Effects of the Vermont Mental Health and Substance Abuse Parity Law

Endnotes

¹ Adverse selection may result when those who are older or sicker opt to enroll in or continue insurance to a greater extent than those who are younger or healthier. Moral hazard may occur when reduced cost sharing through insurance coverage reduces the incentive for individuals to economize in their use of health care.

² Specifically, the Vermont parity law defines a mental health condition to mean "any condition or disorder involving mental illness or alcohol or substance abuse that falls under any of the diagnostic categories listed in the mental disorders section of the international classification of disease, as periodically revised."

³ In addition, the evaluation included focus groups with a convenience sample of providers and consumers. This report does not present the results of the provider and consumer focus groups. However, the case study findings presented in Chapter II include provider and consumer perspectives (along with those of other stakeholders) that were gathered in the site visits.

⁴ BCBSVT used MBC for its health maintenance organization (HMO) product (The Vermont Health Plan) prior to passage of the MH/SA parity law. Some employers also requested an option to offer an indemnity product for MH/SA services.

⁵ For the sake of simplicity, this chapter refers to the organization as MBC, despite its subsequent name change to Magellan.

⁶ One common analogy suggested was how health plans approach the management of diabetes. Providers noted that health plans routinely pay for long-term treatment to maintain functioning in people with diabetes and to prevent acute episodes that may require hospitalization. They contrasted health plan management of chronic mental illness by pointing out that health plans - through their medical-necessity criteria - typically are willing to authorize only short-term mental health treatment (8 to 10 therapy sessions), unless the patient is in the midst of an acute episode.

⁷ The law was intended to build on Vermont's existing managed health care consumer protection law (Rule 10), which proponents believed was not adequate to address concerns about managed care arrangements for MH/SA services. Specifically, Act 129 required annual filing of medical loss ratios specifically for MH/SA conditions, as well as annual filing of a report card showing: (1) annual inpatient MH/SA discharge rates; (2) average length of stay for inpatient treatment and number of outpatient visits for MH/SA services; (3) percent of covered lives receiving inpatient and outpatient MH/SA services; (4) number of denials of MH/SA services; (5) number of denials appealed by consumers and/or providers; (6) rates of readmission for inpatient MH/SA services; and (7) patient satisfaction measures.

⁸ The empirical analysis presented in Section B includes all contracts, regardless of the level of enrollment or benefit design.

⁹ Partial/day treatment is a form of intensive outpatient treatment for MH/SA disorders that require moderate to high - intensity services. Treatment includes a minimum of 5 hours per day within a structured therapeutic milieu (Merit Behavioral Care Corporation, 1997).

¹⁰ The Mental Health Parity Act took effect on January 1, 1998 (concurrent with Vermont's parity law), and prohibited insurers from applying annual and lifetime dollar limits to mental health benefits that differed from those applied to general health benefits. Refer to Chapter I for a comparison of the terms of the Vermont and Federal parity laws.

¹¹ Prior to parity, the VFP individual plan provided MH/SA services through a managed care carve-out, whereas the VFP group plan covered MH/SA services on an indemnity basis.

¹² The Vermont parity law does not require out-of-network benefits to conform to the parity law. Comp and VFP contracts with an out-of-network MH/SA benefit covered the following services when provided by out-of-network providers: inpatient mental health - 25 days annually and 50 days lifetime at 50 percent coinsurance; outpatient mental health - up to 20 visits annually, subject to a $5,000 lifetime maximum benefit; inpatient substance abuse - 30 days per occurrence and 60 days lifetime at 50 percent coinsurance; and outpatient substance abuse - 90 hours per year and 180 hours lifetime at 50 percent coinsurance.

¹³ This result was derived from the multivariate analysis and is based on the odds ratio signifying the independent effect of parity on the probability of obtaining outpatient MH services. See Appendix C for the complete multivariate results (Appendix Tables C.1 and C.2).

¹⁴ This result was derived from the multivariate analysis (see Appendix Tables C.5 and C.6).

¹⁵ The multivariate results for Kaiser/CHP suggested that this decline was due to a secular time trend independent of the implementation of the parity law (see Appendix Table C.8).

¹⁶ These results should be interpreted with caution for two reasons. First, Kaiser/CHP unit costs may differ from those of BCBSVT. Second, out-of-pocket spending levels among Kaiser/CHP members (both pre-and post-parity) may differ from the aggregate assumptions applied based on BCBSVT member experiences. Therefore, these results should be considered illustrative of the potential effects of parity on spending for MH/SA services.

¹⁷ Firm size is defined as very small (fewer than 10), small (10 to 25), medium (26 to 50), and large (more than 50).

¹⁸ Fully insured employers purchase coverage for their employees from an insurance company or health plan. Employers who are self-insured, or self-funded, pay the claims directly (or under an arrangement with an administrative-services-only contract). Most self-insured plans were exempt from the Vermont parity law due to the Federal preemption under ERISA, although they were subject to the more limited requirements of the Federal parity law. However, self-insured plans administered by Blue Cross Blue Shield of Vermont (BCBSVT) were not exempt from the Vermont parity law because BCBSVT issued a certificate to the subscriber. (The State considers such policies as insured by BCBSVT rather than self-funded by the employer.) For the purpose of this analysis, businesses that offered both fully insured and self-insured plans were classified as fully insured plans, since at least one of their plans was subject to parity. There were not enough businesses with both types of plans to perform a separate analysis.

¹⁹ The interview was conducted with the person who was most familiar with the parity law. When a respondent reported that he or she had not heard about the parity law, the interviewer asked if there was anyone else in the firm who might be familiar with it. In such cases, the interviewer called back to talk to the most knowledgeable person.

²⁰ It is possible that this estimate understates the level of knowledge about parity to the extent that there were others in the firm who knew about parity but who were not interviewed during this survey. This would be especially plausible if the level of knowledge were lower in the large firms, in which there is greater division of labor for employee benefits, health insurance purchasing, employee relations, and other functions. In small firms, however, it is more likely that the survey would have identified someone who was knowledgeable about parity, given the multitude of probes asking to speak with an individual who was familiar with the parity law.

²¹ Specifically, the survey asked: "Was it increased utilization, cost-of-living adjustments, changes in types of health insurance plans offered, or another factor?" Nearly half of the initial responses were coded as "other," and a verbatim response was recorded. Where possible, these open-ended responses were recoded into the specified categories, and two additional categories were created: costs in general and government regulation. In addition, the category called "changes in types of health insurance plans offered" was expanded to include changes in the Vermont health insurance market.

²² Due to the small number of employers discontinuing coverage, it is not possible to develop reliable estimates of their characteristics.

²³ This estimate overstates the proportion of Vermont employees actually affected by the shift to self-insured plans, to the extent that some employees obtained coverage through fully insured products that continued to be offered or did not take up health insurance coverage through the employer.

²⁴ This estimate is a composite of the percent reporting that parity played a role in increasing the number of health plan choices (13 percent of 8 percent) plus the percent reporting that parity played a role in decreasing the number of health plan choices (24 percent of 6 percent).

²⁵ Employee assistance programs are designed to provide counseling and referral services to assist workers with personal problems that may adversely affect their performance on the job. EAPs generally address a wide range of problems, including those related to drug and alcohol abuse and mental health conditions, as well as marriage and family issues and financial and legal problems (Zarkin and Garfinkel, 1994).

²⁶ On the other hand, it is possible that the level of knowledge about parity was understated among employers. The survey took place more than 18 months after the parity law went into effect; thus, it is possible that awareness was heightened during the period of early implementation - particularly when advocacy efforts led to increased public education and proactive response by BCBSVT to address transition problems.

²⁷ The 1996 Federal parity law applied only to health insurance sponsored by employers with more than 50 employees. It also only required that annual and lifetime dollar limits for mental health be equal to those for physical health coverage. The law did not eliminate disparities in deductibles, coinsurance, and visit or day limits for mental health services, nor did it cover substance abuse treatment (USGAO, 2000).

²⁸ In 1998, about a quarter of the privately insured were covered by self-funded plans offered by employers or other purchasers not subject to Vermont's parity law (BISHCA, 1999).

²⁹ BCBSVT markets an HMO product through The Vermont Health Plan, an affiliated, licensed HMO.

³⁰ The odds ratio shows the probability of obtaining treatment post-parity compared to pre-parity. An odds ratio of 1 indicates there was no difference in the probability of obtaining treatment before and after parity. An odds ratio greater than 1 indicates that the probability of obtaining treatment was higher after parity than before parity, while an odds ratio less than 1 indicates that the probability of obtaining treatment was lower after parity than before parity.

³¹ An "enterprise" is the unit representing the entire corporation, including all divisions, subsidiaries, and branches. An "establishment" is the physical location of a single business, which typically produces a single good or provides a single service. An enterprise may consist of multiple or single establishments. According to Zarkin et al. (1995), "Because multi-establishment enterprises generally make health insurance decisions for the enterprise as a whole rather than for individual establishments, enterprise surveys are the most appropriate source of data for information on the factors affecting the decision to provide health insurance coverage and the rationale for designing health plans."

³² The KFF/HRET survey, for example, defines large firms as those with more than 200 workers.

Previous | Table of Contents | Next