*pact of employee assistance programs on substance abusers and workplace safety Karen Elliott Kyna Shelley

Businesses have dealt with substance abuse in different ways. Some organiza- tions have established Employee Assistance Programs (EAPs) to address these problems. One large national company chose to fire employees with positive drug screens, offer them EAP services, and then consider them for rehire after treat- ment. A study of performance records for 12,167 employees with safety incidents revealed that rehired employees had a significantly higher incident rate than the company’s general population. Results indicated no difference in pre- and post- EAP incident rates for rehired workers, and the post-EAP incident rate fluctuated for 2 years. Implications of hiringlrehiring individuals who abuse drugs and alcohol are discussed, and suggestions are made for future research. ....

Organizations want their employees to work hard, show initiative, and be produc- tive day after day, year after year. These objectives can be met only when the em- ployer recognizes that employees are individuals with personal lives and problems. Some U.S. businesses have realized the importance of intervention and have estab- lished Employee Assistance Programs (EAPs) to address substance abuse and psy- chological problems of their employees. Early detection and treatment of problems may benefit both the employee and the employer. Employees are better able to balance their work and personal lives, and the company may retain a valued worker, enhancing the workplace environment and the organization’s value to the general public (Rotarius, Liberman, & Liberman, 2000).

HISTORY OFTHE EAP

The origin of employee assistance can be traced to the founding in 1935 of Alco- holics Anonymous (AA), an organization that fostered the concept of alcoholism as a disease and promoted a long-term treatment for recovery (Riley & Zaccaro, 1987). By the 1940s, several major corporations were actively promoting helping relation- ships between alcoholic employees and AA members (Trice & Sonnenstuhl, 1985).

a m . .

Karen Elliott, Department of Human Sciences and Humanities, University of Houston-Clear lake; Kyna Shelley, Department of Educational Leadership and Research. University of Southern Mississippi. Karen €//iott is now with the Academic Success Team, College of the Mainland. Correspondence concerning this article should be addressed to Karen Elliott, 1200 Amburn Road, Texas City, TX 77591 (e-mail: ke/liotf@com.edu).

journal of employment counseling September2005 Volume 42 125

Because of increased interest in dealing with deteriorating job performance during the 1960s and 1970s, EAPs expanded their services to cover other problems that employees might have. In 1979, 57% of Fortune 500 companies had some type of program that provided assistance to employees with alcohol problems (Normand, Lempert, & O’Brien, 1994), and by 1998, an estimated 48% of companies with more than 100 employees and 15% of small businesses had EAPs (Stieber, 2000).

EAP ACTIVITIES AND SERVICES

Although EAPs are not all the same, all EAPs provide the same essential functions: direct service delivery to employees (e.g., treatment, referral services, follow-up) and system-maintenance activities (e.g., program evaluation, training; Erfurt & Foote, 1977). A company may establish a n EAP or may contract with an outside provider for EAP services. Some programs address substance abuse only, whereas others assist with any personal concerns. No other system outside the family or legal sys- tem exists that has this kind of ongoing contact that allows a long-term relationship (Normand et al., 1994).

HISTORY OF SUBSTANCE ABUSE AND THE WORKPLACE

Alcoholism in the 1970s

Alcoholism was presumed to be the most prevalent problem in organizations and was reported to cause deterioration in job performance, which can be reflected in absenteeism, sporadic and reduced production, poor decision making, and low morale of coworkers (Erfurt & Foote, 1977). Presnall (1976) estimated that 35% of employee problems identified by deteriorating job performance were alcohol related, with a n additional 10% linked to other drug use. These figures produced a rate of alcoholism in the workforce of 8.7%, which is higher than the National Council on Alcoholism’s 1968 rate of 5.3%.

Drugs and Alcohol in the 1980s

Substance abuse escalated throughout the 1980s. Bernstein and Mahoney (1989) reported that 47% of industrial injuries and approximately 40% of industrial fatali- ties were linked to alcohol. In addition, users of illegal drugs were 1.7 times more likely to be involved in workplace injuries than nonusers and 5 times more likely to have a vehicle accident than nonusers (Crouch, Webb, Peterson, Buller, & Rollins, 1989; Hingson, Lederman, & Walsh, 1985).

Drugs and Alcohol in the 1990s

The National Clearinghouse for Alcohol and Drug Information estimated that by 1998, 12.8 million people used illegal drugs, 32 million were binge drinkers, and

126 journal of employment counseling Sepiember2005 Volume 42

11 million were heavy drinkers (Bryan, 1998). Workplace problems that are exac- erbated by drug/alcohol abuse include tardiness, absenteeism, turnover, accidents, decreased productivity, crime, and violence (Bahls, 1998). Individuals who abused drugs and alcohol had 4 times as many accidents as nonusers (Goldstein, 1997), and Bahls found that employees who abused substances used 16 times as many health care benefits and were 6 times more likely to file a workers’ compensation claim than nonabusers.

Substance Abuse in Today’s Workforce

Atkinson (2001) indicated that 12% of the workforce reported being heavy drinkers and that 47% of industrial injuries and 40% of workplace deaths were linked to alcohol consumption. Almost 14 million Americans use illegal drugs; as workers, they are 3.6 times more likely to be involved in an accident at work and 5 times more likely to file for workers’ compensation benefits than nonusers (Nighswonger, 2000). In 2002, nearly 15 million adults had alcohol-related problems (Substance Abuse and Mental Health Services Administration, 2002), ranging from missing a day of work to serious acci- dents, and approximately 100,000 American lives are lost each year to the effects of alcohol use, either through diseases or accidents (Goplerud & Cimons, 2002).

PURPOSE OF THE STUDY

The present study examined employees who abused drugs and alcohol for the pur- pose of evaluating reportable safety incidents (i.e., any event resulting in a loss or damage) and risk in the workplace. Three research questions were considered. First, work-related safety incident rates of return-to-work employees (former employees with positive drug screens who were rehired after EAP intervention) and the company’s general population (individuals without positive drug screens) were compared to test the hypothesis that the incident rate of the return-to-work employees would be higher than the rate of the company’s general population. Among the items examined were recordable (as defined by the Occupational Safety and Health Organization [OSHA]) and multiple incidents, including lost-time, medical, restricted, vehicle, and fatality cases. Second, safety incident rates of return-to-work employees were compared before and after EAP intervention. It was hypothesized that the safety incident rate would be lower after EAP services. Finally, post-EAP incident rates were analyzed to test the hypothesis that the incident rate would decrease after returning to work.

METHOD

Participants

Participants in the study were 12,167 employees whose records showed at least one reportable safety incident: 12,092 in the company’s general population with no positive drug tests (random or postaccident), and 75 who had tested positive for drugs and/or alcohol (random or postaccident), had been fired, and had completed EAP requirements to return to work.

journd of employment counseling Septrmber2005 Volume 42 127

Procedure

Safety incident records for all employees were obtained through the company's Busi- ness Objects database. Drug test results and work status were obtained through the company's MedComp database and EAP. The data encompassed all events occurring between January 2000 and December 2002. Data were analyzed with a z test to exam- ine the difference between proportions of two independent samples (return-to-work employee and general population groups; pre- and post-EAP incidents).

Safety incidents were classified into categories based on type and whether or not the case was recordable. A recordable case is defined by OSHA criteria and includes lost-time, medical, restricted, vehicle, and fatality cases. Lost-time cases refer to employees who are not able to return to work the day after an injury or illness resulting from an incident. A medical case indicates that the employee required treatment from a physician following a safety incident. A restricted case is one in which the employee is able to come back to work after a safety incident with restrictionslaccommodations. A vehicle case is one in which an employee is driving or operating a vehicle when an accident occurs. Multiple incidents include any combination of categories. Incident rates for each group were obtained in each category.

RESULTS

Recordable Cases

A comparison of safety incident rates of the return-to-work and general population employees, including total and multiple incidents, are presented in Table 1. The over- all rate includes all safety incidents (recordable and reportable). The recordable rate includes recordable cases only. The total incidents designation includes employees with one or more safety incidents, whereas the multiple incidents designation includes employees with more than one safety incident. Results indicated that return-to-work

TABLE 1

Safety Incident Rate Comparisons for Return-to-Work (RTW) and General Population (GP) Employees

Total Incidents Multiple Incidents

Category RTW GP RTW GP Overall rate" 22.4* 18.0' 6.5" 0.53** Recordable cases 6.2" 6.6** 0.59 0.44

Lost-time cases 3.2* 2.1' 0.29' 0.06* Medical cases 2.6 2.6 0.29' 0.07' Restricted cases 1.1 2.1 0.00 0.05 Vehicle cases 0.29 0.19 0.00 0.02 Fatality cases 0.00 0.02 0.00 0.00

"Includes all safety incidents (recordable and reportable). ' p < .05. " p c .01.

128 journd of employment counseling SeplemberZOOS Volume 42

employees had a significantly higher incident rate overall (z = 2.00. p < .05) than the general population group. In contrast, return-to-work employees had a signifi- cantly lower incident rate of recordable cases (z = 6.16, p < .01) than did the gen- eral population group. A separate examination of recordable cases showed a difference between groups for lost-time cases only; return-to-work employees had a signifi- cantly higher incident rate for lost-time cases (z = 1.87, p < .05) than did the gen- eral population group. Analysis revealed that the return-to-work employees had a significantly higher incident rate of multiple cases (z = 14.89, p < .01) and of lost- time (z = 1.87, p < .05) and medical cases (z = 1.76, p < .05) than did the general population group.

Pre- and Post-EAP Incidents

Investigation of the records of return-to-work employees revealed no difference in incident rates before and after treatment overall or in separate categories. Analysis of post-EAP incidents revealed that a significantly higher number of incidents occurred in the first 3 months after returning to work (z = 1 . 7 1 , ~ < .05) than at any other time after rehire. However, after decreasing and remaining low for 18 months, the post-EAP incident rate increased at 22 months to its second highest level.

DISCUSSION

Interpretation of Results

The results indicate that between 2000 and 2002, the return-to-work employee group had a lower rate of recordable incidents, overall, than did the general popu- lation group. This is inconsistent with previous literature (Goldstein, 1997) that has suggested that individuals who abuse drugs and alcohol have 4 times as many accidents in the workplace as nonusers. Analysis of specific types of recordable cases showed that whereas return-to-work employees had a higher rate of lost-time incidents than did the general population group, which is consistent with previous literature (Bross, Pace, & Cronin, 1992; Hoffman, Brittingham, & Larison, 1996), they had a lower fatality rate than did the general population group. In addition, the multiple-case incident rate was higher for the return-to-work employee group than for the general population group in two categories. These findings indicate that both individuals who abuse substances and nonusers are likely to have multiple incidents, and the type of incidents will vary.

The results of the analysis indicated that there were no differences in the incident rates, in all categories, of the return-to-work employees before and after EAP inter- ventions, suggesting that rehired employees are likely to have workplace accidents even after treatment is provided. Unfortunately, personal counseling goals are often at odds with organizational goals. As a result, EAP services may have little impact on individuals who abuse substances and on associated safety incidents. Simply provid- ing EAP services does not guarantee resolution of problems, decreased workplace

journal of employment counseling September 2005 Volume 42 129

accidents, or increased productivity. Each program must be designed to reflect the specific environment of the organization and the needs of' employees to increase the likelihood of recovery and success.

After treatment, return-to-work employees had the highest incident rate in the first 3 months after returning to work. Incidents decreased after 3 months but then increased at about 2 years. This is inconsistent with previous literature (Ryan, Zwerling, & Jones, 1992) that reported a decline in adverse outcomes after 1 year. The fluctuating incident rate after returning to the job suggests that individuals who abuse substances may be unpredictable and may have workplace accidents regardless of intervention and follow-up services.

The results obtained in this study might be influenced by the sample size of the return-to-work employee and general population groups. Even though the number of return-to-work employees is very small compared with the general population group, results do not support the current literature that states that there is a differ- ence between these groups in all categories. Early findings (Crouch et al., 1989; Goldstein, 1997; Nighswonger, 2000) have suggested that individuals who abuse substances have significantly more workplace injuries, vehicle accidents, and fa- talities than do nonusers. These results were not found in this study. No real differ- ences between groups were found in the above-mentioned categories, resulting in the argument that substance abuse might not be as significant a factor as once believed when analyzing accidents in the workplace.

Supporters of hiring and rehiring an individual who has abused substances could use the results of this study to make the following arguments in favor of this decision: (a) Return-to-work employees had a lower rate of recordable incidents than did the general population group, and (b) return-to-work employees had a lower rate of fatal- ity cases than did the general population. On the other hand, people opposed to hiring or rehiring an individual who abused substances could counter with the fol- lowing information: (a) Return-to-work employees had a significantly higher overall incident rate than did the general population group; (b) return-to-work employees had a significantly higher overall rate of lost-time cases than did the general popula- tion; (c) return-to-work employees had a significantly higher rate of multiple lost- time and multiple medical cases than the did the general population group; (d) after returning to work, return-to-work employees continued to have workplace accidents after EAP interventions for substance abuse; (e) return-to-work employees had a higher rate of post-EAP incidents in the first 3 months after returning to the job; and (f) the post-EAP incident rate for return-to-work employees fluctuated for 22 months and then increased to its second highest level.

As shown, arguments can be made to support each decision. However, the results in this study appear to oppose hiring or rehiring individuals who abuse substances. An important issue to consider when faced with this decision is risk. Drugs and alcohol alter mood and behavior (Bennett, 2003; Normand et al., 1994), and the result can be an unpredictable individual. The organization needs to consider if it is willing to employ an individual known to use drugs and/or alcohol and one who may then be a safety risk to self and others as well as a risk to productivity. Determining

130 journal of employment counseling September 2005 Volume 42

which decision should be made is difficult and cannot be done based on just one study. More research in this area is needed before trying to answer this question.

Limitations

This study is not without limitations. For example, it examined only the safety incidents for employees with positive drug andor alcohol tests and excluded other factors that might explain the incidents. Results of drug tests can be misleading, because they determine whether an individual has used drugs and/or alcohol and not current use or level of impair- ment. As a result, a positive drug test after an incident does not necessarily mean drug use was responsible for the incident, because it is possible that the substance was ingested weeks before the test andlor incident and the cause of the incident is some other factor.

Another limitation is that EAP clients with problems that were not drug or alco- hol related (e.g., depression, marital conflict) were excluded from the study. Many employers do not view individuals with psychological or other problems as a safety risk to self or others. As a result, few studies examine other factors and their rela- tionship to workplace accidents (Macdonald, 1995); therefore, many people who may be a risk are overlooked when examining safety incidents.

Finally, this study examined employees who received services through the EAP after testing positive for drugs and/or alcohol after a safety incident or during a random screening, but the study did not include employees who self-reported a substance abuse problem but had not tested positive for drug and/or alcohol use. These individuals were asked to obtain treatment for substance abuse, but they were not terminated. Some of these employees may have had useful information that could have enhanced this study, but they were excluded because of their self-disclosure. These individuals were con- sidered part of the general population group, but it is possible that they had incidents considered recordable by OSHA criteria and would have been part of the return-to- work employee group if they had not self-disclosed. Furthermore, it is possible that the individuals who self-disclosed may have been the only individuals with safety inci- dents in the general population group. The results may have been different and more accurate if the safety incidents of all individuals who abused substances were included.

Future Research

This study raises several issues for future research. First, there is a need for additional work on the relationship between substance abuse and recordable incidents in the work- place. It is easy to assume that substance abuse leads to safety incidents. Other problems, however, may also be factors and should not be overlooked. Examining employees who abuse substances, workers with other problems, and employees before and after treat- ment may bring some clarity to factors that contribute to safety incidents at work.

A second direction for future research is to compare safety records of employees who receive EAP services with the safety records of employees who receive services outside the EM. This could provide useful information and one way for organizations to measure the effectiveness of their EAPs.

journal of employment COUnSehg September2005 Volume 42 131

Finally, it would be valuable for future research to address employees’ overall work per- formance before and after EAP services instead of assessing just one factor (i-e., drug test results) when examining safety incidents and risk. Trying to explain an incident and/or determine risk based on only one factor may lead to incorrect assumptions, inconclusive results, and inappropriate actions by the organization. Research including all EAP clients, all identified problems, and a complete work history would allow a more precise and de- tailed understanding of safety incidents and risk among employees in organizations.

REFERENCES

Atkinson, W. (2001). EAPs: Investments, not costs. Textile World, 151(5), 42-44. Bahls, J. E. (1998). Drugs in the workplace. HR Magazine, 43(2), 81-87. Bennett, J. B. (2003). Introduction. In J. B. Bennett & W. E. K. Lehman (Eds.), Preventing workplace substance

abuse: Beyond drug testing to wellmv (pp. 3-28). Washington, DC American Psychological Association. Bernstein, M., & Mahoney, J. J. (1989). Management perspectives on alcoholism: The employer’s

stake in alcoholism treatment. Occupationul Medicine, 4(2), 223-232. Brass, M. H., Pace, S. K., & Cronin, J. K. (1992). Chemical dependence: Analysis of work absentee-

ism and associated medical illness. Journal of Occupationnl Medicine, 34, 16-19. Bryan, L. A., Jr. (1998). Drug testing in the workplace. Professional Safety, 43(10), 28-32. Crouch, D., Webb, D., Peterson, L., Buller, P., & Rollins, D. (1989). A critical evaluation of the Utah

Power and Light Company’s substance abuse management program: Absenteeism, accidents, and costs. In S. W. Gust & J. M. Walsh (Eds.), Drugs in the workplace: Research and eualuution data (NIDA Research Monograph No. 91). Washington, DC: US. Government Printing Office.

Erfurt, J. C., & Foote, A. (1977). Occupational empluyee assistameprogram for substance abuse and mental health problems. Ann Arbor: The University of Michigan, Institute of Labor and Industrial Relations.

Goldstein, T. F. (1997). Employee assistance programs. Journal of Compensation &Benefits, 13(2), 23-26. Goplerud, E., & Cimons, M. (2002). Workplace solutions: Treating alcohol problems through employ-

ment-based health insurance. Medical Benefits, 20(7), 6. Hingson, R., Lederman, R., & Walsh, D. (1985). Employee drinking patterns and accidental injury:

A study of four New England states. Journal of Studies on Alcohol, 46, 298-303. Hoffman, J. P., Brittingham, A., & Larison, C. (1996). Drug abuse among U S . workers: Prevalence ond

trends by occupution und industry categories. Rockville, MD: Substance Abuse and Mental Health Services Administration.

Macdonald, S. (1995). The role of drugs in workplace injuries: Is drug testing appropriate? Journal of Drug Issues, 25, 703-723.

National Council on Alcoholism. (1968). Prevalence of alcoholism among employees. New York Author. Nighswonger, T. (2000). Just say yes to preventing substance abuse. Occupationr~l Huzurds, 62(4), 3 9 4 2 . Normand, J., Lempert, R. O., & O’Brien, C. P. (Eds.). (1994). Under the influence? Drugs and the

Americun workforce. Washington, DC: National Academy Press. Presnall, L. P. (1976, April 1). fresentution to a meeting ofthe Research Committee of the Greater Detroit

area Chapter of the Association of Labor-Management Administrators and Consultants on alcoholism Riley, A. W., & Zaccaro, S. J. (Eds.). (1987). Occupationul stres.s und organizational effectiveness. New

York: Praeger. Rotarius, T., Liberman, A,, & Libeman, J. S. (2000). Employee assistance programs: A prevention and

treatment prescription for problems in health care organizations. Health Care Munuger, 19(1), 24-31. Ryan, J., Zwerling, C., & Jones, M. (1992). The effectiveness of pre-employment drug screening in

the prediction of employment outcome. Journal of Occupational Medicine, 34, 1057-1063. Stieber, G. (2000). How can you measure the effectiveness of EAPs? Employee Benefit News, 14(3), 30. Substance Abuse and Mental Health Services Administration. (2002). Nutiunnl household survey on drug

Trice, H. M., & Sonnenstuhl, W. J. (1985). A.A. and the EAP: The historical link. The Almrtcun, 15(5), 10-12. obuse 2001. Washington, DC: US. Department of Health and Human Services.

132 journal of employment counseling September ZOO5 Volume 42