This Web site is a component of the SAMHSA Health Information Network

This Site Mental Health at HHS All Mental Health

About the Program Mental Health Parity Resources Journal Articles Featured Publications In the News Related Links Organization & Financing Homepage  CMHS Programs:   Go to... Children's Campaign Community Support Emergency Services Financing HIV/AIDS Homelessness Managing Anxiety Mental Health Statistics Protection & Advocacy Special Programs State Planning Traumatic Events Suicide Prevention Programs Homepage  CMHS Activities:   Go to... 15+ School Violence Stigma Suicide Prevention Surgeon General

printer friendly page e-mail this page bookmark this page shopping cart  current or new account

This Web site is a component of the SAMHSA Health Information Network.

Organization & Financing

Journal Article Administration and Policy in Mental Health

Mental Health and Substance Abuse Services in Ten State Medicaid Programs

Jeffrey A. Buck
Judith L. Teich
Jay Bae
Joan Dilonardo

Jeffrey Buck, Ph.D., and Judith Teich, M.S.W., are with the Center for Mental Health Services, SAMHSA; Jay Bae, Ph.D., is with the Office of Strategic Planning, the Centers for Medicare and Medicaid Services; and Joan Dilonardo, Ph.D., is with the Center for Substance Abuse Treatment.

Address for correspondence: Jeffrey A. Buck, Ph.D., Center for Mental Health Services, SAMHSA, 5600 Fishers Lane, 15-87, Rockville, MD 20857. E-mail: jbuck@samhsa.gov.

Abstract
Program data for 1993 on Medicaid mental health and substance abuse (MH/SA) services and expenditures were developed from the Centers for Medicare and Medicaid Services (CMS) research files for 10 states. These data show that MH/SA service users are 7 to 13% of Medicaid enrollees. The percentage of Medicaid enrollees accounted for by MH/SA users increases with age, reaching a fifth of the 45-64 age group. Across the 10 states, MH/SA spending represents 11% of total Medicaid expenditures. When their expenditures for non-MH/SA services are also considered, MH/SA users account for 28% of total Medicaid expenditures.

KEY WORDS: Medicaid; mental health services; substance abuse

The Medicaid program accounts for a third of the public spending for mental health and substance abuse (MH/SA) treatment (McKusick, Mark, King, Harwood, Buck, Dilonardo, & Genuardi, 1998). Because it focuses on those in poverty or with disabilities, it is a particularly important program for adults with serious mental illnesses and children with serious emotional disturbances.

Despite the importance of the Medicaid program, only a few studies have examined MH/SA services within one or more states. One study of Michigan and California for 1984 found that spending for alcohol, drug abuse, and mental health services was 11 to 12% of total Medicaid expenditures. Persons using such services accounted for 9 to 10% of the Medicaid population (Wright & Buck, 1991). An update of this study for 1992 found that MH/SA recipients were 7 to 9% of enrollees. MH/SA expenditures were 7 to 10% of all Medicaid spending (Wright, Smolkin, & Bencio, 1995). However, the total of MH/SA and non-MH/SA spending for MH/SA users made up 21 to 24% of all Medicaid expenditures. More recently, Larson and colleagues (1998) examined Medicaid MH/SA services use for Michigan, New Jersey, and Washington for 1993. Using a broad definition of MH/SA utilization, this study found that 11 to 13% of enrollees had some MH/SA use.

The amount of this literature is limited for several reasons. First, the the Centers for Medicare and Medicaid Services (CMS) typically does not compile program statistics by diagnosis. Thus, information on MH/SA or other major conditions is not readily available. Second, due to the size and complexity of data sets, statewide Medicaid studies are extremely expensive. Third, until recently, Medicaid research files maintained by CMS were limited to four states: California, Georgia, Michigan, and Tennessee. Therefore, we do not know to what degree the experience of these states may be shared by others.

In 1992, however, CMS expanded the number of states for which it developed research files. The Larson et al. (1998) study is the first to use some of these data to examine statewide MH/SA services supported by Medicaid. Nevertheless, the expense of these studies still limits the number of states that can be studied. To address this problem, we developed a basic set of Medicaid MH/SA program statistics that could be inexpensively generated from these data. This approach permits a broader examination of Medicaid MH/SA services across states with varying eligibility and service coverage policies.

These statistics are generated, however, by sacrificing detail about some issues. Some services or service recipients may not be identified, and limited resources prevent detailed investigation of possible data anomalies. Further, we have not designed this approach to examine specific hypotheses or to explain possible interstate differences in results. Nevertheless, we believe that these limitations are minor compared with the benefits to be gained from increasing our general knowledge about MH/SA services in Medicaid.

Method
Data in this study were developed from CMS's State Medicaid Research Files, which provide uniform claims and eligibility data for selected states within the Medicaid Statistical Information System. Of the 27 states for which 1993 research files were available, 10 were selected: Delaware, Georgia, Kentucky, Maine, Missouri, Tennessee, Utah, Vermont, Washington, and Wisconsin. These states were selected based on criteria related to geographic diversity, completeness and quality of data, and limited penetration of Medicaid managed care. This latter criterion was necessary because 1993 data generally only reflect capitation payments for managed care enrollees, and do not allow classification of services or expenditures by diagnosis.

Within each state's files, we excluded data for certain individuals from analysis. These exclusions were most commonly due to the lack of information that would allow a complete picture of service utilization and expenditures. Individuals falling into any of the following categories were excluded:

1. Dually eligible for Medicaid and Medicare (mostly elderly)
2. Aged 65 and over
3. Enrolled in capitated plans
4. Missing gender or date of birth information

For all but Wisconsin, these exclusionary criteria eliminated between 10 to 23% of enrollees, mostly due to dual eligibility status (largely elderly). For Wisconsin, 45% of enrollees were eliminated from analysis, the chief reason being higher enrollment in capitated plans.

As part of the study, comparisons were made to all enrollees meeting the exclusionary criteria (including MH/SA users), regardless of diagnosis or service use. These are referred to as the "equivalent Medicaid population," and their associated expenditures as "equivalent Medicaid expenditures."

For several analyses, persons were classified and grouped based on Medicaid eligibility status. The "blind/disabled" group consisted of individuals who met eligibility standards through assistance under the Federal Supplemental Security Income (SSI) programs and related programs. Those in the "child" category included children in the Aid to Families with Dependent Children (AFDC) program as well as other children eligible through programs such as those for foster care children or for pregnant women and children. "Caretaker Relative or Pregnant Women" primarily consisted of AFDC adults and adults qualifying through the program for pregnant women and children. Other eligibility groups, such as aliens receiving emergency assistance, and those for which eligibility data were not available, were included in the "other/unknown" group.

MH/SA services were identified through a primary MH/SA diagnosis or category of service indicating MH/SA specialty care. Previous work has shown that primary diagnosis alone accounts for about 95% of cases that are identified through more sophisticated methods (Wright & Buck, 1991). Diagnoses were those that most payers consider to be MH/SA conditions (ICD-9 codes 291-292, 295, 296, 297-299, and 300-314). These did not include Alzheimer's disease, dementias and cognitive disorders; mental retardation and developmental delays; medical conditions related to alcohol or drug disorders (e.g., alcoholic cirrhosis of the liver); or MH/SA-related V codes (e.g., observation for mental conditions). MH/SA categories of service were inpatient psychiatric services for 21 and under and institutional psychiatric care for the aged. Further details about the data and the states in this study may be found in Buck, Miller, and Bae (2000).

Results
For the 10 states, most MH/SA service users received only MH services (86.4%), while less than 10% received only SA services (Table 1). Five percent of users had at least one instance of each type of service. The percentage of persons in the MH/SA group only using mental health services varied from a low of 80.5% to a high of 93.0%. Percentages using substance abuse treatment services were more variable across states, ranging from a low of 3.5% to a high of 14.0%.

¹ "MH Users" and "SA Users" are individuals with only MH or SA services use, and are not counted in the "MH and SA Users" category

Across all states, females represent approximately 60% of the MH group, and approximately the same percentage of equivalent enrollees. There was much more variability across states in the SA group, where the percentage of females ranged from 45 to 67%. In 6 of the 10 states, the proportion of females in the SA group was greater than the percentage of females among equivalent Medicaid enrollees.

While the under-21 group constitutes a significant majority of the Medicaid recipient population, they represent a smaller proportion of the MH/SA user group in each state. Approximately 45% of MH users in all 10 states were under age 21, compared with 65% of Medicaid enrollees for those states. Substance abuse service recipients were an even smaller part of the under 21 recipient population. Across the 10 states, only about 16% of MH/SA recipients with a substance abuse problem were younger than 21.

Health care expenditures for users of MH/SA services are displayed in Table 2. Equivalent Medicaid expenditures are defined as the total expenditures for all individuals, both MH/SA and non-MH/SA, meeting the study's selection criteria. The other columns show MH and SA expenditures, and non-MH/SA expenditures for users of MH/SA services. (Non-MH/SA expenditures represent spending for services that are often labeled as "physical" health care, such as surgery or treatment for infections.) This table indicates that users of MH/SA services account for a high share of Medicaid expenditures. Across the states, expenditures for MH and SA services for these users represent 11.1% of total equivalent Medicaid expenditures. When their expenditures for non-MH/SA services are also considered, they account for 27.9% of total equivalent expenditures. We did not break out non-MH/SA amounts separately for the mental health and substance abuse groups. However, other studies have shown that substance abuse is associated with multiple medical consequences and high costs (Harwood, 1998; Fox, Merill, Chang, & Califano, 1995).

Table 2
Expenditures for MH/SA Users as a Percentage of Equivalent Medicaid Expenditures, 1993

Source: State Medicaid Research Files, 1993

¹ Equivalent Medicaid expenditures represent total Medicaid spending for all enrollees meeting the selection criteria for the study (see Methods section). Amounts are in thousands of dollars.

² Percentages are the proportion of equivalent Medicaid expenditures accounted for by each category.

³ The non-MH/SA amount represents spending for non-MH/SA services for MH/SA users.

Table 3 reports MH/SA users as a percentage of equivalent Medicaid enrollees for five age groups. "Equivalent enrollees" are all enrollees meeting the study criteria, including those with MH/SA disorders. In general, the user rate increases with age. In the under 6 year-old group, MH/SA users represented only 3.3%, compared to the mean for all ages, which was 10.2%. In contrast, about one out of five Medicaid enrollees between ages 45 and 64 are users of MH/SA services. When recipients with substance abuse problems are segregated from the mental health group, the user rate generally peaks in the 21 to 44 age group. Adolescents with substance abuse problems represent a small part of the equivalent Medicaid group, constituting only 1.8% of equivalent Medicaid adolescents across all 10 states.

Table 3
MH/SA Users as a Percentage of Equivalent Medicaid Enrollees, by Age Group, 1993¹

Source: State Medicaid Research Files, 1993

¹ Percentages represent the proportion of equivalent Medicaid enrollees represented by MH/SA users in that category.

Table 4 presents expenditures for MH/SA services as a percentage of equivalent Medicaid expenditures, by age group. While the overall mean for all ages is 11%, there is a fair amount of variability-from a low of 3.0% for the under 6 group, to a high of nearly a quarter of total dollars for the 6 to14 year-old group. The data clearly demonstrate that MH/SA expenditures account for a significant share of Medicaid expenditures, particularly for the child and young adolescent population. Since the substance abuse benefits offered by Medicaid have traditionally been limited, SA expenditures for persons with substance abuse problems are relatively small, representing only 1.5% of all equivalent Medicaid expenditures across the 10 states.

Table 4
MH/SA Expenditures as a Percentage of Equivalent Medicaid Expenditures, by Age Group, 1993

Source: State Medicaid Research Files, 1993

¹ Percentages are the proportion of equivalent Medicaid expenditures accounted for by each category.

² Equivalent Medicaid expenditures represent total Medicaid spending for all enrollees meeting the selection criteria for the study (see Methods section). Amounts are in thousands of dollars.

MH/SA users are fairly evenly distributed among the major eligibility groups, with 36% in the blind/disabled group, 32% in the child group, and 29% in the group consisting of caretaker relatives or pregnant women. Because the total size of each of these groups differ, however, they constitute different proportions of each. Table 5 shows the percentage of equivalent Medicaid enrollees in each eligibility group who are users of MH/SA services. These data show some degree of variability across the three major eligibility groups, presumably due to differences in state eligibility thresholds. MH/SA service users make up a higher proportion of the blind/disabled category than they do of other eligibility groups: percentages range from 16% in Delaware to 31.2% in Vermont. In contrast, the corresponding percentages for the child eligibility category range from 2.1% in Delaware to 9.4% in Vermont. A similar relationship was noted when recipients with substance abuse problems were considered separately; that is, the percentage of equivalent enrollees was highest in the blind/disabled category, while the child eligibility category represented less than 1% of equivalent enrollees in each state.

Table 5
MH/SA Users as a Percentage of Equivalent Medicaid Enrollees, by Basis of Eligibility, 1993¹

Source: State Medicaid Research Files, 1993

¹ Percentages represent the proportion of equivalent Medicaid enrollees represented by MH/SA users in that category.

While Table 5 shows that the MH/SA users represented 26.3% of equivalent enrollees in the blind/disabled category, Table 6 indicates that MH/SA expenditures for this group represent a much lower proportion (13.2%) of equivalent Medicaid expenditures. A similar relationship was observed in expenditures for persons with substance abuse problems in the blind/disabled and the caretaker relative/pregnant women eligibility categories. While persons with SA represented 4.1% of equivalent enrollees in the blind/disabled category, SA services represent 1.7% of equivalent Medicaid expenditures. Similarly, while persons with SA problems in the caretaker/pregnant women category represented 2.3% of equivalent enrollees, expenditures for SA treatment services represented only 2% of equivalent Medicaid expenditures. This is probably due to the need for high-cost physical health care for the disabled and pregnant women groups as a whole, and/or the provision of less costly or fewer mental health and substance abuse treatment services in comparison to other types of medical services.

Table 6
MH/SA Expenditures as a Percentage of Equivalent Medicaid Expenditures, by Eligibility Group

Source: State Medicaid Research Files, 1993

¹ Percentages are the proportion of equivalent Medicaid expenditures accounted for by each category.

² Equivalent Medicaid expenditures represent total Medicaid spending for all enrollees meeting the selection criteria for the study (see Methods section). Amounts are in thousands of dollars.

Discussion

Several qualifications apply in interpreting the program statistics that we have presented. First, in using simplified procedures for identifying MH/SA services and users, our statistics represent a conservative assessment of MH/SA utilization and expenditures within these programs. However, other work tends to show that the effect of this approach is minor (Larson et al., 1998; Wright & Buck, 1991). Second, the chief effect of the selection criteria for the study was to eliminate enrollees 65 and over. Additionally, we limited our definition of MH/SA services to diagnoses used by most payers. These two features probably explain much of any differences between our study and others, which may have included the elderly and/or those with dementias or related disorders in their results. Third, by focusing on actual service use, our study provides no information about the level of prevalence of MH/SA disorders in this population.

Despite differences in methods, the results of this study generally agree with other multi-state studies of Medicaid MH/SA services. These studies have shown that MH/SA service users constitute 7 to 13% of the Medicaid population, depending on the particular state and the broadness of identification criteria. Our study showed individual state results almost wholly within this range.

Comparisons of relevant findings for expenditures were more variable, but still fairly similar. In our study, spending for MH/SA services across all states was 11% of the total equivalent Medicaid amount. The comparable range from the 1992 study by Wright and his colleagues (1995) was 7 to 10%. For all spending on MH/SA users, the Wright study found a range of 21 to 24%, while our study found 28% for all states combined.

While some of our findings generally support those from other studies, others offer new information about the characteristics of Medicaid MH/SA users and expenditures. First, similar to the general Medicaid population, MH/SA users are most likely to be females. Second, because there is little or no MH/SA services utilization among preschool children, MH/SA users are older than the typical Medicaid enrollee. The percentage of Medicaid enrollees accounted for by MH/SA users increases with age, reaching a fifth of the 45 to 64 age group. Third, although few Medicaid dollars go for MH/SA services for preschoolers, nearly a quarter of all Medicaid spending for the 6 to 14 age group goes for such spending. Most of this amount is for MH, not SA services. Finally, although MH/SA users make up a quarter of the blind and disabled eligibility group, MH/SA expenditures only represent about half of that percentage for Medicaid spending for that group. This may be due to higher expenses for the care of those with physical disabilities.

A major benefit of this study is the extension of what we previously knew about Medicaid MH/SA services to many more states. Since previous studies have been based on the experience of only two or three states, their generalizability has been questioned. Our study shows that, despite variability in eligibility and service provisions among the states that we studied, there is considerable consistency for most results. Combined with similar studies, it greatly increases support for the conclusion that spending for MH/SA services and MH/SA service users is an important component of overall Medicaid spending. However, it also documents that the large proportion of this spending is for MH, not SA services.

The demonstration that MH/SA users account for a significant proportion of Medicaid spending suggests that they could be significantly affected by cost-saving measures under managed care. Our findings are based on claims data for a period prior to the extensive employment of managed care plans in Medicaid. Therefore, they provide a benchmark by which to assess the impact of subsequent managed care programs. Since many Medicaid managed care plans are targeted to particular age or eligibility groups, our study is particularly relevant for this purpose. In addition to providing information for specific states, the range of results suggests minimum criteria for evaluating plans for which pre-managed care data may not be available.

References
Buck, J.A., Miller, K., & Bae, J. (2000). Mental health and substance abuse services in Medicaid, 1993 (DHHS Publication No. SMA 99-3368). Rockville, MD: Center for Mental Health Services, Substance Abuse and Mental Health Services Administration.

Fox, K., Merill, J.C., Chang, H.H., & Califano, J.A., Jr. (1995). Estimating the costs of substance abuse to the Medicaid hospital care program. American Journal of Public Health, 85,48-49.

Harwood, H.J., Fountain, D., & Livermore, G. (1998). The economic costs of alcohol and drug use in the United States (NIDA & NIAAA report). Washington, DC: U.S. Department of Health and Human Services.

Larson, M.J., Farrelly, M.C., Hodgkin, D., Miller, K., Lubalin, J.S., Witt, E., McQuay, L., Simpson, J., Pepitone, A., Keme, A., & Manderscheid, R.W. (1998). Payments and use of services for mental health, alcohol, and other drug abuse disorders: Estimates from Medicare, Medicaid, and private health plans. In R. Manderscheid & M. Henderson (Eds.), Mental health, United States, 1998 (DHHS Publication No. SMA 99-3285), pp. 124-141. Washington, DC: U.S. Government Printing Office.

McKusick, D., Mark, T., King, E., Harwood, R., Buck, J.A., Dilonardo, J., & Genuardi, J.S. (1998). Spending for mental health and substance abuse treatment, 1996. Health Affairs, 17(3), 147-157.

Wright, G.E., & Buck, J.A. (1991). Medicaid support of alcohol, drug abuse, and mental health services. Health Care Financing Review, 13, 117-128.

Wright, G., Smolkin, S., & Bencio, D. (1995). Medicaid mental health and substance abuse 1992 use and expenditure estimates for Michigan and California: Final report (Reference No. 8231). Washington DC: Mathematica Policy Research.