy concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Hispanic or Latino Origin and Race and Past Month Heavy Alcohol Use: 2002-2004 Combined

Note: Due to low precision, estimates for American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, Asian, and Two or More Races full-time workers with past month heavy alcohol use are not shown. Estimates for Native Hawaiian or Other Pacific Islander full-time workers with no past month heavy alcohol use also are not shown.

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.14 Workplace Maintains an Employee Assistance Program concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64, by Hispanic or Latino Origin and Race and Heavy Alcohol Use: 2002-2004 Combined

Note: Due to low precision, estimates for American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, Asian, and Two or More Races full-time workers with past month heavy alcohol use are not shown. Estimates for Native Hawaiian or Other Pacific Islander full-time workers with no past month heavy alcohol use also are not shown.

• Heavy alcohol use during the past month was significantly associated with the reported availability of an EAP among more highly educated workers. An estimated 32.9 percent of past month heavy alcohol users with less than a high school diploma reported that they worked in an environment where an EAP was available, while 61.2 percent of college graduates with past month heavy alcohol use reported that they had access to an EAP in their workplace (Figure 4.15 and Table  4.8).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 4.15 Workplace Provides Educational Information, Prepares a Written Policy, or Maintains an Employee Assistance Program concerning Drug or Alcohol Use among Full-Time Workers Aged 18 to 64 with Past Month Heavy Alcohol Use, by Education: 2002-2004 Combined

• There was a significant difference between past month heavy alcohol users and non-heavy alcohol users in the reported availability of a written drug policy or drug education at higher levels of education, but not at lower levels of education. Among college graduates, 35.5 percent of those with past month heavy alcohol use reported drug and alcohol education in their workplace compared with 46.5 percent of non-heavy alcohol users. Similarly, 75.3 percent of past month heavy alcohol users reported that their employer had a written policy about drug and alcohol use, compared with 81.5 percent of non-past month heavy alcohol users among college graduates. Similar patterns were observed for EAPs (Table  4.8).
• There were no significant differences between geographic regions or county types in the relationship between heavy alcohol use and the reported availability of employer-offered educational information about drug and alcohol use (Table s 4.7 and 4.8).

4.4. Working for Employers with Educational Programs, Written Policies, or EAPs, by Illicit Drug or Alcohol Dependence or Abuse during the Past Year

Dependence or abuse during the past year represents a more complex, chronic, and serious form of substance use compared with past month use alone. Dependent and abusing individuals may be the most likely to be affected (and subsequently helped) by drug and alcohol programs in their places of employment. If these programs are effective deterrents, then dependent or abusing individuals would be less likely to work (or continue to work) in an environment with strict drug policies. Previous research suggests that those with dependence on or abuse of illicit drugs or alcohol are more likely to miss workdays due to their substance abuse issues, are more likely to have work-related accidents, and are more likely to experience health-related consequences due to their substance problems (Normand, Lempert & O'Brien, 1994; Mangione et al., 1999).

• Among those who had a past year illicit drug dependence or abuse disorder, there were few differences between age categories in the likelihood of having access to employer-sponsored programs. For example, 26.6 percent of 18- to 25-year-olds, 29.8 percent of 26- to 34-year-olds, and 34.3 percent of 35- to 49-year-olds with a past year disorder reported the availability of educational information about drug and alcohol use in their workplace. These percentages among age groups were not significantly different (Table  4.9).
• However, across most demographic comparisons, those who met criteria for dependence or abuse were less likely than those who did not meet the criteria to report working for an employer who had educational programs, EAPs, and written policies about drug and alcohol use. Among non-Hispanic white adult workers who had past year illicit drug dependence or abuse, 27.9 percent reported working for an employer who had educational information, 39.3 percent reported working for an employer who offered an EAP, and 68.4 percent reported working for an employer with a written drug or alcohol education program. Among non-Hispanic white adults who did not meet criteria for past year illicit drug dependence, 43.1 percent reported access to educational information at their workplace, 61.1 percent reported that their employer had an EAP, and 78.7 percent reported working for an employer with a written drug and alcohol policy (Table  4.9).
• Among those with past year illicit drug dependence or abuse, 70.6 percent reported working for an employer with a written policy (Table  4.9), compared with 71.0 percent of current users who reported working for employers with a written policy about drug and alcohol use (Table s 4.3 and 4.4).
• Alcohol dependence or abuse has a similar association with reported availability of employer-provided information and programs regarding drug and alcohol use when compared with illicit drug dependence or abuse. An estimated 37.2 percent of those who had past year alcohol dependence or abuse reported access to educational information about drug and alcohol use at their workplace, 51.0 percent reported access to an EAP at their workplace, and 74.9 percent reported that they had a written drug and alcohol use policy at their workplace (vs. 44.4 percent, 59.2 percent, and 79.1 percent, respectively, among those with no alcohol dependence or abuse during the past year) (Table s 4.10 and 4.11).

4.5. A Final Note about EAPs

EAP programs are generally established to provide short-term counseling and problem solving and may provide a greater deterrent to illicit drug and alcohol use than educational information and written policies. EAPs may be considered by some employers to be too costly to offer, but they may serve as a first line of access to the health care system with relatively few barriers when appropriately implemented. There is generally no cost for the employee, appointments may be by phone or in person, and the EAP may serve as a referral source for longer term needs. However, there does appear to be unequal access to EAPs across demographic groups. Specifically, adult workers aged 18 to 25 were less likely than older adult workers to report access to an EAP. Younger workers had an increased risk for substance use disorders and yet were least likely to report access to an EAP. Further research will be necessary to assess the contribution that EAPs make to the reduction of substance use disorders, as well as the development of appropriate policy and programming to develop programs targeted at specific demographic groups. Hispanics also were distinctive in reporting less access to EAPs. This may be indicative of other underlying differences among racial/ethnic groups such as educational attainment, occupation and industry opportunities, and age distribution. Further analysis should be conducted to clarify this finding and to improve access for these underserved groups.

4.6. Summary

Workplace drug and alcohol policies and programs serve to communicate a "no-drugs-allowed" attitude that may deter current users from applying and working for employers with this position and also may encourage current users to leave the organization.

Educational attainment was consistently associated with higher access to information resources, and those with higher levels of education were more likely than those with a high school education or less to report access to drug and alcohol information or an EAP. Males were more likely than females to report having used illicit drugs in the past month, but females were more likely than males to report working for employers who offered EAPs or had a written drug and alcohol policy. Non-Hispanic white past month heavy alcohol users were less likely to report working in a setting with an EAP available than their non-heavy alcohol user counterparts. Across most demographic comparisons, those who met criteria for dependence or abuse were less likely than those who did not meet the criteria to report working for an employer who had educational programs, EAPs, and written policies about drug and alcohol use.

The workplace programs were all, to a greater or lesser degree, associated with a lower likelihood that current illicit drug users would work in settings with any of the programs. However, the effect of these programs on potential new hires and existing staff cannot be evaluated with cross-sectional data.

Worker Substance Use and Workplace Policies and Programs

5. Workplace Testing

This chapter examines the relationship between worker substance use and testing programs while controlling for age, race/ethnicity, gender, education, income, geographic setting, occupational group, and establishment size. Workplace drug testing was implemented as an effort to deter substance abuse and its effects on productivity, health, and safety in the Nation's workforce. To date there is limited evidence about the effectiveness of this deterrent effect (Normand, Lempert, & O'Brien, 1994; Hoffmann, Larison, & Sanderson, 1997; Office of Applied Studies [OAS], 1999; French, Roebuck, & Alexandre, 2004). In the National Survey on Drug Use and Health (NSDUH), respondents were asked to indicate whether their employer ever tested employees for alcohol use and/or illicit drug use. Those who indicated at least one of these testing parameter policies then were asked to identify whether testing occurred during the hiring process and/or for employees on a random basis. It is important to note that NSDUH does not collect data related to job tenure and is not a longitudinal survey. Thus, it is not possible to draw conclusions about the causal direction of the relationship between testing and substance use. However, prehire- and random-testing programs in establishments are likely to reduce the prevalence of worker substance use through exclusionary screening and early termination of users.

An estimated 38.7 million full-time workers reported that their employer ever conducted testing for alcohol use. Reported testing for alcohol use was more common among workers aged 35 to 49 than workers aged 18 to 25, 26 to 34, and 50 to 64 (37.5 vs. 30.5, 34.2, and 35.3 percent, respectively), among males than females (37.7 vs. 32.1 percent), and among those living in the South than among those living in the Northeast, Midwest, or West (39.6 vs. 26.8, 36.7, and 34.3 percent, respectively) (Figure 5.1 below and Table s 5.1 and 5.2 in Appendix E).

Testing for illicit drug use was reported more often than testing for alcohol use. An estimated 54 million full-time workers reported that their employer ever tested for illicit drug use. Reported testing for illicit drug use was more common among workers aged 35 to 49 than workers aged 18 to 25, 26 to 34, and 50 to 64 (50.4 vs. 46.7, 49.9, and 46.0 percent, respectively), among males than females (51.4 vs. 45.1 percent), and among workers who were black than among those who were white, American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, or Asian, or reported two or more races (63.1 vs. 46.9, 59.9, 56.5, 44.5, and 49.1 percent, respectively) (Figure 5.1 and Table s 5.1 and 5.2).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.1 Workplace Drug or Alcohol Use Testing Practices among Full-Time Workers Aged 18 to 64: 2002-2004 Combined

5.1. Prehire-Testing Programs among Full-Time Workers

• Among the Nation's full-time workers, 42.9 percent reported that tests for illicit drug or alcohol use occurred at their place of employment during the hiring process, or "prehire" testing. This equates to more than 47 million adults who worked in settings where testing for illicit drug or alcohol use occurred during the hiring process (Figure 5.1 and Table  5.1).
• The youngest and oldest adult workers (18 to 25 and 50 to 64 years) were less likely to report working for employers with prehire drug testing than workers aged 26 to 49 years. An estimated 40.8 percent of 50- to 64-year-olds and 39.0 percent of 18- to 25-year-olds reported prehire testing, compared with 44.3 percent of workers aged 26 to 34 years and 44.7 percent of workers aged 35 to 49 years (Table  5.1).
• Males reported prehire testing in their place of employment more often than females (45.8 vs. 39.1 percent) (Table  5.1).
• The prevalence of prehire testing varied by race/ethnicity. Black (58.1 percent), Native Hawaiian or Other Pacific Islander (53.5 percent), and American Indian or Alaska Native (50.8 percent) adults reported greater rates of prehire testing than Hispanic whites (40.5 percent), Asians (41.8 percent), or adults who reported two or more races (42.8 percent) (Table  5.1).
• College graduates were least likely to report working for employers with prehire testing compared with all other levels of educational attainment (35.0 percent of college graduates vs. 42.5 percent of those with less than a high school diploma, 48.5 percent of high school graduates, and 45.7 percent of those with some college) (Table  5.2).
• The lowest family income category had the lowest prevalence of prehire drug testing. An estimated 36.3 percent of workers with family incomes that were less than $20,000 reported prehire testing (vs. 44.0 percent of incomes of $20,000 to $49,999, 45.9 percent of incomes of $50,000 to $74,999, and 41.9 percent of incomes of $75,000 or more) (Table  5.2).
• An estimated 37.7 percent of residents of noncore rural counties and 41.7 percent of workers in large metropolitan statistical areas (MSAs) reported working full time at places of employment with prehire drug testing. These are significantly smaller proportions than those reported in small MSAs (45.1 percent) or micropolitan statistical areas (MiSA; 44.7 percent) (Table  5.2).
• Workers in protective service (76.2 percent) and transportation and material-moving (73.3 percent) occupational categories had the highest prevalence of reported workplace testing for illicit drug or alcohol use during the hiring process. Workers in legal occupations (14.0) were the least likely category to report workplace testing for illicit drug or alcohol use during the hiring process (Figure 5.2).

Below is a bar graph, click here for the text describing this graph.

Figure 5.2 Workplace Tests Employees for Drug or Alcohol Use during Hiring Process among Full-Time Workers Aged 18 to 64, by Major Occupational Categories: 2002-2004 Combined

• The likelihood of working in a prehire-testing environment was steadily higher as the size of an establishment increased. An estimated 19.0 percent of employees who worked in an establishment with fewer than 10 employees reported a prehire drug-testing program. This is a significantly smaller proportion of workers who reported testing than was observed in other establishment sizes. Among employees who worked for the largest establishments of 500 or more employees, 70.6 percent reported prehire testing (Table  5.12).

5.2. Prehire Testing among Full-Time Workers, by Substance Use and Dependence and Abuse

• For age groups, current illicit drug users were less likely than nonusers to report working for employers who conducted prehire drug or alcohol tests (29.4 vs. 41.3 percent of 18- to 25-year-olds, 32.0 vs. 45.8 percent of 26- to 34-year-olds, 34.2 vs. 45.5 percent of 35- to 49-year-olds, and 31.3 vs. 41.0 percent of 50- to 64-year-olds) (Figure 5.3 and Table  5.3).
• For each racial/ethnic category, current illicit drug users were less likely than nonusers to report working for employers who conducted prehire drug or alcohol tests (Figure 5.4).
• For both males and females, current illicit drug users were less likely than nonusers to report working for employers who conducted prehire drug or alcohol tests (32.7 and 30.2 vs. 47.2 and 39.7 percent, respectively) (Figure 5.5 and Table  5.3).
• For all education categories, current illicit drug users were less likely than those who did not use current illicit drugs to report working for employers who conducted prehire drug or alcohol tests (less than high school: 31.5 vs. 43.9 percent; high school graduate: 35.8 vs. 49.8 percent; some college: 33.7 vs. 46.8 percent; and college graduate: 23.7 vs. 35.7 percent) (Figure 5.6 and Table  5.4).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.3 Workplace Tests during Hiring Process among Full-Time Workers Aged 18 to 64, by Age Group and Past Month Illicit Drug Use: 2002-2004 Combined

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.4 Workplace Tests during Hiring Process among Full-Time Workers Aged 18 to 64, by Hispanic or Latino Origin and Race and Past Month Illicit Drug Use: 2002-2004 Combined

Note: Due to low precision, estimates for American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, and Asian full-time workers with past month illicit drug use are not shown. Estimates for Native Hawaiian or Other Pacific Islander full-time workers with no past month illicit drug use also are not shown.

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.5 Workplace Tests during Hiring Process among Full-Time Workers Aged 18 to 64, by Gender and Past Month Illicit Drug Use: 2002-2004 Combined

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.6 Workplace Tests during Hiring Process among Full-Time Workers Aged 18 to 64, by Education and Past Month Illicit Drug Use: 2002-2004 Combined

• For all income categories, current illicit drug users were less likely than nonusers to report working for employers who conducted prehire drug or alcohol tests: 26.2 versus 37.9 percent for workers with family incomes that were less than $20,000, 33.3 versus 45.1 percent for workers with family incomes between $20,000 and $49,999, 34.5 versus 46.9 percent for workers with family incomes between $50,000 and 74,999, and 31.5 versus 42.6 percent for workers with family incomes of $75,000 or more (Table  5.4).
• For all geographic settings, current illicit drug users were less likely to report working for employers who conducted prehire drug or alcohol tests than those with no current illicit drug use: 23.8 versus 35.4 percent for workers in the Northeast, 35.9 versus 45.3 percent for workers in the Midwest, 35.3 versus 49.3 for workers in the South, and 30.5 versus 40.9 percent for workers in the West (Table  5.4).
• For all county types, current illicit drug users were less likely than those who did not use current illicit drugs to report working for employers who conducted prehire drug or alcohol tests (Figure 5.7).
• The likelihood of working for an employer who conducted prehire drug testing was higher among older age groups than younger age groups who were past month heavy alcohol users. The pattern differed among those who were not past month heavy alcohol users. Among heavy alcohol users, 34.4 percent of 18- to 25-year-olds reported prehire testing in their employment setting, compared with 46.4 percent of 50- to 64-year-olds. Among those who did not report past month heavy alcohol use, the oldest and youngest full-time workers (18 to 25 and 50 to 64 years) were less likely to report prehire drug testing by their employer compared with 26- to 34-year-olds and 35- to 49-year-olds (Figure 5.8 and Table  5.6).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.7 Workplace Tests during Hiring Process among Full-Time Workers Aged 18 to 64, by County Type and Past Month Illicit Drug Use: 2002-2004 Combined

Note: Due to low precision, estimates for Noncore Adjacent, No Town and Noncore Rural, Not Adjacent among full-time workers with past month illicit drug use are not shown.

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.8 Workplace Tests during Hiring Process among Full-Time Workers Aged 18 to 64, by Age Group and Past Month Heavy Alcohol Use: 2002-2004 Combined

• Among workers aged 18 to 34, past month heavy alcohol use was associated with a lower likelihood of working for an employer who conducted prehire drug or alcohol testing in comparison with their same-age peers who did not report past month heavy alcohol use (Figure 5.8).
• Females were less likely than males to report working in an employment setting that conducted prehire drug or alcohol testing for both heavy alcohol users and non-heavy alcohol users. An estimated 33.1 percent of females who reported past month heavy alcohol use reported working in a prehire drug and/or alcohol testing environment, compared with nearly half of males who did so (43.5 percent). The pattern was similar among males and females who did not report past month heavy alcohol use (Table  5.6).
• There were no significant differences by race/ethnicity in the likelihood of working in a prehire-testing employment environment between those who reported past month heavy alcohol use and those who did not (Table s 5.5 and 5.6).
• Individuals with the lowest family incomes of $20,000 or less, individuals with incomes of $20,000 to $49,999, and individuals with the highest family incomes of $75,000 or more who were current heavy alcohol users were less likely to report working in a prehire-testing environment than nonusers in the same income brackets (30.8 vs. 36.9 percent of lowest family incomes, 42.5 vs. 44.1 of incomes of $20,000 to $49,999, and 40.7 vs. 42.0 percent of highest family incomes) (Table  5.7).
• For all categories of establishment size, current illicit drug users were less likely than nonusers to report working for employers who conducted prehire drug or alcohol tests: 15.2 versus 19.4 percent for establishments with fewer than 10 employees, 21.2 versus 35.0 percent for 10 to 24 employees, 34.4 versus 44.9 percent for 25 to 99 employees, 55.9 versus 62.0 percent for 100 to 499 employees, and 61.7 versus 71.2 percent for 500 or more employees. However, there was no difference for current heavy alcohol users (Figure 5.9).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.9 Workplace Tests during Hiring Process among Full-Time Workers Aged 18 to 64, by Establishment Size and Past Month Illicit Drug or Alcohol Use: 2002-2004 Combined

5.3. Random-Testing Programs among Full-Time Workers

• A total of 29.6 percent, or 32 million, of full-time workers in the United States, reported random drug testing in their current employment setting during the study period. The youngest workers (18 to 25 years) were less likely than all other age groups to report working for an employer who conducted random drug testing (27.3 vs. 29.6 percent of 26- to 34-year-olds, 30.6 percent of 35- to 49-year-olds, and 29.1 percent of 50- to 64-year-olds) (Table  5.1).
• Nearly 21 million male workers reported that they were aware of random drug testing for illicit drugs or alcohol in their workplace, while 11.3 million females reported that they knew of random drug testing. Proportionately, more males reported random drug testing than females (33.0 vs. 24.9 percent) (Table  5.1).
• There were differences in the prevalence of random drug testing reported by race/ethnicity. Among white adults, 28.3 percent reported random drug testing in the workplace. While this was comparable with the proportion of Hispanic workers reporting random testing (28.7 percent), it was significantly less than the proportion of black and American Indian or Alaska Native workers (41.9 and 48.0 percent, respectively). Only 17.4 percent of Asian workers reported random drug testing in their current place of employment (Table  5.1).
• An estimated 32.9 percent of workers with less than a high school diploma reported random testing in their current place of employment, while 35.0 percent of high school graduates and 32.1 percent of those with some college reported working for employers who tested for illicit drugs or alcohol on a random basis. College graduates were the least likely to report random drug testing compared with all other levels of educational attainment (20.4 percent) (Table  5.2).
• Among workers, residents of the South were most likely to report working in an employment setting with random testing for illicit drug or alcohol use: 36.7 percent compared with 20.4 percent in the Northeast, 27.9 percent in the Midwest, and 27.3 percent in the West (Table  5.2).
• Workers in the transportation and material-moving (62.9 percent) and protective service (61.8 percent) occupational categories were the most likely to report working for employers who conducted random testing. Workers in legal occupations and arts, design, entertainment, sports, and media occupational categories were the least likely to report working in settings in which employees were tested for illicit drug or alcohol use on a random basis (Figure 5.10).
• There were significant county type differences in the likelihood of reporting a "random-testing" work environment. Residents of large MSAs were the least likely to report random testing (26.5 percent). The proportion of workers who reported random drug testing in their place of employment ranged from 26.5 to 37.7 percent. An estimated 36.8 percent of the most rural workers (noncore, nonadjacent residents) reported random drug testing in their work setting (Figure 5.11).

Below is a bar graph, click here for the text describing this graph.

Figure 5.10 Workplace Tests Employees for Drug or Alcohol Use on a Random Basis among Full-Time Workers Aged 18 to 64, by Major Occupational Categories: 2002-2004 Combined

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.11 Workplace Tests on a Random Basis among Full-Time Workers Aged 18 to 64, by County Type: 2002-2004 Combined

• Among workers, the relationship between family income and reported employment with an employer who tested for illicit drug or alcohol use among workers with no current illicit drug use on a random basis was not linear. The highest category of family income, $75,000 or more, was associated with a lower risk of working for an employer who conducted random drug testing (27.5 percent), compared with family incomes of $20,000 to $49,999 (31.9 percent) and $50,000 to $74,999 (33.2 percent). Workers with family incomes that were less than $20,000 were somewhat less likely to report random testing (28.3 percent) than higher income groups (Figure 5.12 and Table  5.4).
• There was a steady increase in the likelihood of working in a random-testing environment as the size of an establishment increased. An estimated 14.5 percent of employees who worked in an establishment with fewer than 10 employees reported that their employer conducted random drug and alcohol testing. This is a significantly smaller proportion of workers who reported testing than was observed in other establishment sizes. Among employees who worked for the largest establishments of 500 or more employees, 42.6 percent reported random drug testing (Table  5.12).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.12 Workplace Tests on a Random Basis among Full-Time Workers Aged 18 to 64, by Family Income and Past Month Illicit Drug Use: 2002-2004 Combined

5.4. Random Testing among Full-Time Workers, by Substance Use and Dependence and Abuse

• Current illicit drug users were less likely to report working for employers who conducted random drug or alcohol tests than were nondrug users. For 18- to 25-year-olds, 19.7 percent of illicit drug users reported that they worked in a random-testing environment compared with 29.1 percent of nonusers. The relationship was consistent for all age groups: 20.0 versus 30.8 percent of 26- to 34-year-olds, 22.6 versus 31.2 percent of 35- to 49-year-olds, and 20.4 versus 29.3 percent of 50- to 64-year-olds (Figure 5.13 and Table  5.3).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.13 Workplace Tests on a Random Basis among Full-Time Workers Aged 18 to 64, by Age Group and Past Month Illicit Drug Use: 2002-2004 Combined

• Among both male and female full-time workers, past month illicit drug users were less likely to report working for an employer with a random drug-testing program compared with those who had not used illicit drugs in the past month. An estimated 18.7 percent of females and 21.9 percent of males with past month illicit drug use reported working in a random-testing employment setting, compared with 25.3 percent of females and 34.2 percent of males who were not past month users of illicit drugs (Table  5.3).
• Similarly, past month illicit drug use was associated with a reduced likelihood of working for an employer with a random-testing policy across racial/ethnic groups. An estimated 17.8 percent of Hispanic workers and 21.2 percent of non-Hispanic workers who were past month illicit drug users reported working for an employer who conducted random testing, while 29.5 percent of Hispanic and 30.6 percent of non-Hispanic workers who did not use illicit drugs in the past month reported working for an employer who tested randomly. Black workers with past month illicit drug use were significantly more likely to report working in a setting with random testing than white past month illicit drug users (29.4 vs. 20.0 percent, respectively). A similar difference existed between black and white workers who reported no past month illicit drug use (43.0 vs. 29.1 percent, respectively) (Figure 5.14 and Table  5.3).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.14 Workplace Tests on a Random Basis among Full-Time Workers Aged 18 to 64, by Hispanic or Latino Origin and Race and Past Month Illicit Drug Use: 2002-2004 Combined

• For all levels of education, current illicit drug users were less likely than nonusers to report working for employers who conducted drug or alcohol tests on a random basis: 24.0 versus 34.0 percent for those with less than a high school diploma, 23.7 versus 36.1 percent for high school graduates, 20.8 versus 33.2 percent for those with some college, and 13.9 versus 20.8 percent for college graduates (Figure 5.15 and Table  5.4).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.15 Workplace Tests on a Random Basis among Full-Time Workers Aged 18 to 64, by Education and Past Month Illicit Drug Use: 2002-2004 Combined

• The association between reported employment in a setting that tests for illicit drug or alcohol use on a random basis and past month marijuana use among full-time workers mirrored findings reported for past month illicit drug use. Within each demographic group, a smaller proportion of individuals with past month marijuana use reported working for employers who conducted random testing compared with workers who did not report current use of marijuana (Table s 5.8 through 5.11). There was no relationship between heavy alcohol use and the likelihood of working for a random-testing employer.
• Findings associated with illicit drug or alcohol dependence or abuse were consistent with those observed for current illicit drug and heavy alcohol use. Individuals who met criteria for illicit drug dependence or abuse were less likely to report working for an employer with a random drug-testing policy compared with individuals who did not meet dependence or abuse criteria (Figure 5.16).
• Alcohol disorders were associated with a lower likelihood of working in a random-testing environment among the youngest adult workers––those who did not complete high school and those who resided in large MSAs (Figure 5.16).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.16 Workplace Tests during Hiring Process or on a Random Basis among Full-Time Workers Aged 18 to 64, by Past Year Illicit Drug or Alcohol Dependence or Abuse: 2002-2004 Combined

• For all categories of establishment size, current illicit drug users were less likely than nonusers to report working for employers who conducted random drug or alcohol tests (less than 10 employees: 11.4 vs. 14.8 percent; 10 to 24 employees: 16.7 vs. 27.2 percent; 25 to 99 employees: 24.5 vs. 33.1 percent; 100 to 499 employees: 33.7 vs. 40.7 percent; and 500 or more employees: 30.2 vs. 43.4 percent). However, there was no difference for current heavy alcohol users (Figure 5.17).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 5.17 Workplace Tests on a Random Basis among Full-Time Workers Aged 18 to 64, by Establishment Size and Past Month Illicit Drug or Alcohol Use: 2002-2004 Combined

5.5. Summary

This chapter addresses the question "Are current substance users less likely to work for employers who conduct prehire or random drug testing?" It is important to understand that these data cannot inform a discussion about the cause-and-effect nature of drug testing in the workplace. Respondents are interviewed at one point in time and are not asked to provide information about testing in relation to job tenure, their own personal testing experience, or testing in previous employment settings. It also is possible that respondents may not be clear about differences in testing programs. For example, there may be some confusion about random testing in contrast to cause-based testing in response to a specific event or employee behavior. While there is strong evidence here that individuals with current or long-term illicit substance use problems are less likely to report working in an employment setting with either prehire or random drug testing, it cannot be said that drug testing caused the worker to work in one environment or another or that testing reduced the use patterns of an existing employee.

The relationship of drug testing to substance use was impacted in some of these analyses by characteristics of the respondent or the respondent's reported employment setting. Specifically, race/ethnicity appeared to influence the relationship between use status and working in a setting with a drug-testing program. Among current illicit drug users, non-Hispanic black workers were more likely to report working in a prehire- or random-testing environment when compared with non-Hispanic white workers. This may be an effect of other characteristics such as differences in occupational or educational opportunities. There also were differences among county types. This may reflect differences in employment options in cities compared with towns and rural places.

Policy makers and those who work in substance abuse programs will benefit from ongoing research in this area. It will be important to clarify an understanding about whether these programs keep substance users out of the workplace or whether these programs serve as a tool to route those who use illicit substances into healthier and more productive work lives. It also will be important to understand more clearly how demographic characteristics impact the effect of testing programs. As was noted in the earlier report by the Institute of Medicine (Normand, Lempert, & O'Brien, 1994), more could be learned by combining survey data, such as the NSDUH data used in this report, with drug test results collected in work settings.

Worker Substance Use and Workplace Policies and Programs

6. Workplace Behaviors and Attitudes toward Drug Testing

This chapter examines the work-related behaviors and attitudes toward drug testing of full-time workers in order to describe the potential impact of substance use on worker productivity and the role of drug testing in the workplace. Previous chapters in this report have provided estimates of the prevalence of illicit drug use, heavy alcohol use, and illicit drug and alcohol dependence or abuse among full-time workers aged 18 to 64 in the United States. The report also has detailed the proportion of workers in the United States reporting the presence of programs designed to identify and assist workers with substance use issues. In this final chapter, several related matters are presented:

• Workplace behaviors in association with substance use
• Attitudes held by full-time workers about their willingness to work for an employer with a drug-testing program
• Multivariate analysis of employer drug-testing and the willingness of adults to work for an employer with a drug-testing program, by substance use, dependence, and abuse

6.1. Workplace Behaviors among Full-Time Workers

The National Survey on Drug Use and Health (NSDUH) does not collect direct measures of worker productivity or on-the-job safety. However, three work-related measures that are related to productivity and safety are available:

• Worked for three or more employers in the past year
• Missed 2 or more days in the past month due to illness or injury
• Skipped 1 or more days of work in the past month

All of these indicators were significantly associated with current illicit drug use, past year illicit drug dependence or abuse, and past year alcohol dependence or abuse. Two of the three indicators (working for three or more employers in the past year and skipping 1 or more days of work in the past month) also were positively and significantly associated with past month heavy alcohol use.

• Among full-time workers who reported past month illicit drug use, 12.3 percent reported working for three or more employers in the past year, compared with 5.1 percent of workers without past month drug use (Figure 6.1).
• Workers with past month illicit drug use were more likely to report missing 2 or more workdays in the past month due to illness or injury when compared with workers without current use (16.4 vs. 11.0 percent) (Figure 6.1).
• An estimated 16.3 percent of workers who used illicit drugs in the past month reported skipping 1 or more days of work in the past month (vs. 8.2 percent of workers who did not use an illicit drug during the past month) (Figure 6.1).
• The pattern for past month marijuana use mirrored that found among all past month illicit drug users (Figure 6.1).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 6.1 Workplace Behaviors among Full-Time Workers Aged 18 to 64, by Past Month Substance Use: 2002-2004 Combined

• Among workers who reported past month heavy alcohol use, 8.9 percent reported working for three or more employers during the past year versus 5.4 percent of those who did not report past month heavy alcohol use (Figure 6.2).
• An estimated 13.6 percent of past month heavy alcohol users employed full time reported skipping work 1 or more days during the past month (vs. 8.4 percent of workers with no past month heavy alcohol use) (Figure 6.2).
• Past month heavy alcohol use was not significantly linked to missing 2 or more days of work in the past month due to illness or injury (Figure 6.2).
• The patterns of work outcomes related to dependence on or abuse of illicit drugs or alcohol were generally similar to those found among current users, although the magnitude of the relationship was greater for dependence or abuse. All three outcomes were significantly related to dependence or abuse. Thus, the more chronic conditions were associated with a greater risk for absenteeism and frequent job changes during the past year than for substance use per se.

Below is a vertical bar graph, click here for the text describing this graph.

Figure 6.2 Workplace Behaviors among Full-Time Workers Aged 18 to 64, by Past Month Heavy Alcohol Use: 2002-2004 Combined

6.2. Workplace Attitudes toward Drug or Alcohol Testing among Full-Time Workers

NSDUH respondents who were employed at the time of the interview were asked the following questions:

• Would you be more or less likely to work for an employer who tests its employees for drug use as part of the hiring process? Would you say more likely, less likely, or it would make no difference to you?
• Would you be more or less likely to work for an employer who tests its employees for drug or alcohol use on a random basis? Would you say more likely, less likely, or it would make no difference to you?

The responses to these questions were examined for full-time workers aged 18 to 64 to determine if the presence of a prehire or random drug-testing program would deter users from seeking employment at that establishment.

• Among workers in the United States, 52.5 million (46.0 percent) indicated that they would be more likely to work for an employer who tests before hiring, and an additional 56.2 million (49.1 percent) workers reported that prehire testing would not influence their decision to work for an employer. Only 5.6 million (4.9 percent) workers indicated that they would be less likely to work for an employer who conducts prehire drug testing (Table s 6.1 and 6.2 in Appendix E).
• An estimated 45.5 million (39.8 percent) workers reported that they would be more likely to work for an employer who tests randomly for drug or alcohol use, while 10.0 million (8.7 percent) workers reported that they would be less likely. An estimated 58.8 million (51.4 percent) workers indicated that random testing would not influence their decision to work for an employer (Table s 6.1 and 6.2).
• Workers reporting current illicit drug use indicated that they would be less likely to work for employers who conduct prehire and random testing than those workers who did not report current illicit drug use (Figure 6.3 and Table s 6.3 through 6.14).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 6.3 Employees' Feelings toward Working for Employers Who Test during Hiring Process among Full-Time Workers Aged 18 to 64, by Past Month Illicit Drug Use: 2002-2004 Combined

• Workers who reported past month illicit drug use indicated that they would be less likely to work for employers who conduct random testing. An estimated 29.1 percent of workers with past month illicit drug use reported that they would be less likely to work for employers who conduct drug testing randomly, while only 6.9 percent of workers who did not report past month illicit drug use selected this response category (Figure 6.4).
• These findings also held for workers who were heavy alcohol users or who were dependent on or abused illicit drugs or alcohol (Table  6.9).

Below is a vertical bar graph, click here for the text describing this graph.

Figure 6.4 Employees' Feelings toward Working for Employers Who Test on a Random Basis among Full-Time Workers Aged 18 to 64, by Past Month Illicit Drug Use: 2002-2004 Combined

6.3. Multivariate Analysis of Drug Testing in Current Employment Setting and Willingness to Work for an Employer Who Tests for Drugs

It is not possible from the analyses presented in this report to determine whether drug testing deters those who would be illicit drug users from using drugs. However, evidence presented in previous chapters confirms that those who use illicit drugs are less likely to report working for employers who have a drug-testing program compared with workers who do not use illicit drugs. In an effort to clarify some of the relationships reported in previous chapters, multivariate analyses were conducted.

Multinomial logit models were used to estimate the odds that workers would report that they were "more likely" or "less likely" to work for an employer with a prehire or random drug-testing program compared with the reference category of "it would make no difference." In the multivariate models presented here, full-time workers were categorized into four recency of illicit drug use categories:

• Never used an illicit drug
• Used illicit drug in lifetime but not in the past year
• Used illicit drug in lifetime and in the past year but not in the past month
• Used illicit drug in the past month

Workers in the last three categories were compared with those individuals who never used an illicit drug. It was expected that there would be an increased probability of responding "less likely" and a decreased probability of responding "more likely" because drug use was more recent in time relative to the survey. These models were adjusted for age, gender, race/ethnicity, educational attainment, family income, region, and county type (metropolitan statistical area). All workers with information on their willingness to work for employers with testing practices were included in these analyses.

In the multinomial logit analysis for prehire drug testing, there was a significantly higher probability of membership in the "less likely" group in comparison with the "makes no difference" group for all categories of drug user compared with those who had never used, and the odds of reporting "less likely" were higher among "more recent" than "less recent" drug users. Similarly, for "more recent" versus "less recent" drug users, there was a reduced probability of membership in the "more likely" group compared with the "would make no difference" group. A similar pattern was observed in models for random drug or alcohol testing (Table s 6-A, 6.15, and 6.16).

The potential effect of drug testing was further evaluated using three logistic regression analyses. The dependent variables for the three models were (1) working for an employer with a drug- or alcohol-testing program, (2) working for an employer with a prehire-testing program, and (3) working for an employer with a random drug- or alcohol-testing program. The independent variables were the same as the multinomial models. The first model, presence of a drug- or alcohol-testing program, included all full-time workers with valid responses to the questions:

• Does your workplace ever test its employees for alcohol use?
• Does your workplace ever test its employees for drug use?

For inclusion in the other two logistic models, individuals also must have given valid responses to two follow-up questions in the following order:

• Does your workplace test its employees for drug or alcohol use as part of the hiring process?
• Does your workplace test its employees for drug or alcohol use on a random basis?

In each model, there was a decreased probability of working for an employer with a testing program among those who had used in the past year or past month compared with those who had never used illicit drugs. Compared with workers with no lifetime use of illicit drugs, there was a decreased likelihood of working for an employer who tested for drug or alcohol use. There was not a significant difference between no lifetime use and lifetime but no past year use. Compared with workers with no lifetime use, there was a significant decrease in the probability of working for an employer with a prehire-testing program among past year–no past month and past month illicit drug users. Findings were similar for random drug testing (Table s 6-B and 6.17).

---

End Notes

¹ RTI International is a trade name of Research Triangle Institute.