Mental Health and Substance Abuse Services in Medicaid, 1995

Jeffrey A. Buck
Kay Miller

U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES

Substance Abuse and Mental Health Services Administration

Center For Mental Health Services

ACKNOWLEDGEMENTS

Jeffrey A. Buck is with the Center for Mental Health Services (CMHS), Substance Abuse and Mental Health Services Administration. Kay Miller is with The MEDSTAT Group. Eva Witt and Jon Blake of The MEDSTAT Group contributed valuable computer programming assistance in creating these tables.

This report would not have been possible without the assistance and support of the Centers for Medicare and Medicaid Services, U.S. Department of Health and Human Services (DHHS).

DISCLAIMER

Material for this report was prepared under contract by The MEDSTAT Group for the Substance Abuse and Mental Health Services Administration (SAMHSA), U.S. Department of Health and Human Services (DHHS). The content of this publication does not necessarily reflect the views or policies of CMHS, SAMHSA, or DHHS.

PUBLIC DOMAIN NOTICE

All material appearing in this report is in the public domain and may be reproduced or copied without permission from SAMHSA or CMHS. Citation of the source is appreciated. However, this publication may not be reproduced or distributed for a fee without the specific, written authorization of the Office of Communications, SAMHSA, DHHS.

ELECTRONIC ACCESS AND COPIES OF PUBLICATION

This publication can be accessed electronically through the following Internet World Wide Web connection: www.samhsa.gov. For additional free copies or this document, please call SAMHSA's Mental Health Services Information Clearinghouse at 1-800-789-2647.

RECOMMENEDED CITATION

Buck, J.A. and Miller, K. Mental Health and Substance Abuse Services in Medicaid, 1995. DHHS Pub. No. (SMA) 02-3713. Rockville, MD: Center for Mental Health Services, Substance Abuse and Mental Health Services Administration, 2002.

ORIGINATING OFFICE

Office of the Associate Director for Organization and Financing, Center for Mental Health Services, Substance Abuse and Mental Health Services Administration, 5600 Fishers Lane, 15-87, Rockville, MD 20857.

DHHS Publication No. (SMA) 02-3713
Printed 2002

CONTENTS

• Introduction
• 1995 Table Notes
  • Population Selection Criteria
  • Definition of Terms
  • Exclusions
  • Categorization Variables
  • Description of Tables
  • Data Quality Review
  • Reasonability Analysis
  
  
  
• Tables (1995)
  • All States
  • Alabama
  • Arkansas
  • Delaware
  • Georgia
  • Kansas
  • Kentucky
  • New Hampshire
  • New Jersey
  • Vermont
  • Wyoming
• Supplemental Table Notes

Introduction

The Medicaid program accounts for one-third of public spending for mental health and substance abuse (MH/SA) treatment (Mark et al., 2000). Because it partly focuses on those with disabilities, it is a particularly important program for adults with serious mental illness and children with serious emotional disturbance.

Despite the importance of the Medicaid program, only a few studies have examined MH/SA services within one or more States. One study of 1984 Michigan and California data found that spending for alcohol, drug abuse, and mental health services was 11 to 12 percent of total Medicaid expenditures. Persons using such services accounted for 9 to 10 percent of the Medicaid population (Wright & Buck, 1991). An update of this study using 1992 data found that MH/SA recipients were 7 to 9 percent of enrollees. MH/SA expenditures were 7 to 10 percent 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 percent 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 percent of enrollees had some MH/SA use.

There are several reasons for this limited literature. First, the Centers for Medicare and Medicaid Services (CMS, formerly HCFA), the agency that administers the Medicaid program, 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.

To address these problems, a basic set of Medicaid MH/SA program statistics was developed for the non-elderly population that could be inexpensively generated from Medicaid data that States submit to CMS. These statistics are presented in a uniform set of tables for each of 10 selected States, plus a set of tables that aggregates data across all 10 States. These tables provide a range of information on user characteristics, service utilization, and expenditures for each State. The fundamental purpose is to provide policy makers, interest groups, and others with basic information on MH/SA services and expenditures in Medicaid, and associated trends. Table sets for 1986 - 92, 1992, 1993, and 1994 have already been published (Buck, Miller, & Bae, 2000a, 2000b, 2000c, 2000d).

States were selected for analysis based on geographic diversity, completeness and quality of data, and limited penetration of Medicaid managed care. This latter criterion was included because Medicaid managed care data generally reflect only capitation payments for managed care enrollees, and do not allow classification of services or expenditures by diagnosis. Within each State's files, certain individuals were excluded 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 the following categories were excluded:

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

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 percent of cases identified through more sophisticated methods (Wright & Buck, 1991). Diagnoses were those considered by most payers to be MH/SA conditions (ICD-9 codes 291-292, 295, 296, 297-299, and 300-314). These did not include Alzheimer's disease, other dementias and cognitive disorders; mental retardation and developmental delays; medical conditions related to alcohol or drug disorders (e.g., alcoholic cirrhosis of 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.

General Recommendations for Using Tables

In the following sections, notes are provided on terminology, selection criteria, categorization variables, table descriptions, and data quality issues. Users of these tables should review this material with particular attention to the following:

1. Users should exercise caution in interpreting these statistics. Tables are based on data that States submit to CMS and that have not been validated in any way. Where possible, notes identify anomalies or other issues to be aware of when interpreting the statistics. However, possible causes of these anomalies have not been investigated.
   
   
2. Some tables may be difficult to understand at first glance. An effort has been made to provide necessary explanatory material within each table. Nevertheless, users should review term definitions and table-specific comments to ensure proper interpretation of the statistics. Sometimes definitions may vary slightly across tables (e.g., the definition of a "user").
   
   
3. In some tables, comparisons are made to all Medicaid enrollees in the State who meet the exclusionary criteria (including MH/SA users), regardless of diagnosis or service use. These are referred to as "total equivalent enrollees," and their associated expenditures as "total equivalent Medicaid expenditures." These categories constitute the denominator for some percentages. In other tables, the term "equivalent Medicaid population" refers to all enrollees in the study who share a particular characteristic (e.g., have at least one inpatient stay of any kind).
   
   
4. Some care also should be taken when comparing statistics for 1995 with those published for 1994, 1993, 1992, or 1986-1992. Sometimes changes over time can be attributed to changes in category definitions rather than to changes in patterns of care. For example, both the Type of Service and Basis of Eligibility categories are subject to change on a yearly basis.
5. This document contains tables that are judged to be most useful for those interested in Medicaid MH/SA services. In addition to these, other tables were produced through this project which may be of interest to some groups. These tables provide additional information on racial/ethnic characteristics, detoxification claims, institutional services, and high-cost users. They may be accessed at the following web address: www.samhsa.gov.

References

Buck, J.A., Miller, K., & Bae, J. (2000a). Mental Health and Substance Abuse Services in Medicaid, 1986-1992. DHHS Pub. No. (SMA) 99-3366. Rockville, MD: Center for Mental Health Services, Substance Abuse and Mental Health Services Administration.

Buck, J.A., Miller, K., & Bae, J. (2000b). Mental Health and Substance Abuse Services in Medicaid, 1992. DHHS Pub. No. (SMA) 99-3367. Rockville, MD: Center for Mental Health Services, Substance Abuse and Mental Health Services Administration.

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

Buck, J. A., Miller, K., & Bae, J. (2000d). Mental Health and Substance Abuse Services in Medicaid, 1994. DHHS Pub. No. (SMA) 00-3284. Rockville, MD: Center for Mental Health Services, Substance Abuse and Mental Health Services Administration.

Harwood, H. J., Fountain, D., & Livermore, G. (1998). The Economic Costs of Alcohol and Drug Use in the United States. NIDA & NIAAA report, US Department of Health and Human Services, Washington, D.C.: US Government Printing Office.

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 Manderscheid, R. and Henderson, M., (Eds.), Mental Health, United States, 1998 (pp. 124-141). DHHS Pub. No. (SMA) 99-3285. Washington, DC: U.S. Government Printing Office.

Mark, T., Coffey, R. M., King, E., Harwood, H., McKusick, D., Genuardi, J., Dilonardo, J., & Buck, J. A. (2000). Spending on mental health and substance abuse treatment, 1987-1997. Health Affairs 19(4), 108-120.

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. Ref. no. 8231, Washington DC: Mathematica Policy Research.

Back to Contents

1995 Table Notes

A set of 10 tables provides an overview of Medicaid utilization and expenditures for people being treated for mental health and/or substance abuse (MH/SA) conditions during calendar year 1995. Each set of tables was created for 10 states both separately and as part of an "All States" table set. The data were derived from the Centers for Medicare and Medicaid Services' (CMS's) State Medicaid Research File (SMRF) database.

This section describes the definitions and methodologies used to create the tables. It compiles all decisions made during the design process to help users understand the data and how the tables were created. It contains the following sections:

• Population Selection Criteria
  
• Definition of Terms
  
• Exclusions
  
• Categorization Variables
  
• Description of Tables
  
• Data Quality Review
  
• Reasonability Analysis

Population Selection Criteria

The intent of these tables is to present utilization and expenditure statistics using a conservative and easily interpreted definition of the MH/SA population rather than a broader and potentially more complicated one. The goal of this selection process was to find people who were being treated for a MH/SA condition, not to find every person with a MH/SA condition. To this end, no secondary diagnosis codes were used. Because not all claims carried a diagnosis code, the definition was expanded to include Type of Service codes. Therefore, the person/claim was classified as MH/SA if any one of the following criteria were present:

Figure 1: Description of Diagnosis Codes Included

Back to Contents

Definition of Terms

Recipient
A recipient is a person enrolled in Medicaid for at least one month and who had at least one Medicaid service in the calendar year. Recipients with partial year enrollment (less than 12 months) were NOT excluded from these tables.

Enrollee
An enrollee is a person enrolled in Medicaid for at least one month during the calendar year regardless of any service use. Persons with partial year enrollment (less than 12 months) were NOT excluded from these tables.

Total Equivalent Recipients/Enrollees/Users/Medicaid Expenditures
To compare to the MH/SA sample, information on the full population (including MH/SA users) is reported on most tables, as applicable (i.e., Recipients and/or Enrollees, Users, or Expenditure totals). To make the populations comparable, the same exclusion rules are applied (see Exclusion section) that are applied to the MH/SA sample.

Crossovers
These are people dually enrolled in Medicaid and Medicare with at least one claim during the calendar year paid by both payers.

Users
These are counts of people who met the criteria described in the MH/SA User Selection Criteria section. No adjustments are made in user counts for partial year enrollees/recipients (thus, a person counts once regardless of length of enrollment). Identification numbers used in these files are unique to an enrollee regardless of gaps in enrollment.

Expenditures
These are Medicaid payments, shown in thousands of dollars

MH/SA Expenditures
These are Medicaid payments made for care for MH/SA conditions. These services are identified based on either primary diagnosis or type of service (see MH/SA user selection criteria). Therefore, if no diagnostic information is reported for a particular service and it does not fall into specifically included types of service (e.g., inpatient psychiatric), the expenditures will not be reported as MH/SA expenditures, but rather reported in as "Non-MH/SA Service" expenditures. This is particularly important to note when considering that some types of service do not routinely report a diagnosis code, but could be related to MH/SA care. For example, certain drugs are prescribed for the treatment of MH/SA conditions but pharmacy claims do not carry a diagnosis code. While drug claims are the most obvious example of this, any claim that does not include a diagnosis code (or categorized based on type of service (TOS)) cannot be attributed to a MH/SA condition.

Length of Stay
This represents the number of inpatient days from the beginning date of service to the ending date of service for each hospital stay. These tables use customary hospital rules to calculate the length of stay, which count the day of admission, but not the day of discharge. In addition, if, on the last day represented on the claim, the patient has not been discharged, that day is counted as well. Finally, when a stay begins and ends on the same day, the length of stay is set to one. As with Days of Care, all days are included in these totals, even those from stays reporting the recipient was "still a patient."

Inpatient Stays
Stays reported in these tables reflect only acute inpatient admissions in general hospitals. They do not include inpatient psychiatric or long-term care facility stays. They also do not include stays in which discharge status indicated the recipient was "still a patient." These two exclusions allow comparability across tables and take into account nuances, that if not incorporated, could cause misinterpretation of results.

Inpatient Psychiatric and Long Term Care (LTC) services are reported in separate categories from Acute Inpatient. No report of "stays" are included for these services because these services are not bundled into one claim per stay. Therefore, there may be more than one claim per stay especially for LTC services where the provider usually submits claims to Medicaid on a weekly or monthly basis. These services potentially can span years if all claims are combined, making stay building a difficult, if not impossible task.

The requirement that a claim not show a discharge status of "still a patient" prevents the over-counting of stays where the stay was not determined correctly and actually resides on the claims file as more than one claim per stay. However, if benefit limits or other restrictions cause a final piece of a claim to never be submitted, the potential exists for under-counting stays. Situations where there are unusually high numbers of these claims should be investigated during analysis ( Data Quality Review-Stays with Discharge Status of "Still a Patient" section for more details).

Outpatient Visits
Visits reported are derived from the quantity field on the outpatient claims file, that represent the number of visits paid on each claim.

Top 10% of Equivalent Medicaid Enrollees
These enrollees are those with the highest annual costs (the highest decile), based on individuals' total Medicaid expenditures for all services. They also meet the same criteria defined above for the "equivalent Medicaid enrollee" (e.g., enrollees age 65 and over, etc.).

High Cost MH/SA Users
These users are those MH/SA users who are included in the "Top 10% of Equivalent Medicaid Enrollees" (see above).

Back to Contents

Exclusions
Certain populations were excluded from these tables. The most important reason was the absence of complete data for these populations that would prevent a full picture of their utilization. While these tables are designed to represent full utilization and expenditures for the populations included, they may not fully represent an entire state's Medicaid population. The following are the population groups excluded and the basic reason for exclusion (Figure 2 gives the counts of people excluded, by state, for each category):

Figure 2: Number of Enrollees Excluded from Full Medicaid Population

Note: Because a person could fall into more than one category, the above counts are listed in the order in which the rules were applied (in order of priority) to ensure against duplication.

1. Crossover Recipients: Crossover recipients are excluded because their expenditures would not reflect the total cost of their care. For these recipients, Medicare is their primary payer and, therefore, the Medicaid expenditures would represent only those services not covered by Medicare either in total or in part. Also, the quality of data on crossover claim is usually of a lower quality than non-crossover claims.
   
   
2. Enrollees age 65 and over: Those enrollees over 65 who are not crossovers (see above) would not truly reflect those people in this population group. In most cases, they represent a very small percentage of the over 65 enrollees. It was felt that the data would be more accurate if they were excluded.
   
   
3. People who enrolled in capitated plans at any time during the year: Encounter data was not available for those services received under capitated arrangements. The only services included in the data available would be those services that occurred during months they were not enrolled in managed care, fees related to their managed care (premium payments, etc.) and services outside of the managed care program that were paid fee-for-service. Therefore, the services that would be reported would not fully reflect their health care under Medicaid and were excluded. Nationally, CMS reports that 28 percent of the Medicaid population was enrolled in managed care in 1995, although not all of these managed care programs included capitation payments.
   
   
4. People who had no enrollment during the year (ineligible recipients): There are situations in some states where claims are adjudicated without a valid eligibility record (e.g., retroactive eligibility, manual adjudication, etc). For these situations, no eligibility record exists, therefore, demographic information was not available for these recipients. Because a number of tables give breakdowns by basic demographic categories, it was decided to exclude them. The numbers here are very small (< 1% in most cases).
   
   
5. Enrollees with missing date of birth or sex: The reasons for exclusion here are the same as the category above.

Back to Contents

Categorization Variables

Age Group
Age is computed using the date of birth. It reflects the age of the recipient/enrollee on December 31 of the file year. The following groupings are used for reporting age:

Under 1
1-5
6-14
15-20
21-44
45-64

Note: People 65 years of age or older are excluded from these tables as well as those with unknown dates of birth.

Sex
Male Female

Note: People whose gender is unknown are excluded from these tables.

Basis of Eligibility
The basis of eligibility defines the broad category under which the enrollee was eligible for Medicaid. The categories used are those defined by CMS for federal reporting requirements, but have been grouped slightly for these tables:

• Blind/Disabled: This includes people who meet the federal requirements for Supplement Security Income (SSI) or other programs on the basis of blindness or disability
  
  
• Child: This includes Aid to Families with Dependent Children (AFDC) children as well as other children eligible through special programs such as Ribicoff Children, Foster Care Children or children eligible under the medically needy or poverty related eligibility categories. This category does not include children who qualify for Medicaid under the Blind/Disabled eligibility determination.
  
  
• Caretaker Relative or Pregnant Woman: This includes adults qualifying under the AFDC rules or other special program such as the medically needy or poverty related eligibility categories.
  
  
• Other Title XIX (includes Aged category): This includes people who were not eligible in the above categories. This will include people such as aliens receiving emergency assistance, 1115 waiver recipients who do not qualify elsewhere and situations where a State has expanded the coverage using more liberal income limits. In addition, if anyone remained in the file after the "age over 65" exclusion who showed an "aged" eligibility category, they were defaulted to this category. Finally, this does not include the non-Medicaid eligible such as the 100% state only eligibles (i.e., general assistance) or 100% federal only eligibles such as Title V or Refugees. This database only includes those who are "Medicaid (i.e., Title XIX) eligible."
  
  
• Basis of Eligibility Unknown: This category, while small, captures those enrollees whose eligibility category was not reported on the eligibility files used for this project.

It is important to note that these eligibility categories represent the "Basis of Eligibility" for medical care as defined by CMS. "Maintenance Assistance Status" categories which are related to determination of income and assets such as "Categorically Needy" (with and without cash assistance) and "Medically Needy" as well as other coverage groups created by specific legislation (e.g., "Poverty Related" categories) cannot be separately identified using these tables. For example, people qualifying under a provision for the medically needy will be found in all "Basis of Eligibility" categories, as appropriate, based on their eligibility group.

MH vs. SA
The diagnosis code ranges listed in the "MH/SA User Selection Criteria" section have been split to separate Mental Health users from Substance Abuse users. If the determination of a MH/SA condition was made solely by using Type of Service the default category used was mental health. Care was taken not to classify an individual claim as substance abuse based on diagnosis and mental health based on Type of Service.

Because only primary diagnosis was used, a claim could only qualify as MH or SA, but a person could qualify for both if at least one claim during the year qualified under each definition. For those claims classified as MH based on Type of Service, there is a small chance of misclassification of a person as having both MH and SA if the only claim that was classified as MH was based on Type of Service.

Diagnostic Category
Each recipient was assigned to a diagnostic category based on their most frequent diagnosis category in a year. In case of a tie, each file type was inspected to find the first MH/SA diagnosis code assigned to a claim. The files were evaluated in the following order: Inpatient, Outpatient and Long Term Care. Therefore, in the case of a tie, if one or more MH/SA claims were found on the Inpatient file, the first one occurring in the year based on date of service was used. If no MH/SA diagnosis was found on the Inpatient file, the same criteria were applied to the Outpatient file, etc. The diagnostic categories used are as follows:

Type of Service (TOS)
Services were categorized into broad categories. These categories were based on CMS's 2082 Type of Service categories. See Figure 3 for details of how the CMS categories are grouped into the broad categories used for these tables.

Back to Contents

Description of Tables

Table 1: MH/SA Users by Sex and Age Group
This table gives a broad overview of the population reflected in this table set. It subsets the MH/SA population into those with only mental health, only substance abuse, both mental health and substance abuse and total by sex and age group. It then compares these users to the total equivalent enrollees and the total equivalent recipient population.

Figure 3: Services Included in Broad Type of Service Categories

Table 2: MH/SA Users by Diagnostic Category and Age Group
This table looks at users within broad diagnostic categories as well as subsets therein. It also looks at the percentage of people within each diagnostic grouping to see what percentage also have a co-occurring condition. The co-occurring condition is defined as either MH or SA depending on the whether the original condition is MH or SA. That is, if the original condition is MH, it looks at the percentage with any SA condition or vise versa. It is important to note that these numbers only represent individuals with dual use of MH and SA services, and not all those with co-occurring disorders.

Table 3: MH/SA Users by Basis of Eligibility and Age Group
This table is similar to Table 1 except that the demographic category used is the "Basis of Eligibility" which defines the enrollment category under which the enrollee/recipient was eligible for Medicaid.

Table 4: Total MH/SA Expenditures (in Thousands) by Basis of Eligibility and Age Group
This breaks the expenditures found for the MH/SA population into Mental Health, Substance Abuse or Non-MH/SA Related expenditures. These expenditures are then compared to the expenditures for the equivalent Medicaid population (see note under Measures for MH/SA Expenditures). The expenditures are the broken down by Basis of Eligibility and age group.

• Expenditure totals are rounded to the nearest thousand for presentation in the tables. The full numbers are used when creating the totals, which may cause some variation in the reported totals.

Table 5: Expenditures for MH/SA Users by Type of Service and Age Group
This table is similar to Table 4 except that instead of Basis of Eligibility categories, the table reports expenditures by major type of service category (see Figure 3) and by age group.

• The outpatient category is divided into two categories depending on whether a diagnosis code is expected on the type of service (e.g., A diagnosis code is expected on Physician claims, but not on dental claims)
• Expenditure totals are rounded to the nearest thousand for presentation in the tables. The rounding is done at the time the tables are populated. The full numbers are used when creating the totals, which may cause some variation in the reported totals.

Table 6: Acute Inpatient Stays for MH/SA Users by Sex and Age Group
This table looks at people who are classified as a MH/SA user (using all claims, not just inpatient claims) who were also inpatient users (see bullet three below). These stays are then categorized into All IP stays, MH, SA and Total MH/SA stays and compared with the equivalent Medicaid population.

• This includes only "Acute Inpatient" stays and not stays in Inpatient Psychiatric facilities.
• "All IP stays for MH/SA Users" represent all inpatient stays by MH/SA users, regardless if they had a stay for MH or SA treatment
• Only Primary Diagnosis was used to determine the cell in the table where stay is reported (see MH vs. SA section for further rules).

Table 7: MH/SA Acute Inpatient Users by Sex and Age Group
This is the compliment to Table 6 in that it shows the counts of users who were included in the counts of stays reported in Table 6. All the rules described in Table 6 apply to this table as well.

• The criterion for inclusion into this table was at least one inpatient "stay" (see Definition of Terms) during the year.
• This differs from Table S4, which defines a user as having at least one paid day of care during the year (see Table S4 section for further details). Therefore, if a person's only stay showed that they were "still a patient" at the end of the year, but reported "days of care", they would be counted in Table S4 and not Table 7. Conversely, if they had an inpatient stay with a discharge, but the claim did not report any covered days, they would be counted as a user in Table 7 and not Table S4. Either of these two situations happens only rarely. See Data Quality Review section for more information regarding the magnitude of these situations.

Table 8: Length of Stay for MH/SA Related Acute Inpatient Stays by Sex and Age Group
This table reports length of stay statistics: mean, median and 90th percentile for mental health and substance abuse related stays by sex and age group. It also calculates comparative statistics for the equivalent Medicaid population.

• This table includes only "Acute Inpatient" stays and not stays in Inpatient Psychiatric facilities.
• Only Primary Diagnosis was used to determine the cell in the table where stay is reported.
• This uses length of stay, calculated from date fields (see Definition of Terms section for more detailed discussion of rules employed).
• This uses the claims level data, not person level data.

Table 9: Outpatient Visits for MH/SA Users by Sex and Age Group
This table is the only table that attempts to quantify outpatient utilization. It breaks the visits into categories based on whether the service was for mental health, substance abuse or non-MH/SA related care. The standard breakdown of sex and age group is used for the row breakdowns.

• It counts visits using the quantity field on claims with a broad Type of Service of "Outpatient."
• Only visits from outpatient claims with a diagnosis codes were included in this table. This should not be confused with the distinction made in Table 6, which breaks the Outpatient TOS into two groups: "with Expected Dx" and "w/o Expected Dx" (see Figure 3 for further definitions). Table 13 includes visits from both of these two groups, but only if the claim reports a diagnosis. Figure 4 shows how the Type of Service breakdown relates to these two broad types of service.

Table 10: High Cost MH/SA Users by Sex and Age Group (new for 1995)
This table gives an overview of the high cost MH/SA users. It looks at users and expenditures for this population by sex and age group. It also looks at what percentage the population represents of both the MH/SA users and the top 10% of Equivalent Medicaid Enrollees. Finally, it looks at the expenditure percentages for this population for MH/SA services, other services and total services compared with the expenditures for the full top 10% of Equivalent Medicaid Enrollees.

Back to Contents

Data Quality Review

This section describes areas that require further explanation or problems that were encountered during the creation of these tables.

Stays with Discharge Status of "Still a Patient."
In most states, the percentages of stays excluded for this reason was less than eight percent. However, for two of the states included in these tables, the number of MH/SA claims with a discharge status of "still a patient" was either high in general or higher in the MH/SA population. The important consideration here is how much difference exists between Non-MH/SA and the MH/SA claims. This may relate to the length of psychiatric admissions as well as potential day limits for psychiatric coverage which would cause a MH/SA claim to never be fully billed (potentially only the covered days were submitted for payment).

Figure 5 shows the difference in the distribution of discharge status values between the MH/SA population groups and the full Medicaid inpatient claims SMRF file in each state.

Figure 5: Distribution of Discharge Status

Reasonability Analysis
As noted earlier, statistics presented in these tables have been developed from Medicaid SMRF data and have not been validated. However, they were reviewed for basic reasonableness, using a cross-state review process. Selected columns from selected tables for each state were combined together to allow review of the same statistics across states. While users always should be aware of anomalies resulting from small cell sizes, other discrepancies also are noted below.

Table 1
The results appeared to be reasonable across states.

Table 2
The results appeared to be reasonable across states.

Table 3
The results appeared to be reasonable across states.

Table 4
The expenditures per user (computed using both Tables 3 and 4) appeared to be reasonable across states except for some dramatic differences in some Blind/Disabled populations due to extremely small sample sizes. These differences were most notable in the Under 1 age category across all states and to a lesser extent across all ages for Delaware, New Hampshire and New Jersey.

Table 5
Georgia is the only one of the ten states that does not provide inpatient psychiatric services.
ICF/MR expenditures are low in New Hampshire, Vermont and Wyoming and there is no service use among the MH/SA population.

Table 6
The results appeared to be reasonable across states.

Table 7
The results appeared to be reasonable across states.

Table 8
Average length of stay (ALOS) for MH/SA acute inpatient stays looked reasonable in all states except Delaware and, to a lesser extent, Vermont. While most states reported ALOS between 1.0 to 22.5, Delaware's ranged from 1.0 to 77.2 with half of the age groups reporting ALOS of over 25 days. Vermont had one high cell of 34.1 (males, 15-20) which particularly affected the total category for that age group as well (21.9 for 15-20, all sexes). There were also some high averages in the Under 1 category for Alabama and New Jersey, due to small sample sizes, which also affected the "All States" numbers for this age group.

Table 9
Vermont and Wyoming showed unusually high percentages of visits being MH/SA when compared to the total visits for the equivalent Medicaid population. For Vermont, it affected most age groups, but for Wyoming, this was mostly observable in the 6-20 age groups.

Table 10
The results appeared to be reasonable across states.

Back to Contents

Tables (1995)