Employers frequently contract with managed behavioral health care organizations (MBHOs) for specialty mental health and substance abuse services through behavioral health carve-outs (1,2). Most large workplaces also offer employee assistance programs (EAPs) (3), either separately or in a combined package with standard behavioral health carve-out benefits from a single vendor (usually referred to in the managed behavioral health care industry as "integrated").
EAPs are workplace-based programs designed to address behavioral health and other problems that affect employees' well-being or job performance (4). They are usually externally contracted to MBHOs and provide outpatient clinical services (for example, assessment or short-term counseling for mental health, substance abuse, work stress, or family problems) as well as nonclinical services and management consultation (5). Employers that purchase combined EAP-managed behavioral health care products (integrated products) from a single source may expect that including EAP benefits will encourage utilization patterns that are different from those in standard managed behavioral health care plans—such as greater and earlier use of outpatient care and less use of expensive higher levels of care.
There is a paucity of published data to address the question of service use differences in these two common managed care products. Studies have analyzed utilization patterns in MBHO-covered populations, but few have examined single-source integrated plans (6,7,8). A study by Cuffel and Regier (9) found a greater increase in access per additional dollar spent on behavioral health benefits in single-source integrated plans compared with plans with nonintegrated EAP benefits; however, the main focus was on the relationship between spending and access. We are not aware of studies describing utilization patterns for specific services under the two types of plans.
This study used data from a national MBHO to compare use of specific categories of behavioral health services in its integrated product and its standard managed behavioral health care product (standard product). The main research question was whether and how much the use of different types of service varied across the two product types. For example, we examined whether use of outpatient substance abuse treatment or mental health care was greater in the MBHO's integrated plan than in its standard plan and whether use of higher levels of care was less in the integrated plan. We examined only services in each type of plan and did not include use of EAP services that might be available to the MBHO's enrollees from some other source.
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 services and EAP services. As described above, these services are offered separately or combined into a single-source integrated plan. We used 2004 administrative data—deidentified claims and eligibility files—to examine the two plans. Claims included those for EAP services and for specialty mental health and substance abuse services covered by MHN. For EAP claims, only clinical services were included; claims for assistance such as legal or financial help were not used. The study received approval from the institutional review board of Brandeis University.
For enrollees in both plan 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 receive approval to see a network provider. In the single-source integrated plan, enrollees call a single toll-free number. After a brief intake, those assessed as needing regular outpatient care are usually offered the opportunity to use the EAP benefit first, with some exceptions (for example, medication evaluation or management). Typically, three to five EAP visits per year are covered at no cost to the enrollee. EAP benefits include some services that are not for a clinical disorder, such as marital therapy in the absence of a psychiatric diagnosis. EAP services are provided by network clinicians in private offices. When an enrollee reaches the EAP visit limit and needs more services, use of the managed behavioral health care portion of the benefit is authorized. Enrollees can usually choose to remain with the same network provider whose services were covered by the EAP portion of the benefit. Some enrollees, such as those who need a higher level of care or who request a medication evaluation, bypass the EAP and access services under the managed behavioral health care part of the benefit. In the standard plan, enrollees call a toll-free number to request authorization. For some enrollees in this plan, an EAP may be available through their employers, but use of these services would be outside the MHN system.
The unweighted sample consisted of 543,964 enrollees in the integrated plan and 166,050 enrollees in the standard plan. Because the integrated product is purchased only by employers, we included only standard products that were purchased by employers (not by health plans) in order to increase comparability across the two products. We excluded plans purchased by employers in an industry that was represented in only one product type. The remaining sample reflected enrollment from three sectors—services, sales, and government. Weights were applied to achieve exact matching on sociodemographic characteristics and to adjust for partial-year enrollment so that the subsamples would be more comparable (see below). The weighted sample consisted of 286,750 enrollees split evenly across product types.
We categorized each enrollee's benefit package according to whether it was an integrated plan or a standard plan. We compared use of services only within the coverage scope of each product: use of EAP and managed behavioral health care services for the integrated product and use of managed behavioral health care for the standard product. Enrollees in the latter group were defined as receiving only managed behavioral health care benefits from MHN, even though they may have received EAP benefits from a non-MHN source. Services were categorized as primarily mental health or substance abuse on the basis of the primary diagnosis and the specialty provider type, and provider type and service codes were used to further classify the services used within these two broad categories. We used the Clinical Classification Software from the Agency for Healthcare Research and Quality to group ICD-9 diagnoses into mental health and substance use-related categories, and we also included behavioral health-related V codes (10).
Enrollees in the two plan types differed on observable variables such as sociodemographic characteristics. Thus, to maximize comparability, we exact-matched the two samples on available sociodemographic variables and reweighted the number of enrollees in selected cells (11). We computed separately by product the number of enrollees in each match cell, defined as a unique combination of the following variables: gender, age group (four values), census geographic region (four values), and spouse or dependent status (yes or no). Next, we computed the ratio of enrollees in each cell in the integrated plan to those in the standard plan, and its reciprocal was used as a weight for enrollees in the integrated plan. The weighted number of enrollees in the integrated plan was thus made equal to the actual number of enrollees in the standard plan. We matched on a key subset of covariates, because using all covariates would have resulted in an excessive number of cells, many with unreasonably large weights because of low numbers. (Unreasonably large weights would have been the case whether we used exact matching or propensity scores, because all variables are categorical with few levels). Matching on the covariate subset is equivalent to creating a propensity score from the four variables.
We also sought to correct for possible bias from censored observation of members enrolled for less than 12 months, who were less likely than full-year enrollees to have any visits, by adjusting utilization rate calculations for length of enrollment. For example, a full-year enrollee would have a weight of 1 in the rate calculation, but an individual enrolled for six months would have a weight of .5.
Results are presented in weighted form. Bivariate tests (t tests) were used to compare utilization measures for the integrated plan and the standard plan. Analyses were corrected for the use of weights that varied by match cell. The correction was accomplished with SUDAAN software, with the match cell specified as a stratum variable (12).
Because of matching, weighted proportions across the two product types were equal in terms of gender (51.9% female), age (28.2% younger than 18, 20.4% aged 18 to 35, 37.4% aged 36 to 54, and 14.0% aged 55 and older), relationship to subscriber (44.3% employee, 22.0% spouse, and 33.7% dependent), and region (8.6% Northeast, 33.0% Midwest, 45.3% South, and 13.1% West) (data not shown). Data on race and ethnicity were not available.
Use of specific service types is shown in Table 1. Among all enrollees, the proportion who used any outpatient mental health or substance abuse services (including EAP services) was higher in the integrated plan than in the standard plan (54.8 versus 46.1 service users per 1,000 enrollees, p<.01), whereas use of non-EAP outpatient services was slightly lower (43.5 versus 46.1 service users per 1,000 enrollees, p<.01). Use of residential substance abuse rehabilitation was lower in the integrated plan than in the standard plan, but use of substance abuse intensive outpatient or day treatment was higher (both p<.05). The mean number of days or visits per user was modestly but significantly higher in the integrated plan for several mental health services, including outpatient mental health office visits (both total visits and non-EAP visits) and intensive outpatient or day treatment. No significant difference between plans was found in the mean number of days or visits for substance abuse services.
In both plans, about half the substance abuse service users also used mental health services—47.7% in the integrated plan and 54.8% in the standard plan (data not shown). Less than 2% of mental health service users also used substance abuse services.
Analysis of treatment patterns in the two plan types revealed interesting differences. For mental health services, patterns in the integrated plan reflected substantially greater use of outpatient mental health office visits (including EAP visits) than in the standard plan. Access to the EAP in the integrated plan may have encouraged more enrollees to seek outpatient services; for example, EAPs are typically promoted to employees in a way designed to have a destigmatizing effect (for assistance with life problems). However, use of non-MHN EAP services by enrollees in the standard plan was not examined, and further research is needed to investigate such use.
In addition, the mean total number of outpatient mental health office visits per user was significantly higher among enrollees in the integrated plan—although the magnitude was very modest—which suggests the possibility that treatment engagement was somewhat enhanced. Longitudinal, episode-level analyses would elucidate how combinations of services are used and whether most EAP service users continue on to use the managed behavioral health care portion of their benefits. For each of the mental health services examined, the proportion of enrollees who used a higher level of service did not differ by plan. However, the significantly greater quantity per user of certain higher-level services in the integrated plan does not readily fit with expectations.
Similarly, for substance abuse treatment, a larger proportion of enrollees in the integrated plan than in the standard plan used outpatient office visits. The proportion of enrollees using intensive outpatient or day treatment was also larger in the integrated plan; however, a smaller proportion used any residential substance abuse rehabilitation. The combination of these differences suggests some shifting to lower levels of care among enrollees in the integrated plan. Such patterns could occur as a result of early identification of substance abuse problems in the EAP, although our data cannot confirm this. More detailed clinical information at the episode level is necessary to explain these findings. The results also indicate a high level of mental health service use by substance abuse treatment clients in both plan types, which underlines the importance of examining use of the full range of behavioral health services for this subpopulation, regardless of plan type.
From a purchaser perspective, greater outpatient service use in the single-source integrated product may reflect desired patterns. Purchasers considering the two product types would need to factor into their decision making the complete costs and likely benefits of each. Cost factors would include the cost of purchasing single-source integrated benefits versus standard managed behavioral health care benefits, taking into account all types of service utilization. Cost factors would also include the cost of stand-alone or internally provided EAP models that could be offered alongside the standard product. Benefits, including improved clinical and productivity outcomes, would accrue with increased use to the extent that services are needed, effective, and timely.
The study has several limitations. Other MBHOs may differ in how they structure the two products, which limits generalizability. Furthermore, the research design does not allow conclusions regarding causality. This study did not analyze total use of specialty behavioral health services for enrollees in each of the two plan types, because some enrollees may have had access to a non-MHN EAP (for example, an internal EAP through their employer). Key informants at MHN indicated that in general, many employers that purchased a standard product had internal EAPs; however, it was not possible to determine this for each employer. Rather, our study provides information only on utilization differences in benefit plans of an MBHO. Because single-source integrated plans cost somewhat more than standard plans, employers that purchase these products may be more generous, which would reflect a systematic difference in purchasers that could have implications for treatment patterns. Conversely, there may not be a difference in overall employer generosity because some employers that purchase the standard product may also pay for separate EAP services not provided by MHN. Because use of EAP visits does not require a copayment, service use in the integrated plan includes the effect of offering free visits for some initial outpatient care. As with all claims data, examination of diagnoses is limited to codes entered by providers. Finally, other factors such as employer size or state parity laws may affect use.
This study found that patterns of service use for specific types of services among enrollees in a single-source integrated plan, which combined EAP services with standard managed behavioral health services, were significantly different from patterns among enrollees in a standard plan. Some differences, such as greater use of outpatient office visits and lower use of substance abuse residential rehabilitation among enrollees in the integrated plan, are in line with purchaser expectations of the potential beneficial effects of EAP availability, although other findings were mixed. Further research is needed to examine product type and other factors in overall access in a multivariate context, to determine differences between the two products in the characteristics of service users (for example, diagnosis and use of a more detailed array of services), and to more closely analyze individual-level service use in various product types with a longitudinal approach.
This study was funded by grant P-50-DA-010233 from the National Institute on Drug Abuse. The authors thank Nancy Pun and Kikumi Usui for analytic file preparation at MHN; Joanna Volpe-Vartanian, Ph.D., and Frank Holt, R.N., M.A., for research assistance; and Laura Altman, Ph.D., and Paul Roman, Ph.D., for helpful comments on an earlier version of this brief report.
Dr. Hiatt is employed by MHN. The other authors report no competing interests.