Employee Responsibilities and Rights Journal

, Volume 22, Issue 4, pp 325–343

Brinksmanship Redux: Employee Assistance Programs’ Precursors and Prospects

Authors

    • Department of Business Administration, Alfred Lerner College of Business and EconomicsUniversity of Delaware
Article

DOI: 10.1007/s10672-010-9144-0

Cite this article as:
Weiss, R.M. Employ Respons Rights J (2010) 22: 325. doi:10.1007/s10672-010-9144-0

Abstract

Originating in the U.S. 70 years ago as industrial alcoholism programs, and promoted for decades by R. Brinkley “Brink” Smithers, these programs finally gained popularity during the 1970s and 1980s, when they were renamed “employee assistance programs” and given a broader mission than identifying alcoholics. Attempting to change their image as instruments of coercion, EAPs added a variety of social work functions to their portfolios but, now primarily operating as external contractors, have seen cash flows fall dramatically in the face of intense competition from similar services. EAP practitioners have called for restoring the approach promoted by academics supported by Smithers, who aggressively maintain that eliminating behavioral health problems is the responsibility of employees rather than employers. This article compares the evidence addressing the efficacy of that approach with that of alternatives, being developed outside of the U.S., that expand employee rights.

Key words

employee assistancesubstance abusestress management

“When Ford Motor Co. launched its employee assistance program in 1976,” according to Ruiz (2006), “its primary objective was straightforward: combating alcohol abuse in the workplace.” Ruiz noted that 30 years later “Ford’s EAP is virtually unrecognizable,” as it was handling “a wide spectrum of issues—ranging from finding day care for children to planning for the purchase of a new home” (p. 46). Not every U.S. organization had shared Ford’s perspective at that time; professional baseball, for example, rejected a proposal for an alcoholism program that year (Norwood 2009). Thirty years later, however, baseball’s approach had changed as much as Ford’s: all teams provided EAPs for substance abusing players and, reflecting the broadening of their mission, the Chicago White Sox’ EAP had referred the team’s manager to “sensitivity training” after he called a sportswriter “a f-----g f-g” (Cowley 2006).

The evolution of employee alcoholism programs into employee assistance programs has been characterized not only by an expanding purview, but also by a changing conception of employees’ responsibilities and rights. The earliest alcoholism programs, beginning with that initiated by DuPont in 1939, held employees responsible for performing their job adequately, while establishing employees’ right to assistance—equivalent to that for physical illnesses—in attempting to resolve addiction problems. Particularly since the global financial crisis of 2008, Greene (2009) has noted, employers have been inclined to “take more drastic steps to shift costs to employees.” A dominant trend in the U.S. appears to be to hold employees responsible for behavior that merely has the potential to impair their performance, whether that behavior is engaged in on the job or off, and even when the potential harm may not occur for many years. Although employees in the U.S. still are seen as having the right to attempt to resolve such problems before facing job termination, they are now more likely to be held responsible for the cost of resolving personal problems.

These recent shifts in EAPs’ conceptions of employee responsibilities and rights are by no means the first such controversial changes. This article examines the stages of EAPs’ development, and contrasts alternative futures for these programs. One is extrapolated from (re)current trends in the U.S. towards increasing employees’ responsibilities for productivity-related behavior. The other is based on research evidence regarding effective interventions for substance abuse and other maladaptive stress coping mechanisms.

From Employee Alcoholism to Employee Assistance

Seventy years ago, a DuPont employee came up with the idea of using his workplace to carry out the last of the twelve steps of Alcoholics Anonymous’ recovery program—proselytizing others to join this fellowship. Collaborating with the company’s medical department, he conducted supervisory training sessions in which he advocated for AA’s view that abuse of alcohol is not a moral failing, but rather a disease, and explained that employees henceforth were guaranteed the right to treatment. Employees in supervisory positions were trained to carry out a new responsibility: to refer subordinates exhibiting this disease’s symptoms to the medical department. Few complied, however, and only a handful of companies emulated DuPont’s program.

The next generation of programs was promoted by the Yale Center for Alcohol Studies, funded by R. Brinkley “Brink” Smithers, an IBM heir who had recovered from alcohol addiction through membership in AA. Smithers had subvented the Yale Center’s publication of a book—by an author whose academic credentials were fraudulent (cf. Page 1997)—claiming to provide scientific confirmation of the disease model (its evidence consisted of selected responses from 98 of 1,600 AA members surveyed). “Yale Plan” industrial alcoholism programs nevertheless did not depend (as DuPont’s had) on convincing employees that alcoholism was a disease, nor on assigning supervisors the responsibility of making medical diagnoses (Henderson and Straus 1953). Rather, alcoholism was to be diagnosed primarily on the basis of absenteeism, after which: “Paycheck indorsements were examined to see if the checks had been cashed in taverns or bars. Foremen . . . were questioned directly and specifically about these men. Finally, visits were made by the plant nurse to the homes” (Straus 1952: 496). Criticized as “witchhunts” (cf. Hirshberg and Cleland 1999; Luthans and Waldersee 1989), few of these intrusive programs were implemented.

A third generation of job-based alcoholism programs was developed by the National Council on Alcoholism—also funded by Smithers. The NCA claimed to have discovered, through “confidential studies” (the details of which they refused to reveal, according to Weiss 1986), that simply observing workplace behavior—rather than spying on employees’ private lives—provided all the evidence needed to diagnose the 20% of any company’s employees who were alcoholics or borderline alcoholics (Cherrington 1983). The NCA maintained (Von Wiegand 1974: 83) that “absenteeism, poor judgment, erratic performance, excessive material spoilage, decrease in productivity, poor interpersonal relationships, lateness and early departures” were reliable bases on which to diagnose the disease of alcoholism, and that its victims were less productive by 25% in its early stage and by 75% in its late middle stage. Although these putative symptoms were behaviors with which supervisors would be most in tune, the NCA, wary of supervisors’ reticence to make a medical diagnosis, absolved them of this responsibility. As with earlier programs, the treatment employees were granted the right to receive consisted of mandatory attendance at AA meetings, which was to continue until the disease’s symptoms (i.e., absenteeism, poor judgment, etc.) went into remission.

When the NCA’s strategy had only slightly greater success than did the Yale Plan, Brink Smithers next hired Harrison Trice, a professor of organizational behavior at Cornell University, as a research consultant. Apparently accepting as fact the NCA’s assertion that poor job performance was a reliable symptom of alcoholism—he neither offered nor cited any research evidence—Trice and his student Paul Roman (1972: 171) stated definitively that “there is no doubt that evidence of alcohol or drug abuse will typically become obvious in impaired performance.” Consequently, they argued, all employees who perform poorly should be presumed to be alcoholics and given the choice of improving their performance or forfeiting their job. They named this strategy “constructive coercion” and stated that it: “involves a simple statement that repetition of this act will lead to termination. There is no referral to a medical department or introduction into therapy because such referrals are not necessary” (Roman and Trice 1968: 249).

In arguing that employees had the responsibility to perform their job adequately, but not the right to treatment for addiction, as it was not a “genuine sickness,” Roman and Trice’s position was inconsistent with Smithers’ financial interests. As a result of lobbying by the Smithers-funded NCA, in 1966 the American Medical Association had defined alcoholism as a disease, and employee health insurance began to cover its treatment. Smithers and other entrepreneurs opened residential alcoholism rehabilitation facilities (popularly known as “rehabs”), designed to treat employed alcoholics (i.e. poor performers), whose insurance would pay their substantial fees (Weisner and Room 1984).

Without abandoning their advocacy for coercion, Trice and his associates quickly harmonized their position with Smithers’, providing a narrative consistent with convincing companies to adopt workplace alcoholism programs, and supervisors to refer poorly-performing subordinates—and thus fill rehabs’ beds. Contradicting their earlier insistence that therapy was inappropriate, their justification for coercing employees into the program was no longer that threats were sufficient to stop their drinking. Rather, it now was that coercion was necessary for getting this disease’s victims to accept treatment, as its chief symptom “was a determination to deny that a problem existed” (Trice and Beyer 1984a: 251). That Trice again did not cite research may have been due to modesty, as the best research on “denial” had been conducted by him. Alternatively, it may have been because his new claims directly contradicted what his research had found: that rather than denying their problem, alcoholics “readily and meekly admitted their problem, tending to throw themselves on the mercy of the boss if they were confronted with concrete evidence” (Trice 1964: 2-3).

Perhaps most consequentially for the success of the rehab business, on New Year’s Eve of 1970 Smithers’ telephone calls had convinced U.S. President Nixon to sign a bill creating the National Institute on Alcohol Abuse and Alcoholism (Alcoholism and Drug Abuse Week 1994). One of Brink’s men was appointed director, and this new agency very actively promoted programs encouraging employers to refer poorly-performing employees to the rehabs. The influence of Smithers and the various men he supported is evident in a later NIAAA director’s statement that the emphasis on workplace programs had been due to “research findings on the effectiveness of such programs by eminent scientists such as Harrison Trice and Paul Roman, major scientific and program initiatives in the 1960s by the National Council on Alcoholism and the Christopher D. Smithers Foundation” (Gordis 1999: 4).

Had there been, in fact, evidence of such programs’ effectiveness, the NIAAA might not have immediately changed strategies. Extant programs had failed to overcome supervisors’ reticence to take responsibility for referring poorly-performing subordinates, and few employees had been exercising their right to use these programs (not voluntarily, at least). Trice and Beyer (1982) acknowledged that the term “constructive coercion” had been too harsh, and rebranded the same procedure “constructive confrontation.” Moderating the claim that poor performers necessarily were alcoholics, the NIAAA instead asserted that “in about half the cases the employee’s problem will be alcohol related” (1976: 8). Further, they acknowledged (1976: 7) that poor performance also may result “from an emotional disturbance, various forms of drug abuse, or other personal problems.” This new “broad-brush” approach, the NIAAA hoped, would enhance cooperation. Thus, by the mid-1970s the National Institute on Alcohol Abuse and Alcoholism had dropped the word alcohol from the programs they were promoting, and had substituted the term “employee assistance program.”

EAPs Catch On

Over the following decade the rate of program adoption in the U.S. increased rapidly, with “employee assistance” rapidly replacing “alcohol” in programs’ titles. Although this sudden success might appear to have been due to the appeal of the NIAAA’s broad-brush, therapy-oriented, approach, data from a national survey indicated that the NIAAA influenced very few companies’ decision to implement an EAP (Weiss 1986). A stronger influence appears to have been the Rehabilitation Act of 1974, which included drug and alcohol addiction in its definition of disabilities for which organizations were required to make accommodations. Companies feared that, without being able to document an offer of assistance, they faced legal action from employees terminated for poor performance, who might claim to have been performing poorly due to their addiction.

The rehabs, standing ready to assist employers, now flourished. Because many of the early rehabs had been started as extensions of AA, they did need to broaden their claimed area of expertise beyond alcoholism. To accommodate employees being diagnosed as drug abusers, rehabs adopted the more inclusive term “substance abuse” to characterize the problem for which they purported to provide solutions and, typically, added Narcotics Anonymous to their treatment offerings. Many patients were so highly satisfied with their 28-day stays at these comfortable facilities that they returned repeatedly.

This high recidivism, paired with high costs, made rehabs a target for the managed care organizations that, increasingly, determined what treatment would be covered by employer-provided insurance. Since the 1990’s, coverage for rehab treatment typically has been limited to a maximum of two stays over an employee’s lifetime, and approval for a 28 day stay for employees diagnosed merely on the basis of poor job performance has become rare. Many U.S. rehabs have failed, and those that remain typically have filled beds by applying their therapeutic regime to a broadened range of addictions. At Hazelden, perhaps the most famous, AA’s idea of a 12-step program through which addicts are guided to abstinence by former abusers is reflected in the advertisement for their smoking cessation program: “Our counselors (who are recovering tobacco users) . . .[use] a Twelve Step, abstinence-based model” (Hazelden 2009). A Florida rehab (C.A.R.E. 2009) boasts of “enormous success” applying AA principles to “overeating as well as gambling addiction, sex addiction, and even internet addiction” (they do not mention whether their treatment for each of these challenges requires abstinence).

With the rehabs’ decline, EAP administrators no longer benefited from the kickbacks—as much as $1,800 per referral—rehabs had been giving them (Smith 1991). Further, curtailing alcoholism treatment undercut much of their justification for their job. Typically, when EAPs were introduced, administrators could identify a number of long-term alcohol addicts whom the organization had not previously had a means of helping, and busy themselves monitoring the quality of treatment and keeping in touch with treatment providers and supervisors. Although supervisors remained reluctant to refer subordinates simply on the basis of poor performance and few employees voluntarily identified themselves as drug–or alcohol-dependent, the difficulty of recovery from long-term addiction kept administrators occupied with treatment efforts that frequently extended for years.

The characterization of contemporary EAPs as “unrecognizable” when compared to those of 30 years ago reflects EAP administrators’ attempts to justify their jobs by expanding the range of services they offered employees whose performance might be suffering as a result of “other personal problems.” Currently, most calls to EAPs concern employees’ financial problems (Harris, Rothenberg International 2009), such as coping with mortgage foreclosures (Larson and Marquez 2007). In addition to locating child care, planning home purchases, and providing sensitivity training on how not to use obscene, homophobic epithets at press conferences, EAPs assist employees with finding day care for elders (Trading Markets 2009) and insurance for pets (Pounds 2009), choosing a college for one’s child (Pallarito 2008), starting a book club (Western News 2009), and breastfeeding (Marquez 2008).

Although only 5% of calls to EAPs in the U.S. now concern substance abuse (Harris, Rothenberg International 2009), those may nevertheless remain far more important to EAPs’ survival than calls regarding sundry “personal problems.” The Rehabilitation Act’s requirement that companies aid employees disabled by an addiction, coupled with managed care organizations’ refusal to pay for lengthy inpatient counseling sessions, strengthened EAPs’ position. Most had always provided brief counseling, and as it became clear to employers that this satisfied their legal obligation (at low cost), EAPs became ubiquitous in the U.S. By the beginning of the 21st century, more than 90% of the Fortune 500 firms had made EAPs available to their workforce (Merrick et al.2003).

The organizational form that EAPs have taken in the U.S., however, is not what the NIAAA had envisioned when they encouraged employers to assign an EAP administrator to work within companies’ human resource or medical departments. Instead, the field is dominated by external EAP providers, which typically deliver counseling by telephone or internet. Since the maturation of the U.S. EAP market in the 1990s, however, these providers have faced severe competition that has led to dramatic reductions in the per capita fees they are able to charge employers (Burke 2008).

A successful competitive strategy has been to expand internationally. In Korea, for example, the number of companies providing EAP services has grown from one in 2001 to 620 in 2009 (JoongAng Daily 2009).

The strategy of broadening the range of service offerings, however, has been less successful, due to its unintentional effect of further intensifying the competition faced by EAPs, as it became apparent that similar programs duplicated many of their offerings. Among the most problematic of these have been “employee wellness programs,” which generally are seen as containing costs associated with physical health—primarily those associated with smoking and obesity—in contrast to EAPs’ focus on mental or behavioral health problems. These two programs’ domains, nevertheless, often intersect. EWPs, many of which were marketed by health club operators, may deal with drinking, drug use, and smoking, but typically view them as maladaptive responses to stress, and advocate physical exercise as a healthy substitute for those behaviors. Integrated Wellness Solutions, Inc. (2009), for example, promises to “show you how exercise is a better coping tool for stress than smoking.”

Although EAPs generally do not ignore stress, their approach to ameliorating the various stresses that can impair employee performance is the same offer of counseling they make to substance abusers. In the wake of the financial crisis of 2008, for example, EAPs have been used to counsel employees suffering from the stress of laying off their subordinates (Elias 2009).

Also competing with EAPs in the U.S. are “chaplain assistance programs” (CAPs), which offer assistance from (not, as the name implies, to) members of the clergy. Like EAPs, CAPs offer advice regarding drug and alcohol dependency, financial matters, child care, stress management, and “the smooth transition of laid off or terminated employees” (Capital Chaplains 2009a). CAPs, however, also preside over weddings and funerals, and provide public prayers. Like EAPs, for-profit organizations such as Capital Chaplains (2009b) promise reductions in substance abuse and absenteeism, and increases in productivity. In addition, CAPs promise higher employee loyalty, trustworthiness, cooperation, and dedication to company values. One customer is said to have described his company’s CAP as “an employee assistance plan on steroids” (Capital Chaplains 2009b) and the Fox News television network observed that “employers find that what’s good for the soul may also be good for business” (Corporate Chaplains of America 2009).

Frequently overlapping with EAP services, as well, are “work-life” programs. They, also, help employees find care for children, elders, and pets, choose summer camps and colleges for their children, and provide legal and financial advice. Additionally, work-life programs provide information regarding (for example) dry cleaners and home remodelers, and offer concierge services such as “family activity coordination” (Perspectives, Ltd. 2009).

Duplication among these programs has led to attempts at integration. In some companies, wellness activities are part of their EAP, in others, EAPs are part of the wellness program, and one of the four courses required for certification as a “work-life professional” is on wellness. Increasingly, the dominant form of delivering these services in the U.S. has been for the major insurance companies to include them—at no additional cost—in packages of group health and disability insurance (Burke 2008).

Back to the Future?

The threat of their business being swallowed by insurance giants has led EAP practitioners in the U.S. to develop a narrative to justify their business’ continuing existence. Aware that they are unable to claim parity with psychologists or clinical social workers as counseling professionals, they nevertheless have endeavored to claim that they, too, deserve to be remunerated like “knowledge workers” (Hughes 2007).

Their emerging strategy appears to be to position themselves as having unique: “knowledge of the relationship between human behavior and workplace performance” (Burke 2008). They contend, for example, that “The emphasis on healthcare (and cost-containment in the U.S.) has often distracted us, as providers, from focusing on the work performance issues that constitute the roots of our field,” and that “in an insidiously subtle way, the shift of focus away from the workplace benefit of EAP has undermined the appreciation of the primary benefit that gave rise to our industry, i.e., improvement of organizational performance” (Tisone 2008).

The professional knowledge that EAP practitioners now claim as their exclusive domain consists of the views that had been promulgated—with distinctly limited success—by Smithers and those he funded and promoted. In 1985 Roman and his spouse had published a list of the ideas he and Trice had long been advocating, calling it the EAP “Core Technology”—a term borrowed from research that had been fashionable when he was a graduate student in organizational behavior in the 1960s. The elements of this “Technology,” which was to be strongly focused on alcoholism, have been summarized (Hughes 2007) as: “performance-based identification, constructive confrontation, clinical assessment, short-term counseling, referral, training, and consultation.” It was largely ignored at the time, as relatively few employers had ever bought into the idea that employees who aren’t working hard enough should be treated for alcoholism and fired if their job performance does not improve.

As the EAP industry’s market position has been usurped by competitors, however, EAP entrepreneurs have rallied around the “Core Technology.” Pompe and Sharar (2008) bemoaned that EAPs had “drifted from Roman and Blum’s original EAP Core Technology.” Beidel and Brennan (2004) argued that EAPs that don’t conform to its principles are not “truly employee assistance strategies.” Mannion (2007) declared that, without it, employee assistance doesn’t exist. Barth (2006) insisted that EAPs urgently need to take heed of the “old tools”—the only one of which she mentioned was “constructive confrontation.”

EAP practitioners represent the “Core Technology” as based on evidence. According to Herlihy (2009), it “arose in the early 1980s from research conducted by Paul Roman and Terry Blum.” The co-chairs of the Standards Subcommittee of the Employee Assistance Professional Association’s Professional Practices Committee stated: “As evidenced by the EAP Core Technology, a vital part of the work of any EAP is the consultation with supervisors and managers to educate them about the constructive confrontation and supervisory referral process.” They concluded that “the supervisory referral process—a longstanding tenet of the employee assistance field—is indeed a critical best practice for our profession” (Beidel and Brennan 2006). Cagney (2006) asserted, similarly, that “the effectiveness of constructive confrontation is very well documented, and it does work.”

Practitioners’ faith that there is evidence supporting the efficacy of the “Core Technology” and, in general, of EAPs, reflects what they have been told repeatedly by Smithers-funded academics. Blum and Roman wrote of “an impressive accumulation of evidence . . . about EAP effectiveness” (1995), and “a variety of published and unpublished studies, conducted with different methodologies, that indicate the cost-effectiveness of EAPs” (1995: 28). Roman told National Public Radio in 2004: “We do know from research that I and others have completed that about 70% of people who do go through such a formal process [i.e., an EAP based on the “Core Technology”] do succeed in getting well.” Sonnenstuhl, Associate Director of Cornell University’s Smithers Institute for Alcohol-Related Workplace Studies (funded by Smithers’ AA-loyalist widow), cited studies by Walsh et al. (1991) and Trice and Beyer (1984b) as demonstrating that programs based on constructive confrontation have been “successful at helping alcoholic employees to gain sobriety, typically reporting recovery rates of 70% or better” (1996: 16).

Mainstream academics, in contrast, have expressed concerns about the absence of evidence confirming the efficacy of EAPs and of the Core Technology’s tenets. When French et al. noted in (1995) that “no study has randomly assigned EAP-eligible employees to a control group and an EAP-treated group” (p. 452), they were echoing a long line of critical commentators. In 2000, Arthur emphasized the “embarrassingly thin, largely anecdotal” nature of evidence on EAP effectiveness, citing five literature reviews that “are critical of the lack of properly controlled and methodologically sound studies.” In 2003, Kirk and Brown observed that “reviews of EAPs over the last 15 years have noted the absence of methodologically rigorous evaluation studies” (p. 140), and that “the majority of these ‘evaluations’ comprise case studies in a single organizational setting, and are often conducted by the providers of the service” (p. 140). In 2005, Sulsky and Smith concluded that EAP evaluation studies of even minimal quality do not exist.

Noting the disconnect between this absence of evidence and the fact that “alcohol programs are unanimous in reporting positive results,” Colantonio (1989) observed that: “individuals who were referred to the program faced possible threat of job action or loss if some improvement was not achieved. Under these conditions, how much improvement in employees can be attributed to the actual program itself?” (1989: 19). Similarly puzzled, Harris and Heft (1992) found “it rather curious that so many companies have adopted EAPs, even though there is no rigorous proof they are cost-effective” (p. 255).

Among the disconfirmed tenets of the “Core Technology” is the contention that poor job performance reliably diagnoses substance abuse. Mangione et al. (1999) found no relationship between alcohol consumption and work performance problems. Ames et al. (1997) found no significant differences in absenteeism between heavy drinkers and nonheavy drinkers, nor between drinkers and abstainers. Foote’s (1990: 234) review of the research evidence concluded that “alcoholism in the early stages does not appear to produce job performance deficits,” and Ames and Delaney (1992) concluded that performance deficits are difficult to detect even among long-term alcohol addicts. Job performance is equally inappropriate for diagnosing drug abuse. Frone (2004: 147) wrote that “despite the widely held belief” that substance use impairs job performance: “In fact, no credible scientific evidence currently exists to suggest that employee substance use is among the major and consistent causes of workplace injuries and accidents (or any other productivity outcome).”

Evidence regarding the effectiveness of constructive confrontation is especially unsupportive. The Walsh et al. study cited by Sonnenstuhl (1996) did not, in fact, address constructive confrontation; rather, it was a comparison of three treatment regimes. It did report a sobriety rate—of 23, rather than 70, percent. The findings of the other study he cited, by Trice and Beyer, contradicted Trice’s principles of applying progressive discipline while making “no referral to a medical department or introduction into therapy” (Roman and Trice 1968: 249). Written warnings and suspensions were associated with poorer performance, and receiving or discussing treatment were the only predictors of higher performance.

Even the Smithers-funded scholars have been critical of EAPs and the pseudofacts on which they are based—when addressing other academics. Trice and Beyer (1984a) told fellow academics: “there are no scientific data on the effectiveness” of EAPs. Roman and his associates reported to academic colleagues that heavier drinkers had less absenteeism and tardiness than lighter drinkers (Blum et al. 1993). Regarding EAP effectiveness, Roman and Blum have criticized EAP research before academic audiences on repeated occasions, as follows:

the knowledge base for both EAPs and wellness/health promotion programs is thin (Roman and Blum 1987: 67);

A natural question to be raised about EAPs is their efficacy. One way to respond is to refer to the oft-quoted figure of 70% “success”. . . Speaking both generally and strictly, there is no answer to the question of whether EAPs are efficacious (Roman and Blum 1994: 379–380);

there is a minimal amount of research about EAP effectiveness (Roman and Blum 1999).

In his 2007 keynote address to the Employee Assistance Society of North America, however, Roman told EAP practitioners what he had been telling academics: that there was no research showing that “real EAPs” are effective. With EAP entrepreneurs having failed to convince customers that more extensive EAP services provided greater value than the minimal EAPs being given away with group insurance plans, Roman now told them that, to save their businesses, they needed to raise money to fund researchers to produce such evidence. Other EAP researchers suddenly agreed that there was no evidence of EAPs’ efficacy and that funding them to produce such evidence was an urgent priority (Pompe and Sharar 2008). Naming Roman as its senior advisor, an EAP entrepreneur made a (tax-deductible) contribution of one million dollars to establish an Employee Assistance Research Foundation (BehavioralHealthcare 2008).

Research sponsored by businesses, and intended to demonstrate that their services are worth purchasing, is unlikely to alter mainstream academics’ dim view of the evidence supporting EAPs. Attridge’s (2007) advocacy for basing EAPs’ value on improved work performance, which, he contended “is best measured with a pre- and post-contact methodology,” suggests that this literature will continue to reveal merely that below-average performers will have regressed toward the mean some time after having entered an EAP.

EAP practitioners’ attempt to reposition themselves as part of the human resource function by emphasizing the Core Technology may be highly compatible with the current trend to hold employees responsible for behavior—both on and off the job—that might reduce productivity. Increasingly, employers are collecting and integrating data on health-related behavior, criminal background, and absenteeism patterns (Tacoma News-Tribune 2009)—representing such actions as intended to make employees healthier and happier. Similarly, EAPs that use the threat of termination emphasize its value to employees. Advocating firing employees who don’t perform well, Wiley (2004) suggested: “sometimes offering to help people find new jobs will help them get on with their lives.” The director of Halliburton’s EAP (Christie 2003), noting that “the clearest lessons we learn in life are taught in ways that are unwelcome and even painful,” explained that “constructive confrontation can be the kindest gift of all. As EAPs have shown from the start, it can save jobs and even lives.” Cagney (2004) discussed return to work agreements for employees completing substance abuse treatment; noting that they mandate attendance at 12-step meetings and abstinence off-duty, she described them as “appropriate alternatives to termination that provide poorly-performing employees a structured opportunity to change their behavior.”

Body and Soul

Skeptics might interpret these trends as fulfilling O’Toole’s prediction (1980: 52) that EAPs had “laid the groundwork for the day when employers could own their employees body and soul”. Prince (1998), similarly, argued that the increasing use of “counselors” providing “treatment” was merely a more subtle cloaking of the EAP’s role in identifying poorly-performing workers:

For employers, the most important use of an EAP is for employees referred to them by supervisors. Consisting of about 20% of all EAP patients . . . supervisors refer these workers to EAPs, where they can receive counseling. Employees who refuse may face dismissal (1998: 15).

The use of EAPs to facilitate dismissals appears to be common knowledge. A British Columbia television station reported (KVOS 2009) that EAPs are “generally seen as a last chance for employees to improve their performance deficiencies. If troubled employees reject that opportunity, companies can strengthen their decision to terminate them.” The use of an employee’s rejection of this “opportunity” as a basis for termination has been discussed in testimony to the U. S. Senate (Peel 2002) regarding an executive pressured to accept EAP treatment by a supervisor who disliked her. Acquiescing—against her own physician’s recommendation—she was fired when she discontinued treatment with the EAP’s clinician.

Those who, like Vickers and Kouzmin, see EAPs as “engaging in coercive practices under a paternalistic and humanitarian guise” (2001: 62), also may note that among the strongest supporters of the Core Technology are companies with controversial reputations. When Halliburton, whose EAP director described constructive confrontation as “the greatest gift,” was assessed the largest corporate corruption fine in U.S. history ($577 million), it was merely one of that company’s many problems, including negligence leading to leaking radioactivity, and use of workers’ pension funds for executive bonuses (Briody 2004). Similarly, the most prominent supporter of Chaplain Assistance Programs, Tyson Foods, has been criticized for numerous felony violations of environmental, racial discrimination, labor, and occupational health statutes, as well as for drastically cutting employee compensation while tripling that of CEO John Tyson1.

Advocates of the “Core Technology” latterly have defended it on philosophical grounds. Mannion (2007) argued that turning EAPs into social work programs, which were allied with the “ideologies” of “diversity and multiculturalism” were part and parcel of the “intellectual and moral carnage” that “have left us a ravaged cultural and moral environment.” In 2004 he had criticized EAP practitioners who had abandoned “constructive confrontation” for their “liberal political persuasion” and their advocating “feminism and environmentalism, and that old standby, socialism,” while opposing the communal ends that are best achieved by “the worksite, the quintessential form of social cooperation.” Responding to the charge that “constructive confrontation” is coercive, the Journal of Employee Assistance (2006) editorialized that it is “misunderstood, since no one is ‘forced’ to comply with the referral or treatment,” and that it is “yet another service offered to the employer and employee” that provides “an alternative consequence to the loss of employment.”

Perhaps the most basic philosophical disagreement about the future of employee assistance, employee wellness, chaplain assistance and similar programs has to do with the relative responsibilities they assign to employees and employers for behavioral health problems. Banerjee (2006) noted concerns that workplace chaplains ignore the workplace’s role in causing these problems. Johnson (2004) observed that in CAPs “individual health behaviors get the attention, rather than the organizational issues—work conditions, deadlines and schedules, job content, coworker and supervisor relations and financial rewards.” Similarly, Chathapuram (1988) reported that EAPs “categorized client problems in personal terms . . . even when clients described problems that were work-related.” Kirk and Brown (2003) noted that the tendency “to ignore the role of the workplace” as a causal factor in employee’s behavioral problems is among “the most vehement criticisms of EAPs.”

These criticisms notwithstanding, the NIAAA has favored approaches to behavioral-medical problems that focus on changing employees rather than organizations (Duster 2006), and the EAP academics associated with Smithers have aligned their views with that perspective. Although Beyer and Trice had represented their monograph, Implementing Change (1978), as describing a successful organization-level change in dealing with substance abuse, they subsequently told an NIAAA conference that even if the “speculation” that workplace conditions contribute to substance abuse were true: “There is little evidence to date that the great bulk of efforts directed toward changing workplaces for any of a variety of objectives have been successful” (Beyer and Trice 1982: 192). Writing in academic journals, Roman and his associates had reported finding that “more satisfied workers emerged as significantly less likely to engage in each of the three potentially problematic alcohol use patterns” (Martin and Roman 1996: 19), and that “employees are more satisfied if their jobs have variety and afford them discretion in their decisions” (Finlay et al.1995: 440). Writing in the NIAAA’s magazine, however, they warned against changing organizations to reduce behavioral-medical problems, explaining: “it is unclear how to implement such changes” (Roman and Blum 2002).

Brink Smithers’ men have been consistent, in their public pronouncements, in assigning responsibility for behavioral-medical problems to employees, while absolving employers. The director of the Smithers Institute and his associates (Bacharach et al.2002), interpreted a finding of high levels of drinking in a stressful occupation not as evidence that stress might be a causal factor in abusive drinking, but rather, as demonstrating that such jobs provide workers an alibi for “norms rationalizing heavy drinking.” Roman and Baker (2002: 376) asserted that “it is critical to underline that the responsibility for change rests completely with the employee.” And Roman told National Public Radio (2004) the following regarding employers’ rights and employees’ responsibilities: “When an employee manifests poor job performance . . . the employer has such a legitimate right to intervene . . . and say ‘you have a performance problem, and we need to do something about that,’ placing the onus on the employee.”

As has been the case repeatedly in the evolution of EAPs, practitioners have endeavored to soften the coercive tone of the policies promoted by Brink’s men. In particular, employers appear to have coupled a broader view of the antecedents and consequences of stress with a more proactive (or, skeptics might say, intrusive) approach to stress reduction. Research asking human resource executives to rank the causes of absenteeism and health care expenditures found that all of those ranked highly were related to stress (Ipsos, 2004). Substance abuse was ranked as one of the top two causes by 20% of the respondents, and two other consequences of stress, depression and anxiety, were ranked (jointly) as a top cause by 60%. A number of sources of stress were rated highly: relationship with supervisor (44%), childcare (35%), co-worker conflict (28%), parenting (21%), and eldercare (19%). And 60% ranked stress, itself, among the top two causes. To reduce the impact of these stressors, organizations not only have expanded EAPs’ purview, but also have deployed additional initiatives. Wellness programs typically focus on aiding employees in quitting smoking and controlling overeating, while encouraging physical exercise as a more benign stress coping mechanism. Work-Life Programs attempt to help employees cope with the stresses of balancing work responsibilities with those of home and family, such as child–and elder-care. CAPs also devote efforts to much the same set of stress-related problems as do EAPs, but differ by offering counseling in very brief (roughly 5 minute) conversations in the workplace, rather than the scheduled office sessions favored by EAPs (Nimon et al. 2008).

Some of these efforts appear to be efficacious. Brief psychological counseling has been studied frequently and, according to McLeod and McLeod’s (2001) review of this literature, can be effective. Whether helping employees find child care reduces stress caused by difficulty finding child care has not been formally assessed; it nevertheless would seem to be a direct and highly plausible solution to that source of stress.

Some of the most popular approaches to controlling these perceived causes of absenteeism and health care costs, however, have been shown not to work. Helping employees to stop smoking would seem—much like helping them find childcare—another obvious way to reduce health care costs, and research indicates that adoption of a workplace anti-smoking policy does reduce overall cigarette consumption (Evans et al. 1999). Workplace-based efforts to break this habit entirely, nevertheless, have not been successful, even among employees who participate voluntarily. A meta-analysis of 19 evaluation studies of smoking cessation programs (Smedslund et al.2004: 200) found that their “effects do not seem to last beyond 12 months.” Similarly, few employees substitute exercise for drinking, smoking or overeating; Shephard’s (1999) meta-analysis of the 26 most rigorous evaluations of exercise programs found that this was the case no matter the type of intervention or the use of incentives.

Even if the evaluation research had found that these programs could entirely eliminate maladaptive stress coping behaviors, it is not certain that employers would find it advantageous to do so. The empirical evidence confirms that behaviors such as drinking do help employees tolerate stressful working conditions (at least in the short term). Cook and Schlenger (2002) reported that half of their subjects drank to relieve stress, and Frone (2003) described employees’ routine use of drugs to get through their workday. A number of studies (Upmark et al. 1999; Vasse et al. 1998; Webb et al.1994) found that, under stress, moderate alcohol users are significantly less likely to be absent than are abstinent employees. Barrett (2002: 79), summarizing the evidence that moderate drinking is associated with positive economic returns, observed that “benefits from moderate alcohol consumption have commonly been found, including reduced stress, anxiety and depression.”

Employers represent all of these programs as employee benefits. Yet, they all remain consistent with Roman’s premises that the employer has “a legitimate right to intervene” so as to ensure that employees live up to their individual responsibilities.

The Evidence-Based Alternative

The antipathy of Brink’s men toward interventions aimed at changing organizations, rather than individual employees, originally had some justification. Despite complaints that organizational factors were being ignored by EAPs and anecdotal evidence of these factors’ influence on stress and stress-related illness, empirical research had not consistently demonstrated their impact. Over the past two decades, however, there has been an accumulation of evidence confirming those relationships, and it now is feasible to offer an alternative approach to employee assistance that differs from the “Core Technology” not only in its consistency with the empirical evidence, but also in its conception of employee and employer rights and responsibilities.

This different conception of rights and responsibilities already has been embodied in state-level actions. In the U.K. for example, the Health and Safety Executive promulgated rules assigning employers responsibility for assessing the risk of stress-related illness arising from work activities and for taking measures to control that risk. This shift of employers’ responsibility from tertiary intervention—providing treatment for individuals after a problem develops—to primary intervention—removing the source of problems—was further institutionalized in the U.K. by the Intel v Daw decision in 2007. Awarding £134,000 to an employee who had protested her excessive workload, was never provided the assistance management agreed she needed, and was subsequently diagnosed as suffering from depression, the court concluded that employees have the right to a job that is not unnecessarily stressful, and that employers’ responsibilities include altering work conditions, rather than merely providing counseling to individuals (Barrett 2007). In that same spirit, in 2007 The Netherlands’ Working Conditions Act was revised to double fines for employers not carrying out their responsibility to prevent excessive “pressure of work.”

The premise of such governmental action—that the workplace can be a source of stress, and that stress, in turn, can be a source of behavioral-medical problems such as substance abuse or depression—now is supported by a research literature in which three key factors having emerged. First is a low level of control, a circumstance in which employees perceive insufficient autonomy regarding how their work is to be accomplished (Frone 1999). Second is poor fit, when work is experienced either as too demanding for employees with limited ability or knowledge, or insufficiently challenging for highly capable employees (Oldham and Gordon 1999). Finally, stress and its consequences are more likely when employees perceive a lack of fairness in the distribution of rewards and in the determination of how they are given (Greenberg 2004).

Emphasizing that employees have a right to a workplace that does not unnecessarily contribute to stress does not mean that employers are responsible for all behavioral-medical problems. Especially in light of the evidence of counseling’s efficacy, it should always have a role in employers’ overall approach to behavioral health. As Huppert (2009) has shown, however, behavioral health symptoms are widely distributed throughout the population, and lowering the population mean of the underlying behavior is a far more efficient method of reducing the level of diagnosable disorders than treating only the individuals at the extreme end of the distribution.

A first step in an approach to employee assistance that, beyond individual counseling, also attempts to alter the conditions that lead some individuals to exhibit unhealthy symptoms, might be a survey assessing perceptions of the fairness of an organization’s policies and how they are carried out. The impact of fairness on behavioral health has been demonstrated in diverse fields’ research literatures. Originating in medical sociology, the body of evidence on employees’ perception of an inequitable balance between their effort and the rewards they receive is now quite extensive; a systematic review by Vegchel et al. (2005) examined 45 high quality studies, and Stansfield and Candy (2006) found 11 studies suitable for their meta-analysis. The evidence indicates that perceived effort-reward inequity leads to increased tobacco and alcohol consumption, and increased levels of mental disorders. Large–sample, longitudinal, studies on organizational justice, a topic originating in organizational psychology, have found that perceived injustice is associated with behavioral-medical problems including psychiatric illness (Kivimaki et al2003) and sleep disturbance—an indicator of prolonged emotional stress (Elovainio et al. 2009). Research by epidemiologists on the concept of unfairness (DeVogli et al.2007) has found that it predicts lower levels of mental functioning.

Practical strategies to enhance fairness hardly need the involvement of an EAP or of any particular “program.” Rather, this means of reducing stress could involve actions that are basic to what most business academics and practitioners would agree is simply “good management,” such as developing clear rules and procedures to provide a standard against which decisions can be assessed, and to which everyone can be held equally.

A second step in an evidence-based effort to provide employee assistance might involve an approach, already adopted by many organizations, that stands in an intermediate position between tertiary interventions, such as counseling, and primary interventions, focused on prevention. Stress management programs, a form of secondary intervention, are designed to reduce the stress experienced by employees—not by changing the organization to eliminate stressors, but rather by enhancing employees’ resistance to stress. Among the various types of such interventions, Richardson and Rothstein’s (2008) review of the extensive research evidence found the most efficacious to be cognitive-behavioral programs, which help individuals cope with stress more directly than do strategies such as encouraging physical exercise. Cognitive-behavioral stress management programs typically begin by helping employees to understand situations that cause them to experience stress, and then train them in skills, such as listening, assertive communication, and conflict management, that can reduce the likelihood that interactions with coworkers and supervisors will result in stress-related illness.

To the extent that the circumstances contributing to stress-related illness are rooted in organizational characteristics rather than interpersonal conflict, however, these interventions’ impacts are likely to be modest. Although they will be insufficient to meet the legal obligation, in some countries, to reduce the stressors inherent in workplace structures and relationships, developing interpersonal skills prior to changing the loci of control within an organization may be especially important, as such changes have the potential to increase conflict.

Another step that might best precede increasing employees’ job control would be to increase the fit between employees and their job. The extensive research literature addressing Karasek’s job strain model (Karasek and Theorell 1990) has made clear that passive jobs, which reflect underutilization of skills, cause employees to experience stress. Most recently, for example, Gimeno et al. (2009) found, in a sample of more than 3,000 U.S. workers, that passive jobs increased the likelihood of heavy drinking. Improving the fit between an employee and a job can most readily be accomplished through careful selection and placement. When poor fit reflects a task that is beyond an employee’s skill level, fit can be enhanced through training, as well as by examining tasks to see if they can be made more compatible with employees.

Taking steps to make counseling available, train employees to better manage potentially stressful situations, and improve the match between employees’ abilities and their task, sets the stage for what the research evidence indicates is the most consistent and robust deterrent to stress—enhancing employee job control. Bambra et al. (2007) reviewed 19 studies of the impact of work reorganization, and found that changes that increase job control were the most important predictors of favorable health outcomes. Stansfield and Candy’s meta-analysis (2006) found that increased job control was associated with a reduced risk of mental disorder. And an umbrella review of seven systematic reviews of the literature on the effects of organization-level changes (Bambra et al.2009) found that interventions that increased control predicted lower risks of health consequences such as anxiety and depression.

The application of these findings on the salutary impact of increasing fit and then enhancing job control is illustrated by research conducted in the Netherlands to evaluate a program of organizational-level changes. The manufacturing facility studied by Maes and his colleagues (1998) had trained production workers to carry out tasks previously reserved for supervisors, and then authorized them to take control over “initiating work orders, arranging supply and transportation of raw materials and finished products, calculating hours spent on tasks, and performing quality checks” (p. 1039). Among the experimental subjects, health improved and absenteeism was halved, whereas control subjects experienced significantly less improvement.

Perhaps the broadest corroboration of the effect of organizational-level changes on behavioral health can be found in the work of scholars located primarily in U.S. business schools—where performance, rather than health, most typically has been the outcome of interest. In the 1960s (while Roman and Trice were advocating “constructive coercion” as a means of motivating employee productivity) Argyris and his colleagues were examining employee motivation from the perspective of Maslow’s theory of self-actualization. Hackman and Lawler (1971) began research to demonstrate that employees thrive when assigned challenging tasks over which they are given control, but the “job redesign” movement it spawned did not succeed, as altering tasks to make them more psychologically “enriching” did not consistently raise productivity. What studies of their “job characteristics model” have shown, however, is that when a task is “enriched” so that employees control outcomes, experience a level of challenge fitting their abilities, and receive fair and accurate information about their performance, their job satisfaction is higher (Behson et al.2000). Disappointingly for management scholars, satisfaction turns out to be only a weak predictor of productivity (Judge et al. 2001), and therefore has been relegated to merely a secondary status.

The health psychology literature, however, leads to a very different conclusion about the value of redesigning jobs to increase control, fit and fairness. A meta-analysis of more than 100 studies, with a total sample size greater than 70,000, indicates a correlation greater than .4 between job satisfaction and both depression and anxiety (Farragher et al. 2005)—human resource executives’ highest ranked causes of absenteeism and excess health costs.

Conclusion

After 70 years, the EAPs promoted by Brink’s men still lack evidence of their effectiveness for ameliorating behavioral-medical problems. It nevertheless is far from certain that they will not reestablish their dominance—at least in the U.S., where employees have no rights under the law to be provided anything beyond the tertiary intervention that is at the core of the “Core Technology.” These EAPs’ low cost and policy of assigning employees the responsibility for removing barriers to maintaining job performance appear to contribute further to their attractiveness to employers.

The accumulation of research evidence over two decades now offers an alternative future for EAPs that contrasts with the insistence that “the responsibility for change rests completely with the employee” and the “denial” that primary interventions are efficacious. EAPs that acknowledge employers’ and employees’ shared responsibility for the development and resolution of behavioral health problems now have the potential to serve not merely the interests of managerial control, but rather the shared interests of employers and employees in greater health.

Footnotes
1

The director of the Yale Divinity School’s Center for Faith and Culture, where Tyson sponsored a conference promoting CAPs, described John Tyson as seeking “to build a company that adheres to certain Christian values” (Banerjee 2006). Yet, although some CAPs boast of bringing employees to Christ—Marketplace Ministries claims to have led 2,600 of their clients to Bible-teaching fellowships—Tyson takes pride in its 128 chaplains’ ecumenicism. At an Iowa pork plant, for example, “the chaplain has a relationship with monks at a local Buddhist monastery” (Banerjee 2006). Presumably, that relationship does not include inviting them to pig roasts.

 

Copyright information

© Springer Science+Business Media, LLC 2010