How Do Private Health Plans Manage Specialty Behavioral Health Treatment Entry and Continuing Care?

Ensuring access to behavioral health care has long been a major concern, reflecting awareness of extensive unmet need and the historical inequities between insurance coverage of behavioral health and general medical care (1). Access to and management of care involve information sharing with patients, patient choice of provider, timeliness of care, and a variety of facilitators of or barriers to care. Health care systems’ management of treatment entry and continuation drives access. Understanding how this works is essential for providers and patients to ensure that they can access needed care.

With the passage of the Patient Protection and Affordable Care Act (ACA) in 2010, most Americans now have health insurance coverage, an important step in ensuring access to behavioral health care and general medical services. It is generally acknowledged, however, that health insurance is necessary but not sufficient to guarantee access, which is also influenced by a broad set of factors, such as availability of specialty providers, utilization management, and medical necessity criteria (2,3). One key element is health plans’ management of treatment entry and continuation, which can either facilitate or hinder access to services. Managed care has traditionally utilized stricter treatment limits for behavioral health care because of concerns that people with behavioral health conditions use more and costlier care than people without those conditions (4,5). However, restrictive behavioral health coverage can present barriers to accessing services, which may raise concerns for patients, clinicians, and policy makers.

State and federal regulations requiring equal insurance benefits for behavioral health and general medical care have aimed to improve equity but historically have not addressed management approaches. The 2008 Mental Health Parity and Addiction Equity Act (MHPAEA), however, requires parity in both quantitative limits, such as copays and deductibles, and nonquantitative limits, which include care management tools, such as prior and continuing authorization and medical necessity criteria. MHPAEA also requires that if medical necessity criteria are used for authorization or denial of service coverage, those criteria must be made available to plan members.

Health plans can implement a wide variety of policies and procedures relevant to access, which is a multifaceted concept. Access involves initial access to care and the ability to obtain ongoing care when it is needed, as well as access to information relevant to obtaining behavioral health services. Health plans can influence both the quantity and the type of behavioral health care that individuals receive. A quasi-experimental study of health plans’ behavioral health care management strategies found that they significantly affected access (6).

This article reports findings from analyses of nationally representative survey data from private health plans regarding various aspects of how they managed specialty behavioral health services in 2010, including utilization management approaches, medical necessity criteria, and actions that plans took to facilitate and monitor timely access to care. This was a key time point to understand how the policy environment influences access to behavioral health services. MHPAEA was passed in 2008, but many implementation details were deferred to the rulemaking process. Interim rules were released in January 2010, and the final rules were not released until 2013. The interim rules specified that parity applied to management practices, and the final rules reinforced this.

During the study period, health plans were not yet required to apply parity to management practices. However, a significant reduction from 2009 to 2010 in prior authorization requirements for outpatient behavioral health care was identified, perhaps due to anticipation of the regulations (7). Prior national findings regarding private health plans’ management of behavioral health services in 1999 and 2003 indicated that prevalence of management approaches varied widely depending on the specific activity and also by plans’ management arrangements for behavioral health (8,9). President Obama’s Mental Health and Substance Use Disorder Parity Task Force reported that access has improved recently, but nonquantitative treatment limitations, including prior authorization, require further attention (10). Given the increasing national focus on behavioral health and emphasis on behavioral health treatment motivated by the opioid epidemic and other trends, such as integration of behavioral and general medical care, it is important to examine management practices in 2010 as a baseline to understanding the impact of these trends.

Methods

Sample

Data were from the third round of a nationally representative U.S. survey of commercial health plans regarding alcohol, drug, and mental health services, which gathered data for the 2010 benefit year. The study used a panel design with replacement. The primary sampling units were the 60 market areas selected as nationally representative by the Community Tracking Study (11). We then sampled health plans within the 60 market areas. A sample frame of health plans for this round was built by updating the sample frame from the second round of the survey by using a national health plan directory and state insurance regulator Web sites. Health plans were selected from the sample frame independently within each site. If a health plan served more than one market area, it was defined as a separate plan in each market area. In these cases, a multisite supplement was administered to the parent plan to identify differences between sites.

This sampling approach seeks to capture what the health plan sells to employers in each specific market area. For each site, health plans were asked about their top three products based on enrollment. A product (for example, health maintenance organization [HMO], preferred provider organization [PPO], and consumer-directed plan) was defined to represent all designs that are similar in all respects except for cost-sharing (for example, HMO or PPO, benefits, and network design). Health plans were not eligible if they had fewer than 300 individuals or 600 covered lives enrolled, did not offer behavioral health benefits or comprehensive health care products, or offered only Medicaid or Medicare products.

Data Collection

Data were collected from September 2010 through June 2011. The telephone survey was administered by trained interviewers to senior health plan executives. One respondent typically answered administrative questions (for example, plan characteristics, network management, and benefit design) and referred us to a general medical or behavioral health director to answer clinical questions (for example, care management, quality improvement, and pharmacy coverage). Data from the clinical module are reported here. The study was approved by the Brandeis University Institutional Review Board.

Data were obtained from 385 plans (88% response rate) reporting on 925 products. Questions were asked at the product level for the 2010 benefit year. Some questions related to the implementation of MHPAEA were also asked for the 2009 benefit year.

Measures

This analysis focused on health plans’ utilization management approaches for enrollees entering and continuing specialty mental health and substance use disorder treatment. The approaches included the availability of information to enrollees seeking treatment, prior and continuing authorization requirements, medical necessity criteria (such as InterQual [12] and mcg Care Guidelines [13]), and waiting time standards.

Four types of arrangements that health plan products used to manage behavioral health services were identified: external, hybrid-internal, internal, and comprehensive. External arrangements involve the health plan’s contracting with an independent managed behavioral health organization (MBHO) for delivery and management of behavioral health benefits. Hybrid-internal arrangements involve the health plan’s “contracting” with a specialty behavioral health organization that is part of the same parent organization as the health plan to manage behavioral health benefits. In internal arrangements, the health plan manages both general medical and behavioral health benefits, and services are provided by plan employees or through a network of providers directly administered by the plan. Comprehensive arrangements involve a health plan’s contracting with a single vendor for both general medical and behavioral health provider networks, but findings are not reported separately for these arrangements because of the small number (N=4); they are included in the total column only.

Analysis

Univariate and bivariate statistics were calculated. Utilization management data are shown by behavioral health management arrangement. The data were weighted to be nationally representative of commercial health plans offered in the United States, resulting in a weighted sample of 8,427 products. The weights were computed at the plan level according to the probability of the site’s being selected and the probability of a health plan’s being selected within a selected site. Analyses were conducted in SUDAAN 11.1 because of the complex sampling design. Significance is based on pairwise t tests with a .05 significance level, adjusted for multiple comparisons (three pairwise tests for each comparison by behavioral health management arrangement) with the Bonferroni correction.

Results

Prior and Continuing Authorization

Prior and continuing authorization approaches were largely similar for mental health and substance use disorder treatment in 2010 (Table 1). Only 4.7% of all products required prior authorization for outpatient care, although 16.9% of specialty external products required such authorization (a larger proportion than for both of the other product categories). Prior authorization requirements were less common in 2010 than in 2009, when 13% to 14% of all products required such authorization (p<.05; data not shown). In 2010, three-quarters of all products (75.0% for mental health; 73.9% for substance use disorders) required authorization to continue outpatient treatment.

Prior authorization for all other types of behavioral health care was required by at least 95% of products. About 95% required continuing authorization for other levels of substance use disorder services and residential mental health services, whereas 86% required continuing authorization for mental health partial hospital or day treatment. Products with internal behavioral health arrangements were more likely than products with the other two arrangements to specify a typical number of visits or days allowed, rather than making determinations as needed. Only products with internal arrangements reported a dollar limit (2.2% of all products and 13.8% of internal plans; data not shown).

Medical Necessity Criteria

Plans may have used multiple criteria and several approaches to sharing medical necessity criteria (Table 2). Self-developed criteria (79.2%) and American Society of Addiction Medicine (ASAM) criteria (14) (69.8%) were the most commonly used criteria for prior authorization of behavioral health care. Among products that used self-developed criteria, 13.6% used only self-developed criteria, and most also used ASAM criteria (70.4%; data not shown). Differences by behavioral health management arrangement were observed in use of all criteria sets.

Medical necessity criteria were made available to providers and enrollees on request and on denial of services by over 90% of products. Hybrid-internal products were much more likely than external and internal products to make medical necessity criteria available to providers and enrollees on a Web site; only about one-third of external products did this. Only 8.6% of products routinely sent medical necessity criteria to enrollees with the plan policy. Although nearly half of internal plans routinely sent criteria to enrollees, no hybrid-internal plans and only 9.1% of external plans did so.

Waiting Time Standards and Monitoring Approaches

Almost all products had formal standards to limit waiting time for routine and urgent treatment, but nearly 30% did not have such standards for detoxification services (Table 3). External plans, however, were as likely to have standards for detoxification services as for routine and urgent treatment. Internal products were significantly less likely than other product types to have formal standards for waiting times for routine, urgent, and detoxification services. The mean maximum wait time specified in products with such standards was 9.9 days for routine treatment, 65.4 hours for urgent treatment, and 13.6 hours for detoxification services. Although all types of products had similar maximum wait times for routine treatment, internal products had higher means for maximum wait times for urgent and detoxification services.

Patient surveys and complaint analysis were the most common methods overall to monitor wait time (Table 4) and were widely used by all product types. Two-thirds of products (66.5%) used a “secret shopper” approach to monitor wait times. This approach was much more common among hybrid-internal products (83.7%) than among internal (21.1%) or external (8.6%) products (p<.01).

Discussion

The results of the study shed light on elements of access that have implications for either facilitating care or presenting potential barriers. Addressing access is particularly important in behavioral health care, where ambivalence about entering treatment is not unusual. From the patient’s perspective, these issues matter because they bear on whether the health plan can and does reliably facilitate one’s access to indicated and effective care while protecting from major financial risk. How health plans structure information about and access to both initial and continuing care is critical.

Requiring prior authorization for outpatient behavioral health care visits was very rare in 2010, except in a notable minority of specialty external products, and it was significantly reduced compared with 2009, a finding discussed elsewhere, along with other parity changes (7). The reduction in requirements for prior authorization may have been an anticipatory response to federal parity at an interim point in the implementation and regulation process. However, it could reflect an ongoing trend.

Earlier rounds of this survey found that prior authorization for outpatient behavioral health care was required by about 68% of products in 1999 and 58% in 2003 (9,15). The absence of prior authorization requirements for outpatient care removes one step and a potential barrier for enrollees or providers in the process of getting outpatient care started. This trend may be attributable to the relatively low cost of outpatient care and recognition of the importance of facilitating earlier outpatient care to avoid crises and expensive acute care. It will be important to track these trends and plans’ responses after final parity regulations were issued, given evidence that parity has led to increases in out-of-network substance abuse treatment utilization and costs (16).

The application of medical necessity criteria to utilization management decision making affects the care enrollees receive. Because MHPAEA includes medical necessity in nonquantitative treatment limits, it is important to look at how health plans use and share medical necessity criteria. Plans reported that by 2010 medical necessity criteria were already made widely available to providers and enrollees, in line with parity law requirements. Disclosing this information to providers and patients adds a layer of transparency when services are not authorized and can help inform care decisions. The large majority of products reported using self-developed guidelines and ASAM criteria. The latter criteria are explicitly for substance use disorder treatment, which suggests that the self-developed guidelines may be primarily for mental health treatment. Internal products were less likely to use these types of medical necessity criteria and relied more heavily than other product types on commercial criteria sets. One possibility is that specialty external and hybrid-internal products had access to more specialized behavioral health management infrastructure so that they were able to refine their specific approaches. However, use of commercially available criteria enables a consistent approach across plans, and such criteria may be made available more easily to patients and providers compared with self-developed guidelines.

It is important to open the door, in a timely fashion, to appropriate treatment when individuals need it (17). Therefore, it is encouraging that nearly all products had standards with relatively quick timelines for maximum wait time for routine and urgent care. However, 15.4% reported standards that permit 14 or more days of wait time for routine care. Also, compared with other product types, a lower proportion of internal products had wait time standards. Thus, in a minority of products, more attention to timely access would be useful. In addition, 29% of products had no standard pertaining specifically to accessing detoxification services. Waiting time is perceived as a barrier to substance use disorder treatment entry (18); consequently, timely access can be critical. Some of these products may have considered detoxification to be a medical service covered under wait time standards for urgent or emergent medical care.

Because determining whether timely access standards are being met is critical to improving care, the fact that multiple methods were commonly used to monitor wait time suggests that plans are taking this seriously. It was striking that two-thirds of products used the secret shopper approach, a potentially potent approach to determining what is actually happening at the front end of the treatment process (19).

The type of arrangement that health plans used to manage behavioral health services can make a significant difference in the management approaches observed in this study. Although we observed some striking variation by behavioral health management arrangement, it was beyond the scope of the survey to identify detailed information about these relationships, including allocation of financial risk. In general, external products can be understood to shift risk to the MBHO, whereas internal and hybrid-internal products retain risk for behavioral health services. Risk allocation and characteristics of external products in this study have been explored: 77% shifted risk of the behavioral health services to the MBHO, whereas 23% hired the MBHO for administrative services only (20). Because a relatively small number of products used an MBHO for administrative services only, it would not have substantially changed our findings, and we did not break them out separately. However, specific criteria applied within behavioral health management arrangements may be as relevant as the structure of the arrangement. For example, the design of contracts between health plans and MBHOs can be used as a mechanism to address restrictive behavioral health care management practices (4). It is likely that hybrid-internal and internal products employed behavioral health management practices that we did not capture. It is also possible that the observed variation by behavioral health management arrangement resulted from unobserved characteristics of these contracts. Future surveys that capture the intricacy of these contracts and that compare them with approaches used in general medical care could aid in understanding changing strategies in response to MHPAEA. In the statistical testing, we made the assumption that health plan products offered by the same plan in different markets were independent. It is nonetheless possible that health plan products cluster somewhat by site or plan. Despite these limitations, the study provides valuable insight into a range of activities that can affect access to behavioral health care.

Conclusions

Study findings indicate that in 2010 health plans used a variety of methods to influence behavioral health treatment entry and continuing care. There was little reliance on prior authorization for outpatient care, but prior authorization for other levels of care and authorization for continuing care was typical. Plans used numerous other approaches to influence, manage, or facilitate access. These data provide an important baseline as we view the changing behavioral health care landscape, which is highly focused on improved care and access to care for patients with mental and substance use disorders. These measures of access continue to be important moving forward; however, new health plan approaches, such as the use of narrow provider networks (21), will also shape patient access to services in the future as the policy and practice landscapes evolve.