Workplace Interventions for Drug Abuse: Part B
Paul D. Steele, Ph.D
University of New Mexico
Hyperlinks to sections within this text:
Drug- and Alcohol-Testing Programs
Employee Assistance Programs
Angarola, R., & Rodriguez, S. (1991). State legislation: Effects on drug programs in industry. In S. Gust & J. Walsh (Eds.), Drugs in the workplace: Research and evaluation data (Monograph 91, pp. 305-317). Rockville, MD: National Institute on Drug Abuse.
According to the authors, dramatic increases in workers use of drugs such as cocaine and the large number of companies conducting drug testing have spurred state legislators to introduce bills and enact laws aimed at regulating, and at times prohibiting, some forms of drug testing in the workplace. These state laws cover not only government employees, but private-sector workers as well. By September 1988, eight states (Connecticut, Iowa, Louisiana, Minnesota, Montana, Rhode Island, Utah, Vermont) had enacted comprehensive drug-testing laws. A number of these statutes were patterned after a model bill drafted by the American Civil Liberties Union. Governors of two states (Maine and Wisconsin) vetoed legislation that would restrict drug testing because they were concerned that these laws would impede efforts to reduce drug abuse.
Comprehensive drug-testing laws have 12 basic provisions. With the exception of Louisiana and Vermont, all states require that the employer have probable cause or reasonable suspicion to test an employee for the presence of drugs. The statutes provide no guidance on the nature of facts that constitute cause or suspicion. Five states place restrictions on testing of applicants based on conditions of equity and information. Most states allow random testing in certain circumstances, such as for safety-sensitive positions or if authorized by federal law. All states require confirmatory tests before an employee is disciplined or discharged. States require that testing laboratories be licensed or otherwise regulated. Five states require the employer to follow reliable chain-of-custody procedures. Seven states require that the employer keep the test results confidential. States also indicate that employees may provide samples in private. Four states require that employers establish employee assistance programs for the treatment and rehabilitation of employees who test positive. Five states require employers to provide workers with a written statement on the substance abuse policy and drug-testing program, and five states provide civil remedies for the employee if the employer fails to comply with statutory requirements. Four of the eight states make it a misdemeanor to violate the statute, punishable by imprisonment, a fine, or both.
State drug-testing laws are based on the constitutional dimensions of an employees privacy and freedom from unreasonable search and of fairness and due process for the employee. A large majority of government drug-testing programs have survived due process challenges. There appears to be no question legally that courts will accept the accuracy and reliability of test results when employers use appropriate procedures.
Axel, H. (1991). Characteristics of firms with drug testing programs. In S. Gust & J. Walsh (Eds.), Drugs in the workplace: Research and evaluation data (Monograph 91, pp. 219-226). Rockville, MD: National Institute on Drug Abuse.
This article reports on the results of a national survey conducted by the Conference Board in 1988 concerning the implementation of drug-testing programs and other workplace initiatives to control substance abuse in the United States. The survey was conducted among 2,675 large corporations included in the Boards sampling frame, with a 25% response rate. A clear majority of corporations reported that they had implemented, or were in the process of implementing, a drug-testing program. Seventy-five percent of the firms reporting that they test for drug use are manufacturers or utilities, and those that do not test are most likely to be banking, insurance, or other financial services. Testing companies tend to be male intensive and include a higher proportion of workers in skilled crafts, production, and labor categories. Unions are more likely to be present in the drug-testing firms. Companies that test are also more likely to have large labor forces and be located in multiple sites.
Executives in nearly two-thirds of drug-testing firms blame alcohol and illegal drugs for significant problems among their employees; only one in four reported alcohol alone as the most critical problem. By contrast, almost two-thirds of the employers without testing programs regard alcohol as the principal substance of abuse, and only one-third say they have serious problems with other drugs. Nearly half of the drug testers describe illegal drug use as a more serious problem than it was 5 years ago. Executives in companies that test for drugs say the most compelling reason for establishing a program is evidence of drug problems at the workplace.
The survey found that companies who have drug-testing programs are more likely than nontesting companies to use other drug control policies and programs, such as formal substance abuse policies and procedures, drug training and education, employee assistance programs, and security procedures.
Blank, D., & Fenton, J. (1991). Early employment testing for marijuana: Demographic and employee retention patterns. In S. Gust & J. Walsh (Eds.), Drugs in the workplace: Research and evaluation data (Monograph 91, pp. 139-150). Rockville, MD: National Institute on Drug Abuse.
This study examines a group of approximately 500 male Navy recruits inducted in 1985 who tested positive for marijuana (THC) at accession and compares them with a matched group who tested negative. At that time, Navy policy dictated that sailors found to test positive for THC be retained in the Navy while those positive for other drugs be returned home. Significant differences between the two groups were noted. The THC-positive group had lower levels of educational achievement and lower Armed Forces Qualification Test scores, and individuals were less likely to be white. No differences were found in age, marital status, or geographical home of origin. Retention data showed that 81% of the THC-negative and 57% of the THC-positive group were still in the Navy after 2.5 years. Members of the THC-positive group were much more likely to be discharged for alcohol and drug abuse or other behavioral performance reasons than were members of the THC-negative group. Sailors testing positive for marijuana at accession were divided into three groups depending on the amount of THC present in their urine. Significant differences in retention patterns for these three groups were not observed.
Blanton, E., Kidwell, R., & Bennett, N. (1992). Application of performance tests to identify workplace drug users: A panacea or a familiar set of problems? Journal of Employee Assistance Research, 1, 350-361.
Urinalysis drug-testing programs have been criticized for poor test accuracy, low test validity, and lack of fairness. Computerized performance testing has been advanced as an alternative approach. Such testing is said to have an advantage over urinalysis because it assesses fitness for duty, rather than simply the presence of drugs in the system. However, critics of this approach question its validity, reliability, and employee acceptance.
When novel policies, practices, and procedures are adopted, employees are concerned about the fairness, or procedural justice, by which outcomes and judgments are reached rather than the outcomes themselves. Thus, the authors use a procedural justice framework to compare urinalysis and performance-testing methods of detecting impaired workers. Perceived fairness is grounded in worker judgments of the accuracy, consistency, freedom from bias, correctability, opportunity for worker input, and relevance in data collection. They argue that, from this perspective, performance testing has no inherent advantage over urinalysis. Faulty implementation of either method can provoke employee perceptions of unfairness. Sound company policies and well-trained managers seem to be the requisite strategy for worker acceptance no matter which form of testing is employed.
Blum, T., Fields, D., Milne, S., & Spell, C. (1992). Workplace drug testing programs: A review of research and a survey of worksites. Journal of Employee Assistance Research, 1, 315-349.
According to the authors, the belief that drug use is prevalent among workers in the United States has influenced public- and private-sector employers to conduct drug tests of prospective and current employees. However, little research has been conducted to support the foundations of public policy that espouse organizational-sponsored drug testing. In other words, although public policy has been successful in influencing the proliferation of drug-testing programs, not much is known about the effects of these programs. This paper reviews extant research on this topic and then presents results of a study conducted at 342 work sites.
Sanctions, rehabilitative programs, and drug testing are the three major approaches used to control drug use in the workplace. Little is known about the range or application of sanctions, although a 1986 study found that only 8% of firms fire employees for testing positive for drugs. Other studies sponsored by the American Management Association found that 22% of companies immediately fire for a positive drug test, 21% suspend or put workers on probation, and 70% refer employees to treatment.
Employers generally take different policy stances on alcohol and drug use. Alcohol abuse is treated as a problem only in cases where it inhibits role relationships and/or work performance. In contrast, any presence of drugs in an employees urine is considered evidence of a drug problem in need of correction. In addition, applicants testing positive for drugs are usually excluded from employment, and current employees testing positive are often referred for treatment. One of the most prevalent rehabilitative responses to workplace drug use is an EAP. There is a growing integration of drug testing and EAP services.
Estimates of the proportion of work locations that perform some type of drug testing have increased substantially since 1986. Prevalence of testing programs is positively associated with workplace size, with most found in manufacturing companies and utilities. Random testing is the fastest growing form of testing, although applicant testing and testing for cause are more prevalent. The most common testing approach by far is urinalysis. Testing programs are implemented because of a concern for job safety, beliefs that drug use is immoral, fear that employee drug use leads to crimes and undesirable work behaviors, concern for the satisfaction of other employees, and belief that testing contributes to the companys image. In addition, the beliefs that use is prevalent enough to warrant testing and that testing is necessary to meet regulations are reasons for implementing programs. The prevalence of drug use among employees appears to be in the range of 5% to 7% nationally. Research results are mixed concerning the relationship between positive drug use and poor work habits and/or job performance. Also, there is no definitive relationship between the implementation of testing programs and the reduction of drug use. Programs have proliferated, however, due to the advent of local, state, and federal regulations mandating drug testing.
The authors report on the results of personal semistructured interviews conducted with the highest ranking human resource manager at 342 medium to large (over 200 employees) work sites in Georgia. At the time of the interview, respondents were also asked to complete a self-administered questionnaire and mail it to the researchers. The response rate was 82%. The researchers found that 78% of work sites conduct some form of drug testing, with 70% of work sites conducting applicant testing and 18% conducting random testing. The mean rate of positive test results at sites doing applicant testing was 6.6%. The mean positive rate for reasonable cause tests of current employees was 33.8%, although the median was only 10%. The mean rate of positive tests for random testing programs was 3.4%. The majority (57%) of work sites allowed applicants who tested positive to reapply for employment after some time had elapsed (usually 6 months to 1 year). Consequences for failing a drug test for current employees varied greatly, although counseling/referral was slightly more common than sanctioning. Most common problems with testing were costs, logistics, and worker opposition. The most common reason for not having a testing program was that employers believed there was no need for one.
Consistent with other research, testing programs were more prevalent in larger work sites with more male, younger, less educated, and minority employees. Drug testing was also more prevalent in rural, manufacturing, and unionized work sites, and those with employee assistance programs. Positive drug test percentages were positively associated with the proportion of black employees and unionization. It was inversely associated with educational level and presence of an EAP. Manufacturing sites had a slightly higher percentage of positive drug tests. The proportion of black employees was related to the likelihood to terminate employees and the likelihood that workplaces do not confirm test results. Work sites that tested for reasonable cause were likely to support EAPs. Substance abuse prevention activities were also more likely to occur in work sites with EAPs.
Cook, R., Bernstein, A., Arrington, T., Andrews, C., & Marshall, G. (1995). Methods for assessing drug use prevalence in the workplace: A comparison of self-report, urinalysis, and hair analysis. International Journal of the Addictions, 30, 403-426.
A random sample of 1,200 employees of a steel manufacturing plant were randomly assigned to four different self-report methods of assessing illicit drug use: individual interview in the workplace, group-administered questionnaire in the workplace, telephone interview, and individual interview off the work site. Of those selected for the study, 928 actually completed it. Urine specimens were collected and analyzed on all 928 subjects participating in the study, and hair analysis was conducted on 307 of the subjects. Although self-reports produced the highest drug use prevalence rate, analyses combining the results of the three assessment methods showed that the actual prevalence rate, measured by urine and hair analysis, was approximately 50% higher than the estimate produced by self-reports. The group-administered questionnaire condition produced prevalence rates that were roughly half those of the other self-reports. The findings cast doubt on the validity of self-reports as a means of estimating drug use prevalence and suggest the need for multiple assessment methods.
Crouch, D., Webb, D., Peterson, L., Buller, P., & Rollins, D. (1991). A critical evaluation of the Utah Power and Light Companys substance abuse management program: Absenteeism, accidents, and costs. In S. Gust & J. Walsh (Eds.), Drugs in the workplace: Research and evaluation data (Monograph 91, pp. 169-193). Rockville, MD: National Institute on Drug Abuse.
The authors report on a study to evaluate organizational costs associated with drug-involved employees and to examine safety records for trends indicative of the deterrent effect of the drug-testing program at Utah Power and Light Company. Demographically, an employee with an admitted or detected drug abuse problem had a 75% likelihood of being male. The individual had a mean age in the range of 32 to 37 years and worked as an operator, a laborer, or a craftsman. He had been employed at the company for 7 to 13 years, but earned less than the mean income for company employees. His drug of choice, as detected in 27 of 28 positives, was marijuana. Drug users were absent on the average of 8 days more than members of a matched nonusing control group of workers. Unexcused absences showed a mean difference of 45.1 hours per year. No clear differences in expenditure for medical benefits were observed. Although vehicle accidents decreased after the advent of the testing program, drug users were five times more likely to have had such an accident. When factoring savings in turnover, absenteeism, sick leave, and accidents, the authors assert that the program has saved $662,140.
Greenberg, E. (1992). 1992 AMA survey on workplace drug testing and drug abuse policies. New York: American Management Association.
This report describes the results of the American Management Associations sixth annual survey on workplace drug-testing and drug abuse policies. Survey results indicate that more companies tested more employees and applicants for controlled or illegal substances and that there is a decline in detected drug use among these groups.
Survey findings show a significant growth in corporate initiatives that deal with drug abuse. In 1987, fewer than half of the responding firms sponsored any policy or program; by January 1992, 90% of responding firms had at least one policy or program in place to combat substance abuse. Although the overall rise in drug testing has been dramatic (250% increase from 1987 to 1992), the greatest change has been in policies that determine who is tested and for what reasons. In 1987, the rule among companies with testing programs was to test only when job performance indicated the possibility of substance abuse. From 1987 to 1992, the prevalence of companies that subject current employees to periodic or random testing increased by 1,000%. Significant growth in the testing of job applicants also occurred by 1992, with over one-half of surveyed firms reporting testing of all new hires. The author attributes increases in testing to new state and local laws, regulations issued by the U.S. Departments of Transportation and Defense, the Federal Drug-Free Workplace Act, and concerns about rising employer health care costs and liability.
Hartwell, T., Steele, P., & Rodman, N. (in press). Prevalence of drug and alcohol testing in the workplace. Monthly Labor Review.
This study presents the results of a recent national probability survey of 3,200 work sites conducted in 1992-1993 that was used to estimate the prevalence of drug- and alcohol-testing programs by work site size, industry, and census region. Overall, it was found that 48.4% of work sites with 50 or more full-time employees test for drugs and 23% test for alcohol. Compared to the 1988 Bureau of Labor Statistics survey that used a similar methodology, this represents a dramatic increase of 32% in 4-5 years. Prevalence of testing programs was positively associated with work site size. Drug and alcohol testing was more prevalent in the South. It was also more prevalent in the communications/utilities/transportation (72.4%) and manufacturing firms (60.2%) than in finance/realty/insurance (22.6%) and services (27.9%) firms. Testing programs are also more likely in unionized work sites with a high proportion of full-time employees. Testing programs are less likely to occur in work sites with older, highly educated employees. Ethnic composition of the work sites labor force was not significantly associated with presence of a drug-testing program. Work sites with drug-testing programs are also more likely to have a written policy on alcohol and drug use (96.0%), have an employee assistance program (45.9%), and be subject to the Federal Drug-Free Workplace Act (86.8%). Less than 15% of work sites where employees are subject to testing actually conduct these tests on a regular basis, and less than 5% test after treatment. In contrast, approximately 47% test on a random basis, 37% test for cause, and 26% test after an accident. The percentage of random testing decreases with workplace size and is highest in the communications/utilities/transportation industries. Testing is conducted primarily by contractors, although very large work sites (over 1,000 employees) are slightly more likely to test in their own medical or personnel departments.
McDaniel, M. (1991). Does pre-employment drug use predict on-the-job suitability? In S. Gust & J. Walsh (Eds.), Drugs in the workplace: Research and evaluation data (Monograph 91, pp. 151-167). Rockville, MD: National Institute on Drug Abuse.
Drug testing is increasingly used in the screening of applicants for employment. Despite the growth of drug testing, there is little research that examines the value of preemployment drug use information in the prediction of postemployment suitability.
This research examined the criterion-related validity of preemployment drug use information. Drug use items were included in the militarys Educational and Biographical Information Survey. During the spring of 1983, the survey was administered to approximately 34,000 applicants for the military services. Those who entered the military within 1 year of completing the survey were defined as the study sample (N = 10,188). Employment unsuitability was defined as discharge from military service for reasons classified as failure to meet minimum behavioral or performance criteria on or before September 30, 1987.
For all drugs examined, the greater the frequency of use and the earlier the age at which the drug is first used, the greater the probability of a person being classified as unsuitable after hire. Those who use drugs more frequently than the rates at which they are used by the general population are at much greater risk of being judged unsuitable after hire. Although those who report substantial use are more likely to be discharged for unsuitability, the base rate for drugs, except marijuana, is low. These low base rates contribute to the low predictive validity of the drug measures. In this sample, marijuana has a moderately high base rate (31-38%), yet its validity is low (.07). Used alone as a predictor of suitability, self-reported marijuana use has positive utility but may be less useful than other predictors of unsuitable employee behavior.
The limited operational validity of preemployment drug use measures found in this study suggests that employers who presently rely solely on drug use measures for screening applicants for suitability should consider supplementing or replacing their drug-screening programs with selection systems that more optimally predict employee unsuitability. For example, the predictive power of the high school graduation dichotomy is higher than the drug use measures found in this study. Typically, the discharge rate for non-high school graduates is roughly twice that of graduates. Also, research studies on worker reliability measures show high levels of validity. Such measures could prove to be superior to drug use measures because they tap a wider range of background and personal characteristics predictive of employment unsuitability.
Murphy, K., & Thornton, G. (1992). Development and validation of a measure of attitudes toward employee drug testing. Educational and Psychological Measurement, 52, 189-201.
The authors developed a 19-item scale measuring attitudes toward employee drug testing based on responses to actual drug-testing policies and practices. Initial items for analysis were based on a survey of literature on the topic produced between 1980 and 1989 and corresponded closely to the categories of who was tested, administrative procedures, program management, and consequences for test failure. Two studies were conducted to develop a measure of attitudes. First, 279 undergraduate students were given a list of 57 characteristics of employee drug-testing programs to select response dimensions and to estimate scale values for each item. Second, two separate mailed surveys were completed by students to develop an initial and a final form of the scale and to assess both the reliability and construct validity of the measure. The initial survey was completed by college seniors, and the final survey was completed by students over 23 years of age. The obtained coefficient alpha for this scale is .90. Results of a maximum likelihood confirmatory factor analysis indicate a four-factor structure, one general and three content-oriented factors (who is tested, administration and management, and consequences of test failure). Evidence of criterion-related validity and discriminant validity is offered.
Murphy, K., Thornton, G., & Reynolds, D. (1990). College students attitudes toward employee drug testing programs. Personnel Psychology, 43, 615-631.
Little is known about attitudes toward various aspects of common testing programs. This study examines college students attitudes toward several aspects of drug-testing programs. Three separate samples of students (190 introductory psychology students, a second group of 80 introductory psychology students, and 101 seniors who were mostly business majors) were included. Questionnaires were not identical in topic and format. Responses for identical items were quite similar across all sample groups. Results indicate that testing is seen, under some circumstances, as appropriate and necessary. However, attitudes toward most aspects of testing are characterized by extreme variability. Virtually all aspects of drug-testing programs are strongly accepted by some individuals and strongly rejected by others. Furthermore, attitudes toward drug testing do not appear to vary as a function of employment experience or qualifications or political inclinations, and are not strongly related to the individuals exposure to others drug use. One of the few consistent correlates of approval of employee drug testing was the individuals drug use: Self-report frequency of drug use was negatively associated with the acceptability of employee drug testing.
Normand, J., Lempert, R., & OBrien, C. (Eds.). (1994). Impact of drug testing programs on productivity. Under the influence? Drugs and the American work force (pp. 215-240). Washington, DC: National Academy Press.
This chapter discusses the impact of drug-testing programs on worker productivity, on the attitudes of applicants and workers toward such programs, and on the potential effects of these attitudes on productivity. Most drug-testing programs do not include alcohol among drugs to be tested. Drug-testing programs can be grouped into three types: preemployment testing, incident-driven (or for-cause) testing, and postemployment testing without specific cause (such as random testing). The prevalence of drug-testing programs has increased rapidly, particularly in larger work sites. Preemployment testing is the most common form of testing. Although managers and executives believe that testing programs are effective tools for improving workplace safety, health, and productivity, there is little empirical evidence pertaining to their efficacy. Rather, the main impetus for drug-testing programs seems to be the attention given by the Reagan and Bush administrations to their "war on drugs" policies, from which the "drug-free workplace" concept emerged. Publicity about tragic accidents and government regulations, coupled with favorable court rulings, also have stimulated the proliferation of testing programs.
Preemployment testing research, although hampered by methodological limitations, consistently indicates that those who test positive for drug use prior to employment demonstrate higher rates of absenteeism, turnover, accidents, and disciplinary actions than applicants with negative test results. However, the cost-effectiveness of these programs has not been adequately established. Individuals tested for cause were also found to have used more medical service benefits and had higher rates of absenteeism and accidents than members of control groups. The authors caution that few studies have been done concerning the impact of for-cause testing, and those that do exist suffer from serious methodological flaws. No studies assessing the impact of random testing programs on productivity were found. Generally, their effect is inferred by the reduction of positive test results over time after the implementation of the program, although this assumption is methodologically quite questionable.
Employee drug testing continues to be controversial, and many researchers have found that applicants and employees react negatively to these programs. Drug testing may affect an individuals job search, job choice, and subsequent satisfaction with the job and the organization. There is greater acceptance of drug-testing programs, however, for safety-sensitive jobs. Programs that are seen as reasonable, not overly punitive, and treatment oriented are more acceptable to employees and applicants.
Normand, J., Salyards, S., & Mahoney, J. (1990). An evaluation of preemployment drug testing. Journal of Applied Psychology, 75, 629-639.
In this article, the authors report on the results of drug testing of 5,465 applicants. A positive drug test prevalence rate of 9% was reportedsomewhat lower than other surveys of applicants. The authors attribute lower detection to the studys restrictive and exclusive operational definition of drug use and the overall Postal Service selection process. Of those testing positive, 68% were detected as marijuana users, 23% tested positive for cocaine, and 9% tested positive for one or more of other drugs. Odds of testing positive were higher for blacks, men, and people in the 25-35 age range.
The authors also report on the relationships between testing results and absenteeism, turnover, injuries, and accidents on the job. After an average of 1.3 years of employment, those who had tested positive for illicit drugs as an applicant had an absenteeism rate 59.3% higher than employees who had tested negative (6.63% versus 4.16% of scheduled work hours, respectively). Employees who had tested positive also had a 47% higher rate of involuntary turnover (15.41% versus 10.51%, respectively). No significant associations were found between drug test results and measures of injury and accident occurrence. A practical implication of these results, according to the authors, is that implementation of a preemployment drug-testing program would not have an adverse effect on hiring policies and rates. The authors estimate that implementing a drug-testing program could save the Postal Service approximately $4,000,000 in undiscounted and unadjusted productivity costs in the first year and that the accrued savings for one cohort on new employees over their tenure would be about $52,750,000.
Sheridan, J., & Winkler, H. (1991). An evaluation of drug testing in the workplace. In S. Gust & J. Walsh (Eds.), Drugs in the workplace: Research and evaluation data (Monograph 91, pp. 195-215). Rockville, MD: National Institute on Drug Abuse.
The authors report the results of a 2-year study, funded by NIDA, intended to identify and validate workplace behaviors that are clear indicators of drug use. The study was conducted at Georgia Power, a major utility company in the Southeast. According to the authors, drug use indicators can be used to estimate drug use prevalence in the workplace and to evaluate Georgia Powers drug-testing program.
Information for the construction of social indicators and assessment available at Georgia Power includes records of drug education, supervisor training, drug testing, and employee assistance programs. An integrated database has been constructed by joining automated and manual data sets on approximately 15,000 employees. The database includes demographic information, work status and employment history, pay history, payroll deductions, insurance files, education files, data concerning dependents, human resource records, accident files, EAP admissions files, drug test results, and health insurance claims. Initial studies have been completed on a database consisting of 5 years of information.
Four types of studies are underway using the integrated database available at Georgia Power: identification of workplace behaviors and measures that are associated with drug use, evaluation of the impact of drug testing, estimation of the use of drugs in the Georgia Power work force, and evaluation of the costs and benefits of drug testing. The integrated database has been used to identify problem employees. These are workers who tested positive for drug use, were tested for cause and tested negative, entered the EAP for drug and alcohol problems, entered the EAP for other problems, and/or were discharged for problems other than drug and alcohol use.
The authors report that 13.4% of all employees in 1986 and 14.8% in 1987 tested positive for drug use. Certain job classification groups were overrepresented in the testing group (particularly semiskilled laborers), possibly due to policies and regulations. The highest positive test rates were found among technicians (38.9%) and skilled labor (25.0%). Because the authors do not distinguish between types of testing programs (i.e., random, for cause), it could be that behaviors indicative of use in these groups are more easily detected, resulting in tests for suspicion more often than with other job categories. Younger and less senior employees are more likely to have tested positive. Those who tested positive were less likely to participate in savings plans and changed jobs more often than the entire work force or those who tested negative for drug use. Employees tested for drugs, regardless of result, had higher absenteeism and docked time (i.e., nonpaid sickness, personal, and disciplinary suspension) rates than the average for the entire work force.
Taggart, R. W. (1991). Results of the drug testing program at Southern Pacific Railroad. In S. Gust & J. Walsh (Eds.), Drugs in the workplace: Research and evaluation data (Monograph 91, pp. 97-108). Rockville, MD: National Institute on Drug Abuse.
The author describes a drug-testing program implemented at Southern Pacific Railroad in 1984 to identify violators of Rule G, a policy that forbids employees from reporting to work with drugs in their system. This rule was not sufficient in itself to deter working under the influence. The study of alcohol use by railroad employees on the job conducted by the U.S. Department of Transportation in 1979 found that 75% of employees consumed alcohol, 25% were problem drinkers, 12% drank on the job, and 10% drank while subject to duty. In the 1970s, the Federal Railroad Administration attributed 48 major train accidents, 37 deaths, 80 injuries, and $34 million in damages to alcohol or drug usean estimate that the author believes is conservative. For several years, the railroad industry tested for reasonable suspicion and after serious accidents. Labor opposed testing vigorously, but moderated its stance in the 1980s.
The Southern Pacific program tested for both alcohol and illicit substances. Initially, the program was directed to operating personnel (i.e., engineers, brakemen, firemen, and switchmen). In addition, it tested all applicants in cases of reasonable suspicion, in cases of minor accidents if the workers action or inaction may have contributed to the accident or its severity, and in the case of serious accidents, irrespective of obvious cause. Subsequently, operating management voluntarily submitted to testing.
Percentage of positive test results dropped from 22.9% in 1984 to 5.8% in 1987. These results are not desegregated by occupational group or criterion for testing, although the author indicates that fewer than 5% of tests were for observation of physical symptoms or other apparent work standards violations. In 1985, members of the Engineering Department were included in the testing program. Percent positives dropped from 12% in 1985 to 5.1% in 1987. Similar results occurred when the Mechanical Department was added (16.1% positive in 1985; 3.6% positive in 1987).
Personal injuries dropped following the advent of the testing program, from 2,234 in 1983 (baseline year prior to program initiation) to 322 in 1988, per 200,000 person/hours worked. Railroad accidents were reduced from 22.2 human factor train accidents per 1 million train miles in 1983 to 2.2 human factor accidents per 1 million train miles in 1987.
Employees who test positive are given the opportunity to participate in a drug rehabilitation program. Employees may return to work on a conditional basis if they successfully complete a rehabilitation program as agreed to with the EAP and if they agree to return without compensation for time lost, be placed on probation for 2 years with immediate dismissal if the probation is violated, and submit to random, unannounced drug tests. At the end of 2 years, the EAP recommends the continuation or termination of probation. Thirty-nine percent of those who have successfully completed the treatment phase and returned to work have tested positivealmost all in their first year of probation.
Trice, H., & Steele, P. (1995). Impairment testing: Issues and convergence with employee assistance programs. Journal of Drug Issues, 25, 471-503.
Drug testing is an increasingly popular corporate response to substance-using employees. It has, however, some shortcomings as a means for identifying and controlling substance use. For example, the linkage between positive drug tests and impairment of job performance is unclear. In this article, the authors describe and discuss drug-testing policies and procedures and conclude that drug testing became popular in the 1980s largely as a workplace response to growing public concern and conservative response to drug abuse, the governments support of testing programs, organizational imitation, and the ritualistic role that drug testing plays in workplace organizations, rather than an actual increase in the prevalence of substance use. Issues related to the implementation of performance testing as an alternative to drug testing are discussed. Performance testing attempts to detect impairment more directly through performance of samples, simulations, or representations of the actual job tasks performed. Finally, the authors consider the areas of convergence between drug and performance testing and employee assistance programs. For example, positive drug tests can provide objective data of substance misuse, and workers who test positive for drug abuse can be referred to the EAP for assessment and counseling.