AN
EVALUATION
OF
STATE
EPSDT
SCREENING TOOLS

The Early and Periodic Screening, Diagnosis and Treatment (EPSDT) mandate in Medicaid requires states to conduct regularly scheduled examinations (screens) of all Medicaid- eligible recipients under the age of 22 to identify physical and mental health problems. If a problem is detected and diagnosed, treatment must include any federally authorized Medicaid service, whether or not the service is covered under the state plan. If problems are suspected, an "interperiodic" screen is also required so the child need not wait for the next regularly scheduled checkup.¹

To implement EPSDT, many states have chosen to develop or use a specific screen to identify mental health problems. Although constructing appropriate and useful mental health screening tools for pediatricians and other practitioners may be difficult, it is important in order to assess Medicaid-eligible children appropriately for mental health as well as physical health problems.² The Bazelon Center recently studied tools in use or under development to identify children's mental health and addiction treatment needs in 15 states. Our findings are summarized in the table at the end of this issue paper.

To assist advocates in promoting appropriate identification of children in need of mental health and addiction treatment services, this report analyzes the strengths and weaknesses of these tools. Their utility is assessed according to the following criteria: rapid administration, acceptance by parents, immediate availability of results, inclusion of age-specific questions, and inclusion of questions about child and family background and substance use.³

A number of states reported using the Denver Developmental II measure as a general screening tool for children under age 6. The Denver Developmental assesses a child's development in four areas: gross motor, language, fine motor-adaptive and personal-social development. However, it does not adequately screen young children for emotional and behavioral problems. Therefore, it is not considered a mental health screening tool for the purposes of this report.

Rapid
Administration

Given the time constraints of pediatricians and other individuals performing EPSDT screenings, the screening tool must be one that can be administered quickly and efficiently. Most of the tools met this criterion; they were only one or two pages long.

One exception was Achenbach's Child Behavior Checklist (CBCL), used optionally in Utah. This measure seems exceptionally long for an EPSDT screen and is more appropriate for children already showing indications of mental or emotional problems.

North Carolina uses the Guidelines for Adolescent Preventive Services (GAPS), an 83-item measure, which also may be too long for adolescents to finish as an initial screen.

Acceptance by Parents
(Child Caregivers)

Caregivers must be able to complete a caregiver's report or screening interview. The screen should be easy to understand and use clear, jargon-free language. A screen that lends itself to an interview format may be better than a written parent report, because caregivers may not be able to read or read quickly. Written measures should also be available in languages other than English.

Generally, except for the unduly long CBCL, the tools reviewed seemed to meet the criteria of acceptance by caregivers.

Immediate Availability
of Results

Screeners should be able to utilize the information from the screening tool to help make clinical decisions at the visit. All the screening tools reviewed seem to meet this criteria.

However, the CBCL is weaker on this measure than other tools. It is divided into a number of sub-scales for children's problems. To effectively utilize the Checklist, the clinician must properly score the measure. Thus, results of the measure may not be immediately available to the screener.

Age-Specificity

Another key feature of an adequate screening tool is the age-specificity of the questions. There should be appropriate questions for particular age groups, especially very young children and older adolescents and young adults. The following tools are the strongest in this regard:

• Tools from Minnesota and West Virginia are modified versions of the Oregon screen, which has age-specific screening forms. All three have five age categories, roughly corresponding to the following stages of childhood: 1) infants, 2) toddlers, 3) preschool and young school-age children, 4) school-age children and 5) adolescents. West Virginia's screening tool is optional for children under 6.

• Texas also uses a modified version of the Oregon screen, but has fewer (four) and broader age categories than the others.⁴

Age breakdowns in other tools seem more problematic.

• Iowa's tool is age-specific, but the 0-5 age range combines infants and toddlers into one group, which may be too broad. The eight questions for this age group also seem inappropriate for very young children.

• The screening tool from Florida has even larger age categories, 2-10 and 11-21 years old. This screen also lacks questions for very young children (0-2 years).

• Georgia has an age-specific child self-report, but only for children 6 and older.

• The CBCL, used in Utah, has separate measures for 2-3 year- olds and 4-18 year-olds. It also has a youth self-report for 11-18 year-olds.

Child and
Family Background

The mental health screening tool should include questions about child and family background. Information on family history of mental health and substance abuse problems and about child abuse and neglect is especially important for screeners and may help them to decide whether or not to refer a child for further evaluation.

• Illinois' tool is the most comprehensive on this measure. The child's family situation, including family history of mental illness, is covered. The tool includes a set of questions concerning the child's "life changes" during the past year: victimization/neglect, death of a family member, new school, lost relationship, serious illness/injury, incarceration of a parent, loss of job, economic loss, residence, and witness of a violent crime. In addition, this tool asks about the child's chronic medical conditions and health history as part of mental health screen.

• Delaware's screening tool also asks questions about the child's current and previous problems, such as exposure to abuse and trauma. This screen has a separate section about "problems in child's environment" covering issues such as familial substance abuse and mental illness, and psychosocial stressors.

• West Virginia checks for "family risk factors" as well as child abuse and neglect.

• Florida has questions about any family history of "severe emotional, behavioral, and/or neurobiological disorder or severe mental illness" and about the child's history of neglect and abuse.

• Screens from Tennessee and Wisconsin ask about prior child (as well as prior family) mental health problems and treatment.

Substance Abuse
Screening

The federal EPSDT law was intended to include mental health and substance abuse screening and treatment under the term "mental health." Accordingly, screening children and adolescents for substance use is another important component to an overall screen.

All tools evaluated include questions about substance use/abuse, except the Pediatric Symptoms Checklist (PSC), piloted in Ventura County, California and used optionally in Arizona. In two states, however, mental health and substance abuse screens are separate:

• Illinois has a separate, comprehensive screening tool for substance use that may be used in conjunction with the mental health screen.

• The Texas tool references more comprehensive substance-use screens that clinicians may use for EPSDT screening.

North Carolina's GAPS measure has questions about an adolescent's friends' substance use, as well as the target child's use of non-prescription drugs, steroids and illegal drugs. This information may be helpful for clinicians, given the effects of peer relationships on substance use, and to investigate other types of drugs that adolescents may be abusing.

One screening approach for adolescents is to have them fill out a separate report or be interviewed separately and use this information in conjunction with a caregiver report. This may give the clinician critical information about the adolescent's problems, such as substance use, not available from the caregiver report.

Overall Evaluation

Mental health screening tools will continue to evolve to meet the needs of individual states and the children they serve. This report is designed to enable advocates and policymakers to begin evaluating current mental health screening tools. The table on page 6 highlights the state tools that best meet the preceding Bazelon Center criteria.

The Most Useful
Screening Tools

• West Virginia's measure met the criteria for a satisfactory screen. Spe-cifically, it was the only tool to include age- specific forms and questions about child and family background and other important components.

• The screens from Oregon and Minnesota included all the key factors except adequate questions about child and family background.

• Although lacking specific age breakdowns, the Illinois screen met the preceding criteria and addressed other important issues, such as the child's physical health condition and life changes during the current year.

• Similarly, Delaware's tool contained questions about child's and family's current and past problems, but did not have age breakdowns.

• The Texas tool lacked the same basic feature as those from Oregon and Minnesota, and it also had less satisfactory age categories.

Other state EPSDT screening tools evaluated had individual useful questions and formats, but did not provide the best overall screen.

Notes

1. Koyanagi, C., & Brodie, J.R. (1994). Making Medicaid work to fund intensive community services for children with serious emotional disturbance. Washington, DC:Bazelon Center for Mental Health Law. Return to text.

2. Jellinek, M. S., & Murphy, J. M. (1990). The recognition of pyschosocial disorders in pediatric office practice: The current status of the pediatric symptom checklist. Developmental and Behavioral Pediatrics, 11, 273-278. Return to text.

3. The first three criteria listed were identified by Jellinek and Murphy (1990), along with clinical utility. The clinical utility of a screen should be assessed by a mental health professional and will not be discussed here. Clinical guidance—whether or not the screen helps the screener to know when to refer a child to appropriate mental health or substance abuse services—is another useful criterion that will not be evaluated. Since we did not specifically ask states to include provider manuals or other directive materials with their screens, we cannot assess this area. However, a number of state tools identified the presence of certain behaviors as requiring a referral to mental health or other social services. Return to text.

4. The substance abuse screening tools referred to are the Michigan Alcoholism Screening Test (MAST); the T-ACE, a four-item questionnaire usable in assessing pregnant women in a clinical practice setting for risk drinking; and the CAGE, a four-item measure that asks: "Have you ever felt you should cut down on your drinking?"; "Have people annoyed you by criticizing your drinking?"; "Have you ever felt bad or guilty about your drinking?"; and "Have you ever had a drink first thing in the morning to steady your nerves to get rid of a hangover (eye opener)?" Return to text.