Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Adm Policy Ment Health. Author manuscript; available in PMC 2010 November 1.
Published in final edited form as:
PMCID: PMC2794127

Integrated Employee Assistance Program/Managed Behavioral Healthcare Benefits: Relationship with Access and Client Characteristics

Elizabeth S. Levy Merrick, Ph.D., M.S.W., Senior Scientist,1 Associate Professor Dominic Hodgkin, Ph.D.,1 Professor Constance M. Horgan, Sc.D., Associate Dean for Research, Director,1 Vice President, Quality Improvement Deirdre Hiatt, Ph.D.,2 Bernard McCann, M.S., C.E.A.P., Research Associate,1 Vanessa Azzone, Ph.D., Biostatistician,3 Galina Zolotusky, M.S., Statistical Programmer,1 Grant Ritter, Ph.D., Senior Scientist, Statistician,1 Sharon Reif, Ph.D., Scientist,1 and Professor Thomas G. McGuire, Ph.D.3


This study examined service user characteristics and determinants of access for enrollees in integrated EAP/behavioral health versus standard managed behavioral health care plans. A national managed behavioral health care organization’s claims data from 2004 were used. Integrated plan service users were more likely to be employees rather than dependents, and to be diagnosed with adjustment disorder. Logistic regression analyses found greater likelihood in integrated plans of accessing behavioral health services (OR 1.20, CI 1.17–1.24), and substance abuse services specifically (OR 1.23, CI 1.06–1.43). Results are consistent with the concept that EAP benefits may increase access and address problems earlier.

Keywords: mental health services, employee assistance programs, managed care, access, substance abuse treatment


Facilitating access to behavioral health services is critical since mental health and substance use disorders are frequently untreated [1]. Recent epidemiological data indicate that only 41% of persons with selected mental health and substance use conditions have received any treatment in the past year, and even fewer receive minimally adequate treatment [1]. Employer-sponsored benefits provide one avenue for intervention. The impact of mental health and substance abuse problems in the workplace is well-recognized, including productivity effects [2, 3]. Employers, bearing much of health and disability costs as well as productivity losses, have an interest in providing resources to promote employee health and contain costs and are in a position to do so. Managed behavioral health care, provided through general managed care plans or through employer “carve-outs” with managed behavioral health care organizations (MBHOs), is ubiquitous in private-sector health care in the United States [4, 5].

Many factors can influence the decision to access behavioral health care, ranging from stigma and financial barriers [6] to gatekeeping arrangements [7]. One important question that has not yet been answered concerns the relationship between inclusion of employee assistance program (EAP) benefits in a managed behavioral health care benefit package and accessing of behavioral health care within the plan, in the context of other relevant factors such as state parity laws or employer firm size. EAPs, which originated as Occupational Alcoholism Programs, are workplace-based programs that aim to address behavioral health and other problems that affect employees’ well-being or job performance [8]. The majority of large workplaces provide EAP benefits [9]. Overall, 42% of workers in private industry and 73% of workers in state and local government have access to an EAP [10]. Employer motivation for offering EAPs includes maximizing productivity, which can be impacted by unaddressed problems in various ways. For example, among EAP clients in the same MBHO as examined in the current study, 80% of costs associated with reduced productivity were attributable to presenteeism (reduced productivity while at work) rather than absenteeism [11]. Contemporary EAPs are typically externally contracted to MBHOs and are “broad-brush” programs addressing a spectrum of mental health, substance abuse, and work/life issues as well as needs such as eldercare or legal problems [12]. These external EAP services can be offered separately or integrated with MBHO carve-out benefits for a more comprehensive option. EAPs often provide 3–8 visits to employees and family members, with no co-pay, for assessment and short-term counseling. EAPs also provide organizational consultation and training. Case finding, early intervention, high acceptability, and easy access are among EAP goals.

Inclusion of an EAP in the behavioral health benefit package may encourage more people to use services, at an earlier stage of behavioral health difficulty, compared to standard behavioral health benefits. EAPs are usually publicized as providing assistance for stress, work/life problems, and other low-threshold concerns as well as behavioral health disorders. There are no financial barriers to EAP access. Workplace personnel may refer employees to the EAP. Referral to non-EAP specialty treatment when necessary may be facilitated, because a single vendor administers the EAP and managed behavioral health care (MBHC) benefits. Finally, in some integrated plans, individuals who receive assessment and/or short-term counseling through the EAP may continue specialty treatment with the same provider. Thus, there are numerous reasons that both the proportion and the characteristics of enrollees accessing care through an integrated EAP/MBHC plan could be different than within a standard MBHC package. Alternatively, if enrollees perceive the EAP negatively, then access in integrated plans may not be enhanced.

Previous studies have analyzed behavioral health care utilization patterns within MBHO-covered populations from several angles, but few have examined the integration of EAP and MBHC benefits. A substantial body of research has focused on behavioral health carve-out evaluations and descriptive analyses of utilization patterns within MBHOs [13-16]. However, these studies typically include MBHC-only arrangements or do not specify whether EAP services may be part of the benefit package. Cuffel and colleagues reported a higher marginal increase in access per additional dollar spent on behavioral health benefits in integrated EAP/MBHC plans; however, the main focus was the relationship between spending and access [17]. A descriptive analysis of the same MBHO studied here, focusing on use of specific types of services, found that there was more mental health and substance abuse outpatient use in the integrated product and less use of substance abuse residential treatment. [18] To our knowledge, published studies in this area have not modeled the effects of integrated EAP/MBHC product type on initial access (use of any behavioral healthcare) using multivariate methods, or compared key client characteristics within the two types of plans.

This study uses data from a national MBHO’s integrated EAP/MBHC and standard MBHC products to address the following research questions:

  1. Is inclusion of EAP benefits a predictor of greater accessing of behavioral health (and, separately, substance abuse) care, in the context of other determinants; and if so, what is the magnitude of the relationship?
  2. Are the characteristics of service users different in the integrated EAP/MBHC product versus the standard MBHC product?

This analysis provides a view of determinants of access (defined as use of any behavioral health services) and client characteristics within alternative behavioral health care products that are commonly offered, but rarely examined. The findings will improve our understanding of how product type may be related to improved behavioral health treatment access.


Study Setting and Data

The data source was Managed Health Network (MHN), a national MBHO covering 11 million members. MHN contracts with employers and other payers to manage and deliver specialty behavioral health and EAP services. These services are offered separately or as an integrated product. In this analysis, we focused on integrated and MBHC-only products. We used 2004 administrative data, including de-identified claims and eligibility files. Claims included EAP, specialty mental health and substance abuse services covered by MHN. For EAP claims, only clinical services were included, not assistance such as legal or financial consultation. The study received Institutional Review Board approval.

Treatment Entry Process

It is important to note that not all MBHOs structure their integrated products the same way, thus we describe the process at MHN in some detail. In both product types, accessing services involves calling a phone center for authorization. Authorization is a routine process in which eligibility is verified, brief intake is performed, and enrollees are approved to see a network provider. In the integrated product, enrollees call a single toll-free number to access care for either EAP or managed behavioral health services. Following a brief intake, enrollees assessed as needing regular outpatient care are typically offered the opportunity to use the EAP portion of the benefit first. Some enrollees, such as those initially needing a higher level of care, requesting a medication evaluation, or continuing with prior treatment arrangements would automatically bypass the EAP and access services under the MBHC part of the benefit. Otherwise, enrollees are generally offered access to the EAP benefit, which carries a financial incentive in that there is no co-pay; however, they are not required to use the EAP. Typically 3–5 EAP visits per year are covered at no cost to the enrollee. When an enrollee reaches the EAP visit limit and needs more services, the MBHC portion of the benefit is accessed, after re-authorization. This is considered an “integrated” product for a number of reasons. Both types of services are co-located within the same benefit package offered by the same vendor. When enrollees seek help, intake staff can authorize appropriate care within the broader range of possible services. Also, enrollees continuing treatment can usually choose to remain with the same network provider seen through the EAP portion of the benefit, meaning potentially better continuity for those in need of ongoing outpatient treatment. In the MBHC-only product, enrollees call a toll-free number to request authorization. MBHC-only enrollees may have a non-MHN-provided EAP available through their employers, but this is not observable in MHN’s data.


The unweighted sample consisted of 543,964 enrollees in the integrated product and 166,050 enrollees in the MBHC-only product. Since the integrated product was purchased only by employers, to increase comparability across the two products, we included MBHC-only accounts purchased by employers (not by health plans). We excluded plans purchased by employers in industries represented in only one product type. The remaining sample reflected enrollment from services, sales and government. Weights were applied in order to accomplish exact matching along sociodemographic characteristics and to adjust for partial-year enrollment in order to make the subsamples more comparable (procedure described in detail below). The weighted sample consisted of 286,750 enrollees split evenly across product types.


Behavioral health utilization

Behavioral health care access was defined as having any specialty behavioral health or clinical EAP claim during 2004. ICD-9 diagnoses were grouped into behavioral health categories using AHRQ’s Clinical Classification Software [19]. We also included all behavioral health-related diagnoses such as ICD-9 v-codes. Services were categorized as either primarily mental health or substance abuse based on primary diagnosis and specialty provider type. Provider type and service category codes were used to determine detailed service categories.

Independent variables

The key explanatory variable was a dummy variable indicating enrollment in the integrated product. Covariates included gender, age category, relationship to subscriber (employee versus spouse or dependent), Census region of residence, and employer size (greater or less than 5000 employees). We also included dummy variables indicating residence in a state with a state mental health parity law (36 states), and residence in a state with a state substance abuse parity law (17 states). Parity laws require that mental health or substance abuse coverage be provided on equal terms as for medical care, so individuals in parity states may face fewer financial barriers. We did not introduce copayment variables because the option of initial EAP visits with no co-pay is a key feature of the integrated product itself.

Statistical Analysis

Enrollees in the two products differed on observable variables. We exact-matched the two samples on selected covariates and re-weighted enrollees in selected cells to maximize comparability [20]. We computed separately by product the number of enrollees in each match cell, defined as a unique combination of these variables: gender, age group (4 values), census region (4 values), and spouse/dependent status (yes/no). Next, we computed the ratio of integrated to MBHC-only enrollees in each cell, and its reciprocal was used as a weight for integrated enrollees. This process made the weighted number of integrated enrollees equal to the actual number of MBHC-only enrollees. We matched on a subset of covariates, because using all would have resulted in an excessive number of cells, many with unreasonably large weights due to low numbers. (This would have resulted whether using exact matching or propensity scores, since all variables are categorical with few levels). This is equivalent to creating a propensity score from the four variables. Variables not used for matching were entered as covariates in the regressions.

We also sought to correct for possible bias from censored observation of members enrolled for less than 12 months, who were less likely to have any service use than full-year enrollees, by adjusting utilization rate calculations for length of enrollment. For example, a full-year enrollee would have a weight of one in the rate calculation, but an individual enrolled for six months would have a weight of 0.5.

All results are presented in weighted form. Bivariate tests (chi-square tests for variables with multiple categories and t-tests for other variables) were used to compare enrollee characteristics and utilization measures for the integrated and MBHC-only products. Two logistic regression analyses were constructed to investigate any behavioral health services, and any substance abuse services specifically, as the dependent variables. Independent variables included integrated product type, gender, age category (reference group = 36–54 years old), spouse/dependent, region (reference group = West), employer size of 5000 or fewer employees, and mental health and substance abuse state parity law variables. All analyses were corrected for the use of weights that varied by match cell. The correction was accomplished using SUDAAN software, with the match-cell specified as a stratum variable [21].


The weighted proportions across the two product types were equal in terms of gender, age, relationship to subscriber, and region (combined sample description is shown in Table 1). Enrollees in the integrated product were more likely to be covered through employers with 5000 or fewer workers (9.1% versus 6.0%), and less likely to live in states with mental health parity (78.4% versus 86.3%) or substance abuse parity laws (19.7% versus 20.1%) (all p<.01, data not shown).

Table 1
Combined Sample Description, Integrated and Managed Behavioral Health Care Enrollees

Table 2 shows that the weighted percent of enrollees using any behavioral health services was higher in the integrated product (5.7% compared to 4.8%, p< .01). In both products, the vast majority of behavioral health service use was for mental health. Use of substance abuse services was low: 0.21% in the integrated product and 0.17% in the MBHC-only product (p <.01). Overall use of any behavioral health services, and use of any mental health services, was higher in the integrated product for almost all characteristics examined. Substance abuse service use within the integrated product was also higher in most categories.

Table 2
Percentage of Enrollees with Behavioral Health Claims

The characteristics of behavioral health service users varied by product type (Table 3). Clients in the integrated product were more likely to be employees rather than dependents (p<.01). They were also more likely to receive a primary diagnosis of adjustment disorder (p<.01), and less likely to receive a mood disorder diagnosis (p <.01).

Table 3
Characteristics of Behavioral Health Service Users

In logistic regression analysis, the integrated product was associated with higher likelihood of behavioral health access (OR 1.20, confidence interval [CI] 1.17–1.25), controlling for other factors (Table 4). Female gender was associated with greater access (OR 1.31, CI 1.26–1.35). Compared to the age 36–54 reference group, other age groups were associated with lower likelihood of access, as was residence in all regions compared to the West, and employer size below 5000 employees (all p<.01). Residence in a mental health parity state was associated with lower likelihood of access (p <.01).

Table 4
Logistic Regression: Any Behavioral Health Claim and Any Primary Substance Abuse Claim

The logistic regression model predicting any primary substance abuse claim indicated that the integrated product was again associated with increased access (OR 1.23, CI 1.04–1.46). Being female was associated with lower likelihood of a substance abuse claim (OR 0.52, CI 0.44–0.62). Being a spouse or dependent brought a higher likelihood of substance abuse treatment and access compared to employees (OR 1.52, CI 1.25–1.85). Age of less than 18 or 55 and older were associated with lower likelihood of substance abuse claims compared to those 36–54 years old. The South had significantly lower likelihood of substance abuse treatment access relative to the West.


The results shed light on each of the key research questions. First, there is evidence of greater overall access in the integrated product. The significantly higher proportion of enrollees using any behavioral health services in the integrated product suggests that including EAP benefits is associated with a broader uptake of services within the plan, compared to in the MBHC-only product. This differential exists for both mental health and substance abuse services. The fact that mental health service access was higher in the integrated product across the majority of enrollee characteristics and other independent variables indicates that this phenomenon is fairly broad in scope. While it may not be surprising to find an increase in the proportion of enrollees using services when an additional benefit (EAP) is included, this effect has rarely been quantified in the published literature. The odds ratios of 1.20 and 1.23 respectively for accessing behavioral health and substance abuse services indicate that the magnitude is substantial. In other words, our results imply that adding an integrated EAP benefit brings additional persons into services within the plan, relative to standard MBHC benefits.

Second, client characteristics did vary across the products in this matched sample of enrollees. Service users in the integrated product were somewhat more likely than those in the MBHC-only product to be employees and less likely to be dependents. To increase utilization among dependents, specialized or targeted promotional materials might need to focus more on publicizing service availability for this group. Alternatively, it may be that when dependents exhibit behavioral health symptoms, they are more appropriately dealt with through specialty behavioral health care. It is noteworthy that overall, spouse/dependent status was positively related to greater substance abuse treatment access; this may reflect less reluctance among family members than employees to use workplace benefits for substance use disorders due to stigma and fear of work-related consequences. Clients in the integrated product were more likely to have an adjustment disorder diagnosis and less likely to have mood or other mental health disorders. This tendency towards less severe diagnoses among integrated product clients is consistent with the hypothesis that EAPs encourage earlier use of services, and that broad-brush EAPs succeed at reaching more persons with less severe disorders.

State parity laws did not have the expected positive impact on service use and in one case showed a negative effect, possibly because some enrollees in parity states were not covered, for example due to their employer’s size or self-insurance status. In addition, state parity laws may affect treatment intensity rather than initial access, because some states apply the laws only to benefit limits, not initial cost-sharing (22). Similarly, a number of states apply parity only to more severe disorders, for which complete absence of treatment is less likely. Finally, some enrollees may view authorization requirements as a barrier. However, this would seem unlikely to vary by product type, given the treatment entry process described.

It is noteworthy that members enrolled with smaller employers (under 5000 employees) were less likely to have any behavioral health claims. This could reflect either lower prevalence of behavioral health problems among those groups, or lower willingness to seek care despite similar prevalence. Smaller firms face stronger disincentives to hire and retain persons with behavioral health problems, given that this has a larger effect on their experience-rated health insurance premiums than at larger firms (where greater risk-pooling dilutes the effect).

The study has several limitations. The study included a large sample of enrollees, but is nonetheless a study of a single MBHO. Generalizability is limited to the extent that other MBHOs differ in how they structure integrated and MBHC-only products. We were not able to examine access in the context of individuals’ need for services as prevalence data for the covered populations were not available. The research design does not allow conclusions regarding causality. This is also not an analysis of total specialty behavioral health-related utilization in each of the two product types, since some MBHC enrollees could have access to an EAP outside of MHN, (e.g., an internal EAP through their employer), and non-MHN utilization was not captured. Key informants at MHN indicated that in general, many MBHC purchasers (employers) had internal EAPs, however it was not possible to determine this for each account. Rather, our study provides information on accessing any services and client differences within one large MBHO’s covered benefit plans. It is also possible that since in this MBHO integrated plans cost slightly more than MBHC plans, those employers may be more generous and this could have implications for treatment patterns; we were not able to control for this possible selection bias. However, it is not clear that MBHC purchasers are in fact less generous, since they may be funding EAP services outside of MHN. Since EAP visits had no co-pay, the effect of integrated product enrollment includes the effects of offering initial visits with no co-pay for much outpatient care. As with all claims data, examination of severity and diagnosis is limited to what providers coded.

The study provides a rare opportunity to examine inclusion of EAP benefits as a predictor of access (in the context of other relevant potential determinants) and client differences in a large sample covered by either integrated EAP/MBHC or standard MBHC benefits. Study findings that enrollment in an integrated EAP/MBHC product was associated with greater use of any behavioral health services and substance abuse services in particular (compared to utilization observed within the standard MBHC product) are consistent with the belief that low-barrier EAP benefits may encourage more individuals to access care. Differences in client diagnoses are consistent with earlier intervention and broader reach among enrollees within the integrated product. Differential use rates for employees versus dependents in the integrated product seem likely to be related to perceptions and awareness of EAP benefits.

Further research is needed to address causality, examine total behavioral health utilization including non-MBHO services, and extend the findings to other MBHOs. Other additional work could include a focus on program promotion, since the degree of program promotion for both integrated and MBHC products could affect utilization. In addition, access is a prerequisite to obtaining high-quality, effective care, but does not guarantee results. Thus, examining how increased access via EAP may translate into differences in quality of care or clinical outcomes would be an informative extension of this study. Longitudinal utilization analyses would be useful to examine clinically relevant patterns such as continuation with the same clinician for both EAP and MBHC portions of service delivery. Individuals’ perceptions of and response to common benefit packages is also crucial to understand.

Understanding the results in light of the study’s limitations and the need for future research is essential. Nonetheless, the current study’s overall results are consistent with the concept that providing EAP coverage integrated with managed behavioral health care benefits may be beneficial in improving access and treatment of behavioral health problems when they are less severe, in comparison to observed patterns within a standard MBHC benefit package. This is important because employers can choose the benefits they purchase. To the extent that these decisions facilitate access, benefit selection (in this case, inclusion of EAP and MBHC benefits in an integrated package) may represent a practical way to help reduce the extensive unmet need for assistance with mental health and substance abuse problems.


This study was funded by the National Institute on Drug Abuse grant # P-50-DA-010233 through the Brandeis-Harvard Research Center on Managed Care and Drug Abuse Treatment. The authors thank Nancy Pun and Kikumi Usui at MHN for analytic file preparation, Joanna Volpe-Vartanian and Frank Holt for research assistance, and Laura Altman and Paul Roman for helpful comments on an earlier version of the manuscript.


Presentation of preliminary findings: Earlier versions of selected findings were presented at the Research Society on Alcoholism meeting (2007) and the American Public Health Association meeting (2007).


1. Wang PS, Lane M, Olfson M, et al. Twelve-month use of mental health services in the United States: results from the National Comorbidity Survey Replication. Archives of General Psychiatry. 2005;62(6):629–640. [PubMed]
2. Mangione TW, Howland J, Amick B, et al. Employee drinking practices and work performance. Journal of Studies on Alcohol. 1999;60:261–270. [PubMed]
3. Kessler RC, Akiskal HS, Ames M, et al. Prevalence and effects of mood disorders on work performance in a nationally representative sample of U.S. workers. American Journal of Psychiatry. 2006;163:1561–1568. [PMC free article] [PubMed]
4. Horgan CM, Garnick DW, Merrick EL, et al. Changes in how health plans provide behavioral health services. Journal of Behavioral Health Services & Research. 2009;36(1):11–24. [PMC free article] [PubMed]
5. Oss M, Jardine EL, MJ P. OPEN MINDS Yearbook of Managed Behavioral Health & Employee Assistance Program Market Share in the United States, 2002-2003. Behavioral Health Industry News, Inc.; 2002.
6. Lindrooth RC, Lo Sasso AT, Lurie IZ. The effect of expanded mental health benefits on treatment initiation and specialist utilization. Health Services Research. 2005;40(4):1092–107. [PMC free article] [PubMed]
7. Hodgkin D, Merrick EL, Horgan CM, Garnick DW, McLaughlin TJ. Does type of gatekeeping model affect access to outpatient specialty mental health services? Health Services Research. 2007;42(1 Pt 1):104–23. [PMC free article] [PubMed]
8. Blum TC, Roman PM. In: Cost-Effectiveness and Preventive Implications of Employee Assistance Programs. Substance Abuse and Mental Health Services Administration CfSAP, editor. DHHS; Rockville, MD: 1995. DHHS RP-0907. 1995.
9. U.S. Bureau of Labor Statistics National Compensation Survey: Employee Benefits and Private Industry in the United States. 2005. March 2005. from
10. U.S. Bureau of Labor Statistics Recent data on employers’ costs and employees’ access. Program Perspectives On Health Benefits. Oct, 2008. National Compensation Survey Benefits Series. Accessed at:
11. Hargrave GE, Hiatt D, Alexander R, Shaffer IA. EAP Treatment Impact on Presenteeism and Absenteeism: Implications for Return on Investment. Journal of Workplace Behavioral Health. 2008;23(3):283–293. 1555-5259. 2008.
12. Merrick EL, Volpe-Vartanian J, Horgan CM, et al. Alcohol & drug abuse: Revisiting employee assistance programs and substance use problems in the workplace: key issues and a research agenda. Psychiatric Services. 2007;58:1262–1264. [PMC free article] [PubMed]
13. Grazier KL, Eselius LL. Mental health carve-outs: effects and implications. Medical Care Research and Review. 1999;56(Suppl 2):37–59. [PubMed]
14. Frank RG, Garfield RL. Managed behavioral health care carve-outs: past performance and future prospects. Annual Review of Public Health. 2007;28:303–320. [PubMed]
15. Sturm R. Tracking changes in behavioral health services: how have carve-outs changed care? Journal of Behavioral Health Services & Research. 1999;26:360–371. [PubMed]
16. Greenfield SF, Azzone V, Huskamp H, et al. Treatment for substance use disorders in a privately insured population under managed care: costs and services use. Journal of Substance Abuse Treatment. 2004;27:265–275. [PubMed]
17. Cuffel BJ, Regier D. The relationship between treatment access and spending in a managed behavioral health organization. Psychiatric Services. 2001;52:949–952. [PubMed]
18. Merrick EL, Hodgkin D, Hiatt D, et al. Patterns of service use in two types of managed behavioral health care. 2009. Unpublished manuscript, under review. [PMC free article] [PubMed]
19. Agency for Healthcare Research and Quality Clinical Classification Software for ICD-9-9-CM Fact Sheet. 2003. from
20. Morgan SL, Harding DJ. Matching estimators of causal effects: Prospects and pitfalls in theory and practice. Sociological Methods and Research. 2006;35:3–60.
21. Research Triangle Institute . SUDAAN User’s Manual, Release 8.0. Research Triangle Institute; Research Triangle Park, NC: 2002.
22. Frank RG, Goldman HH, McGuire TG. Will parity in coverage result in better mental health care? New England Journal of Medicine. 2001;345(23):1701–4. [PubMed]