ON 26 SEPTEMBER 1996, President Bill

Transcription

1 120 Covering Mental Health And Substance Abuse Services Nearly all large employers cover mental health/substance abuse services, but not to the same extent as they cover other medical care. by Jeffrey A. Buck and Beth Umland ON 26 SEPTEMBER 19, President Bill Clinton signed into law a bill that included a requirement for insurance companies to have the same annual and lifetime spending limits for mental health services as they do for primary health care. 1 Understanding the effects of this law is difficult. Insurance plans often impose greater restrictions on mental health and substance abuse (MH/SA) benefits than they do on general medical care. However, our knowledge of the nature of these restrictions and their trends is limited. Most of what we know comes from employee benefit data collected by the U.S. Department of Labor's Bureau of Labor Statistics (BLS) from a sample of firms with or more employees. These data show that limitations have been increasing for MH/SA benefits. The percentage of medium and large firms that reported that coverage for mental health inpatient services was the same as that for other illnesses dropped from 27 percent in 1988 to 1 percent in Equivalent coverage for outpatient substance abuse care dropped from 17 percent to 10 percent over the same period. Benefits can be restricted through limits on days or visits, total dollars spent on care, or cost-sharing requirements. BLS data show increases in the use of nearly all of these methods. In 19, about half or more of medium and large firms used limits on the number of inpatient MH/SA days that were more restrictive than those used for other illnesses. 3 For outpatient care, half or more of medium and large firms had more restrictive limits on expenditures for MH/SA services than on expenditures for other types of services. In this paper, we present data from two different studies of MH/SA services in employer health plans and health maintenance organizations (HMOs). These studies provide more recent data about trends in MH/SA coverage limits and offer more detailed information than was previously available about benefits coverage and financing and differences by type of health plan. Data Sources Two sources of data provide the information presented here. The first is a 1995 panel survey conducted by Foster Higgins, an employee benefits consulting firm. This survey focused on MH/SA benefits and used a convenience sample of 171 employers, which replicated a similar survey of the same employers in The panel study was conducted independently of the more widely known annual survey that Foster Higgins conducts of employer health plans. The survey panel was composed almost entirely of large employers, with a total of,236,680 employees. Only 2 percent of these employers had fewer than 500 employees, and 22 percent had more than 20,000 employees. Nearly half of the panel employers had 5,000 Jef Buck, a clinical psychologist, is the associate director for organization and financing in the Center for Mental Health Services, Substance Abuse and Mental Health Services Administration, in Rockville, Maryland. Beth Umland is a managing consultant for A. Foster Higgins and Company, in New York City. H E A L T H A F F A I R S - V o l u m e 16, N u m b e r 19 The Peopleto-People Health Foundation, Inc.

2 T R E N D S : MENTAL HEALTH AND SUBSTANCE ABUSE or more employees. National data maintained by Dun and Bradstreet indicate that only 10 percent of employers with 500 or more employees are this large. Geographic representation on the panel was similar to that for all employers with more than 500 employees, with a slightly greater representation of those based in the West and a somewhat lesser representation of those based in the South. The second source of data is the annual HMO industry surveys conducted from 1986 to 199 by the American Association of Health Plans (AAHP). (This survey was formerly conducted by the Group Health Association of America before it merged with the American Managed Care and Review Association to become the AAHP.) This survey is mailed each year to all HMOs with at least one year of operational experience. In 199, 373 HMOs responded to this survey; these HMOs accounted for 86 percent of HMO enrollees. Information on benefits is for HMOs' "best-selling" plans, defined as the benefit package to which the largest proportion of an HMO's enrollees subscribe. This approach limits the information that HMOs need to provide while yielding good descriptions of coverage and industry trends. Typically, a large majority of an HMO's members are covered by its best-selling package or packages with more comprehensive coverage. Results In the Foster Higgins survey, the panel respondents reported on the health plan that had the highest employee enrollment in 1995, for each type of plan. For instance, an employer offering an indemnity plan and two HMO plans provided information on the indemnity plan and the largest HMO plan. The 171 employers in the survey provided information on a total of 23 plans that they offer. Of these, 25 percent were indemnity plans, 22 percent were preferred provider organizations (PPOs), 18 percent were point-of-service (POS) plans, and 28 percent were HMOs. Carve-out plans, in which employers contract separately for MH/SA services, constituted 7 percent of the total. For comparison, national data for 1995 show that 27 percent of employees in large firms were enrolled in indemnity plans, 27 percent in PPOs, 15 percent in POS plans, and 31 percent in HMOs. Employer-provided MH/SA benefits. Fewer than a third of respondents said that corporate policy was to cover MH/SA services to the same extent that it covered other medical services. Among the panel study participants, the annual cost of MH/SA benefits in 199 averaged $225 per employee in indemnity plans, $176 in PPOs, $160 in POS plans, and $155 in HMOs. In freestanding mental health (carve-out) plans, MH/SA costs averaged $185 per employee. One-fifth of the panel study employers carved out MH/SA benefits from at least one of their medical plans and provided them through a separate managed mental health care plan. Employers that used a carve-out plan reported that 63 percent of their employees, on average, were enrolled in it; employees who were not in the plan were likely to be in an HMO. Of employers that carved out MH/SA benefits, 0 percent said that the primary reason for doing so was to save money, 12 percent said that it was to better predict and manage MH/SA expenses, and 8 percent said that it was to offer employees more coverage or better care. Employers are most likely to cover traditional forms of MH/SA services. For mental health services, nearly all of the survey respondents covered inpatient psychiatric care and outpatient psychotherapy, but they varied in their coverage of other modes of treatment (Exhibit 1). About two-thirds covered intensive nonresidential treatment, such as partial hospitalization, and about a third covered nonhospital residential care. For substance abuse services, most covered inpatient and outpatient detoxification treatment and outpatient therapy. About two-thirds covered intensive nonresidential treatment and case management and referral services. About a third covered nonhospital residential substance abuse care, whereas less than a fifth 121 HEALTH AFFAIRS - J u! y I A u gu s t

3 EXHIBIT 1 Percentage Of Employers Covering Specific Mental Health And Substance Abuse Benefits, 1995 Type of plan All plans a Indemnity Type of service Mental health Inpatient psychiatric care Outpatient psychotherapy Intensive nonresidential care Crisis services Nonhospital residential care Substance abuse Inpatient detoxification Outpatient counseling Outpatient detoxification Intensive nonresidential care Case management Nonhospital residential care Methadone maintenance Number of employers with plan type PPO POS HMO Carve-out 99% % % % % % SOURCE: Within each type, figures based on the primary medical plan identified in a panel survey of 171 employers, conducted by Foster Higgins in NOTES: PPO is preferred provider organization. POS is point-of-service. HMO is health maintenance organization. a Figures represent the percentage of employers that cover the service in at least one of the plans they offer. covered methadone maintenance. These figures show that while little variation exists for traditional forms of care, notable differences exist for others. Generally, carve-out plans are most likely to cover the listed services, followed by POS plans, and HMOs are generally the least likely to cover a given service. Coverage limits. Employers often restrict MH/SA benefits by placing more limits on their use or imposing greater cost sharing than they do for other health care services. In a majority of plans, these restrictions are the same for both mental health and substance abuse services; accordingly, the information that follows only describes limits for mental health services. Cost sharing may take the form of limits on annual or lifetime benefit payments and/or the use of deductibles, copayments, or coinsurance for services. The panel survey did not collect information on the use of deductibles, copayments, or coinsurance. However, it shows that in 1995 employers most commonly used limits on the maximum dollar amount per lifetime and the maximum number of days per year for psychiatric inpatient care (Exhibit 2). The major exception is in HMO plans, which used limits on annual number of days more than other types of plans did. Limits on outpatient mental health care services in 1995 showed more variation by plan type than did limits on inpatient services. Indemnity and PPO plans most commonly used maximum annual dollar amounts, whereas POS and HMO plans most commonly used visit limits. A majority of carveout plans used both visit limits and lifetime dollar limits for outpatient mental health services. For both inpatient and outpatient mental health services, carve-out plans were less likely to use limits in 1995 than were other types of plans. Changes from Data from 1989 from the same employer panel allow identification of spending trends, the use of employee assistance programs (EAPs), and the use of cover- HEALTH AFFAIRS - Volume 16, Number

4 T R E N D S : MENTAL HEALTH AND SUBSTANCE ABUSE EXHIBIT 2 Percentage Of Employers With Limits On Mental Health Benefits, By Type Of Plan, 1995 Type of limit Inpatient services Maximum number of days per year Maximum number of days per lifetime Outpatient services Maximum dollar amount per visit Maximum number of visits per year Maximum number of visits per lifetime Number of employers with plan type Type of plan Indemnity 29% _a 106 PPO 38% POS 19% HMO 13% Carve-out 37% SOURCE: Within each type, figures based on the primary medical plan identified in a panel survey of 171 employers, conducted by Foster Higgins in NOTES: PPO is preferred provider organization. POS is point-of-service. HMO is health maintenance organization. a Not available. age limits. These data show that MH/SA services accounted for about percent of large employers' total medical plan costs in 1995, down from about 9 percent in For traditional indemnity plans, annual MH/SA costs per employee dropped from $237 in 1988 to $225 in 199. In contrast, total health care costs per employee grew from $2,786 to $3,7 over the same period. Some support for these results comes from the broader, more nationally representative annual surveys of employer health plans conducted by Foster Higgins. These surveys show that among employers with 500 or more employees, MH/SA costs represented 6 percent of total health plan costs in 19 and percent in In 1989, 58 percent of the panel survey employers had an EAP. Although originally designed to help employees address personal problems, by the late 1980s employers were increasingly integrating their EAPs with their health plans and having them serve a gatekeeper function. About a third of the EAPs offered by survey employers in 1989 handled MH/SA utilization review, and 10 percent were part of a mental health care provider network. At that time, 8 percent said that their EAP had decreased MH/SA costs, with roughly equal proportions of the remainder saying that it had increased or had not affected costs. By 1995, the use of EAPs had increased and their role had changed. In that year, 81 percent of panel respondents offered an EAP. A majority of these EAPs had a cost management role: 50 percent handled utilization review, and 27 percent were part of a mental health care network. Only 6 percent of the (1995) panel employers said that their EAP had increased costs; 70 percent said that it had decreased them; and the rest, that it had had no effect. Limits on mental health benefits changed from 1989 to 1995, based on employers' primary health plans (Exhibit 3). For inpatient care, these data show an increase in the use of limits on the maximum days per year and a decrease in the use of limits on maximum dollar amounts per lifetime. The use of other types of limits was unchanged. For outpatient care, use of limits decreased for expenditures and increased on annual visits. 123 HEALTH AFFAIRS - ] u I y I A u g u s t

5 EXHIBIT 3 Percentage Of Employers With Limits On Mental Health Benefits, 1989 And 1995 Type of limit Inpatient services Maximum number of days per year Maximum number of days per lifetime Outpatient services Maximum dollar amount per visit Maximum number of visits per year Maximum number of visits per lifetime % % _a SOURCE: Figures based on the primary medical plan identified in panel surveys of 171 employers, conducted by Foster Higgins in 1989 and a Not available. 12 Fewer than half of the panel employers had separate limits on substance abuse services in 1989 and Consequently, comparisons of changes in types of substance abuse benefit limits are probably not valid. However, in both years, percent of these employers had some type of limits on inpatient services. For outpatient treatment, the percentage of employers who had special limitations of any type for substance abuse benefits grew from 76 percent in 1989 to percent in HMOs Data from the annual AAHP survey provide information about changes in mental health benefits in HMOs since 1986 that are not available from the Foster Higgins panel study. (These data are not available for all years. Also, data for substance abuse services were not collected for most years, and data on other types of plans are not available for comparison.) These data show little change in the percentage of HMOs whose best-selling benefit packages included mental health benefits since At least 99 percent included coverage for outpatient mental health services throughout that period, and a minimum of about 91 percent included coverage for inpatient mental health care. The data do not include information about the small percentage of best-selling packages that excluded coverage for either inpatient or outpatient mental health care. In most of those cases, purchasers probably obtained coverage for mental health care from separate managed behavioral health care organizations. This may have been the purchasers' preference, or the HMOs may not have had affiliations or contracts with any mental health facilities. Benefit limits. AAHP survey data provide limited information about HMOs' use of benefit limits for MH/SA services. What is available, however, suggests that most HMOs use such limits. All but a few of HMOs' bestselling benefit packages had a maximum number of covered mental health outpatient visits per year in 199, most commonly the twenty-visit limit specified for federally qualified plans in the Health Maintenance Organization Act of 13. This represents an increase since 1988, when the percentage of HMOs with such limits was 87 percent. The percentage of HMOs with mental health inpatient day limits over the same period was essentially unchanged, with 9 percent having such limits in 199. The limit on mental health inpatient days also changed little, averaging thirty-one days in 199. For comparison, only HEALTH AFFAIRS - Volume 16, Number

6 T R E N D S : MENTAL HEALTH AND SUBSTANCE ABUSE 1 percent of HMOs had day or visit limits for general health care services in The percentage of HMOs with annual dollar limits for outpatient mental health coverage also was essentially unchanged since 1990, with 1 percent having such limits in 199. No data are available concerning the percentage of HMOs with annual dollar limits for inpatient coverage. Cost sharing. Almost no HMO members were subject to deductibles for either inpatient or outpatient mental health care in 1986 and 199 (Exhibit )7 For inpatient mental health services, the use of copayments nearly tripled over the period, with a slight increase in the use of coinsurance. Although the use of coinsurance for outpatient mental health services decreased, the use of copayments increased to the point that four-fifths of HMO members were subject to them by 199. Since HMOs that use coinsurance do not also use copayments, combining the percentages for each provides a clearer picture of cost-sharing trends. Using this procedure, we found that the use of either type of cost sharing for inpatient mental health care increased from 22 percent in 1986 to 1 percent in 199 and, for outpatient care, from 76 percent in 1986 to 91 percent in 199. Thus, while coinsurance rates or copayment amounts remained constant or declined slightly, the percentage of HMO members with such requirements increased significantly. Parity. A related issue is how the use of these provisions compares with those for general health care in HMO plans. Similar to mental health care, in 199 only a small percentage of HMO enrollees were subject to a deductible for general hospitalization or primary care visits. 8 Differences existed for copayments and coinsurance, however. For general hospitalization, 27 percent of HMO enrollees were subject to one of these forms of cost sharing, whereas 89 percent were subject to them for primary care visits. Most of these requirements were for copayments, with the average copay either $95.66 per day or $ per admission for general hospitalization and $6.92 for primary care visits. Corresponding 199 figures for mental health services were $57.63 per day or $282. per admission for inpatient care and $16.81 for outpatient care. Although the daily copayment amount for mental health hospital care was lower than that for general health care, it characterizes most copayment provisions for such care. In contrast, most copayments for general hospitalization are on a per admission basis. Thus, compared with mental health care, fewer enrollees were subject to cost sharing for general hospitalization in 199, and most commonly paid less when they were. For outpatient care, the percentages subject to cost sharing were similar, but copayments for general health care were less than half those charged for mental health visits. Discussion Although differences in health insurance coverage between physical and behavioral health care are generally acknowledged, we have lim- 125 EXHIBIT Percentage Of HMO Members With Cost Sharing For Mental Health Services, 1986 And 199 Type of cost sharing Copayments Coinsurance Deductibles Inpatient services % % 17 1 Outpatient services % % 10 1 SOURCES: Group Health Association of America, HMO Industry Profile, 1987 ed. (Washington: GHAA, 1987); and American Association of Health Plans, HMO and PPO Industry Profile: Annual Edition (Washington: AAHP, 19). NOTE: HMO is health maintenance organization. HEALTH AFFAIRS - J u 1 y I A u g u s t

7 126 ited information about the nature and extent of such differences. Coverage restrictions can affect the type, scope, or duration of services; annual or lifetime spending on these services; or requirements for cost sharing. Both of the studies whose data we examine in this paper have limitations. The Foster Higgins survey is based on a convenience sample of large employers and overrepresents those with very large numbers of employees. Nevertheless, it provides detailed information on how MH/SA benefits provided by these employers have changed over a six-year period. Although the AAHP survey of HMOs is more nationally representative, it does not include responses from all HMOs. Further, it is limited to information about the best-selling plans of those HMOs that do respond. Despite these limitations, the two studies jointly added to our knowledge of MH/SA insurance coverage by providing information not available from other sources. The results from the Foster Higgins survey suggest that nearly all large employers cover inpatient and outpatient MH/SA services in their health plans and that a majority cover intensive nonresidential treatment such as partial hospitalization. However, only a minority cover nonresidential treatment or say that it is their policy to cover MH/SA services to the same extent as they do other medical services. The percentage of those imposing annual limits on inpatient days or outpatient visits for MH/SA care has increased since 1989, with half or more having such limits in Perhaps partly as a result, the percentage of health plan costs accounted for by MH/SA services in 1995 was less than half that for HMO data from AAHP also show greater restrictions on MH/SA services compared with those on general health care. Most HMO consumers of mental health services in 199 were still subject to service limits that did not apply to primary care or general hospitalization. Cost-sharing requirements for MH/SA services in HMOs have increased since 1986, but we do not know how this change compares with those of other types of plans. These findings illustrate the complexity of insurance coverage and the various means by which health plans limit coverage for MH/SA and other services. They offer additional support for the findings of BLS surveys that show a general increase in the use of limits on MH/SA benefits. However, they do not provide information that is as complete as we might like about the use of specific limits or how they vary by plan or employer characteristics. The Substance Abuse and Mental Health Services Administration (SAMHSA) is now working with Foster Higgins, AAHP, and other organizations to improve these data and to better understand the impact of these provisions on access to and quality of care. The views expressed in this paper arc solely those of the authors and do not necessarily represent those of their respective organizations. NOTES 1. Mental Health Parity Act of 19, P.L U.S. Department of Labor, Bureau of Labor Statistics, Employee Benefits in Medium and Large Firms, 1988, Bulletin no (Washington: U.S. GPO, August 1989); and U.S. Department of Labor, BLS, Employee Benefits in Medium and Large Private Establishments, 19, Bulletin no. 256 (Washington: U.S. Government Printing Office, November 199). 3. Ibid. (19).. Foster Higgins, National Survey of Employer-Sponsored Health Plans (New York: Foster Higgins, Survey and Research Services, 1995). 5. Ibid. (19 and 1995). 6. American Association of Health Plans, HMO and PPO Industry Profile: Annual Edition (Washington: AAHP, 19). 7. Group Health Association of America, HMO Industry Profile, 1987 ed. (Washington: GHAA, 1987); and AAHP, HMO and PPO Industry Profile: Annual Edition. 8. AAHP, HMO and PPO Industry Profile: Annual Edition. HEALTH AFFAIRS - Volume 16, Number