III. SCREENING AND ASSESSMENT OF ALCOHOL AND OTHER DRUG USE

A multistage screening and assessment process is the most cost effective strategy for the comprehensive evaluation of adolescent substance use and associated problems (Babor, 1991; Dembo, 1994a, 1994c; Tarter, 1991). Since assessment of the multifaceted problems of adolescents can be labor intensive and costly, a decision tree format is recommended to efficiently guide a client through the assessment process (Tarter, 1990). There appears to be consensus that a three step model is the most productive when treating juvenile delinquents and or substance abusing adolescents (Dembo, 1994a, 1994c; Babor, 1991; Tarter, 1991). The first stage of the evaluation process when dealing with potential substance abusing adolescents is to screen for the presence of specific problems. The second stage is to perform a more comprehensive evaluation of problem areas identified by the screening evaluation. In the third phase of the evaluation process individuals are referred to outside agencies for evaluations that can not be performed in the detention or treatment setting and a treatment plan is formulated based on the findings of all of the assessments.

The Treatment Improvement Protocol (TIP; McPhail, 1995) consensus panel on adolescent diversion programs recommends that in the evaluation process, the assessor be an appropriate professional trained in and experienced in working with adolescents. The Treatment Improvement Protocol consensus panel also recommends that the following be included in the evaluation process:

history of alcohol and other drug use medical health history and physical examination developmental issues mental health history strengths or resiliency factors family history school history

vocational history sexual history peer relationships juvenile justice system involvement and delinquency social service agency program involvement leisure activities

A. Screening

Screening should be instituted at the earliest point of contact with the adolescent and should address potential problems in multiple areas. The primary goal of screening is to identify youth with a suspected substance use problem and refer them for a more comprehensive evaluation of substance use and related problems. In published studies the most commonly used screening instruments are the Personal Experience Screening Questionnaire (PESQ; Winters, 1992), the Substance Abuse Subtle Screening Inventory-Adolescent version (SASSI-A; Miller, 1990), and the Problem Oriented Screening Instrument for Teenagers (POSIT; Rahdert, 1991). Instruments such as the SASSI-A screen only for problems with substance use. Other instruments such as the he POSIT, screen for problems in multiple areas of functioning . For treatment providers a lack of consistency in screening can mean that different regions of the state are referring youth with very different needs to the same kind of treatment. From a research perspective, a single instrument is virtually essential in order to provide meaningful comparisons of youth from different areas of the state.

In the state of Washington no single screening instrument is consistently used when determining if an adolescent may require chemical dependency treatment. Through a coordinated court administrative effort in Washington State a new risk assessment tool is currently under development. This new risk assessment instrument will be administered by juvenile court intake and probation staff after the adolescent is adjudicated, but before disposition (sentencing). An adolescent's level of risk will be determined based on interviews with the adolescent and his family and will influence the level of supervision and the conditions of sentence, including the type of treatment that an adolescent is given. Verification of this information will be obtained through reports by other involved agencies. This risk assessment tool appears to cover the areas within the CDDA statute, and those recommended by the Treatment Improvement Protocol consensus panel. The risk assessment tool could effectively be used as a screening instrument to refer potentially eligible youth to the CDDA program for more thorough evaluation of chemical dependency and related problems.

Top

B. Comprehensive Assessment Battery

If an adolescent is referred to CDDA, a primary goal of the CDDA comprehensive assessment battery should be to determine whether an individual is dependent on, abusing, or merely using drugs and alcohol. Additionally, the CDDA evaluation should provide a more detailed description of the adolescent's problems in school, family, peer, medical, legal, and psychiatric domains of functioning, as well as use of leisure time. Results from this comprehensive evaluation can be used to formulate the most appropriate and least restrictive individualized treatment plan. This comprehensive evaluation can also determine if additional referrals and/or treatments are required for problems in areas not addressed by the substance abuse program (e.g., an adolescent who appears at risk for suicide may be referred for a full psychiatric evaluation).

The utility of the assessment instrument as a component of measuring outcomes is also important. For instance, the quantity and the frequency of alcohol use needs to be measured at treatment intake and re-evaluated at each follow-up to determine whether there has been a decrease in alcohol use. All of the areas listed in the effectiveness standards will need to be assessed when an adolescent enters treatment and re-evaluated at each follow-up in order to measure change over time. As mentioned previously, many issues in addition to drug and alcohol use have a tremendous impact in the youth's life and will need to be addressed in treatment; therefore, these issues will have to be identified during the assessment process.

1. Substance use disorder diagnoses

Although there is agreement that use of a sequential assessment battery is the most appropriate and least costly alternative for formulation of a comprehensive treatment plan, there is no consensus as to which instruments should be used to determine substance use disorder diagnoses. For research purposes it is essential that a uniform instrument be used to establish formal substance use diagnoses. A formal diagnosis of substance dependence requires a maladaptive pattern of substance use that results in clinically significant impairments in functioning and or emotional distress regarding substance use. It should be noted, that although individuals who meet criteria for a substance dependence diagnosis typically have greater severity of substance use than those without a dependence diagnosis, individuals with 'severe' substance use may not always meet criteria for a dependence diagnosis. The substance use of an individual may be frequent and intense, but may not result in impairments in their functioning or psychological distress. Therefore, severity of use should not necessarily be considered synonymous with a formal substance use diagnosis of dependence.

Formal substance use diagnoses can be determined by self-report, as in the case of the Client Substance Index (CSI; Moore, 1990), by unstructured clinical interviews, computer based structured interviews, or semi-structured interviews (Anfold, 1989). The Adolescent Diagnostic Interview (ADI; Winters, 1993) is a commonly used structured pencil and paper diagnostic interview. There are also several commonly used computer based structured interviews such as the Diagnostic Interview for Children and Adolescents (DICA; Wellner, 1987), and the Diagnostic Interview Schedule for Children (DISC; Costello, 1985). Semi- structured interviews such as the Composite International Diagnostic Interview-Substance Abuse Module (CIDI-SAM; World Health Organization, 1993), the Structured Interview for DSM-IV Axis I Disorders (SCID-IV, First, 1996), and a youth's version of the Schedule for Affective Disorders and Schizophrenia, often referred to as the 'kiddie SADS' (K-SADS; Puig-Antich, 1987), have also been used to determine formal substance use diagnoses.

There are advantages and disadvantages to each method of assessment. The self-report method is the least costly in terms of staff time, requires no clinical skill on the part of the test administrator, and in fact, does not necessarily require that a test administrator be present. The disadvantages of a self-report method are that the adolescent may not be truthful in his responses in order to manipulate referral decisions. Additionally, the adolescent may not understand the nature of the question due to inattention or reading problems and may not always ask for clarification (Meyers, 1995; Winters, 1992). It is therefore, not recommended that a self-report measure such as the CSI be utilized to make formal substance use diagnoses.

Use of structured interviews can be more costly in terms of time and manpower, since an interviewer must be utilized. A strong degree of clinical skill is not required for interviewers, however, since probing of the youth's responses is generally not necessary in structured interviews. If probing is done, the probes are typically provided and it is advised that the interviewer ask the questions exactly as they are written. Not allowing for probing, however, can be problematic. As with self-report instruments, youth may not fully understand the nature or intent of the question, but the interviewer is directed to consider only the response as given. An unskilled interviewer, for example, may not be able to determine if the youth is minimizing the severity of their substance use. Highly structured interviews also do not encourage rapport building and can result in higher levels of defensiveness. This can be especially true when using a computer based interview (Rahdert, 1995). Consequently, it is not recommended that a structured interview such as the DICA, DISC, or ADI be used for the CDDA project.

Use of a semi-structured interview is recommended. Although these interviews require the greatest amount of clinical skill to administer and score, they provide perhaps the most comprehensive evaluation of substance use problems. These interviews are based on Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria sets and usually include standard initial questions for each criterion, but encourage the use of probing to ensure that the information is correct and that the intent of the question was understood. The interviewer can confront the youth if it appears that they are being untruthful. This type of interview allows for greater rapport building between the interviewer and youth which may result in more valid information (Meyers, 1995).

The K-SADS provides Research Diagnostic Criteria substance use diagnoses which are similar to, but not interchangeable with DSM-IV diagnoses. DSM-IV diagnoses can, however, be established by slightly modifying the interview questions. Advantages of the K-SADS are that it allows for the collection of data from parents and for the incorporation of archival data in determining diagnoses. The CIDI-SAM and SCID provide DSM-IV diagnoses, but have been used primarily with adults and older adolescents. These instruments do not allow use of parental and other sources of information to the same degree as the K-SADS. The K-SADS, SCID, and CIDI-SAM all take approximately 20-30 minutes for a trained technician to administer and provide lifetime and current (last month) diagnoses of substance dependence and abuse. The optimum instrument to use for the CDDA assessment of substance use disorders would, therefore, be the K-SADS.

All of these instruments provide only information concerning the degree of impairment in functioning and presence or absence of withdrawal symptoms required for a formal diagnosis of substance abuse or dependency. These instruments do not provide detailed information regarding the amount or frequency of use or assess functioning in other areas such as school performance, family relationships, or peer relationships. Information regarding intensity and frequency of substance use and functioning in other areas must be obtained from an additional comprehensive assessment instrument.

2. Assessment instruments for substance use and other problem areas

Several comprehensive assessment instruments have the potential to meet the needs of the CDDA project. Common components of these comprehensive assessment instruments include a thorough evaluation of current use and history of substance use and assessment of other problem areas, but do not provide formal diagnoses of substance abuse or dependence. The main differences between these comprehensive instruments are the types of problem areas addressed and the format of the instrument. There are three comprehensive interviews cited most often in research on adolescent with substance use problems and delinquents that should be considered for use. They are the Adolescent Drug Abuse Diagnosis Instrument (ADAD; Friedman, 1989), the Personal Experience Inventory (PEI; Winters, 1993), and the Comprehensive Adolescent Severity Inventory (CASI; Meyers, 1995). Each of these is described below.

a. Personal Experience Inventory (PEI, Winters, 1987 )

The Personal Experience Inventory (PEI) is a 276 item self administered pencil and paper, or computer based, questionnaire which takes approximately 60 minutes to complete (Winters, 1993). The PEI has two major sections, the first is the Chemical Involvement Problem Severity (CIPS) section, which measures substance use severity, frequency, and onset. The PEI, however, does not measure the quantity of substance use. The second section, Psychosocial Risk Factors, measures both interpersonal and environmental risk factors as well as some specific clinical problems such as mental health problems, parental drug use, and eating disorders. The PEI also includes validity scales to control for attempts to fake 'good' or 'bad' responses.

[Note: correlations (r's) of at least 0.75 are considered necessary for acceptable reliability, validity coefficients are typically lower, but should at least be above 0.50]

Norms for the PEI are based on interviews with over 6,000 youth, over half of whom were in chemical dependency treatment. The youth in this normative sample were 60% male, 75% Caucasian, and their average age was 15 years old. The current psychometric evidence for the PEI is encouraging. Test-retest reliability correlations are reported to range from 0.84 to 0.91 for the CIPS and between 0.64 to 0.96 for the psychosocial scales. The PEI has demonstrated adequate content, criterion and construct validity (Guthmann, 1990). The internal consistency (coefficients alpha) of the scales ranges from 0.70 to 0.97 for the chemical involvement scales, between 0.66 to 0.91 for the psychosocial section and between 0.49 to 0.82 for the response distortion scales. These results indicate that the PEI has adequate internal consistency for some, but not all, scales. The PEI has been found to have minimal ethnic bias (Guthmann, 1990).

b. Adolescent Drug Abuse Diagnosis Instrument (ADAD; Friedman, 1989)

The Adolescent Drug Abuse Diagnosis Instrument (ADAD) is a structured interview of 150 items which takes approximately 60 minutes to complete and 10 minutes to score. The ADAD assesses seven problem areas: alcohol and other drug use, health status, education, social interactions, family interactions, psychiatric problems, and legal status. The ADAD also includes items to assess response bias. Each life problem area is scored for problem severity on a 10-point scale. Mathematically derived composite scores can also be determined for each of the seven assessed areas. These composite scores can also be used to assess changes over time in problem severity for each of the seven areas. A shorter 83 item version of the ADAD has been created for use as an outcome tool.

The ADAD was standardized using adolescents (average age 15.6) with substance use problems. The sample included 683 outpatients, 157 nonhospital residential, and 202 hospital based residential patients. The majority of the sample was male (73%) and Caucasian (53%). The interrater reliability of the ADAD is adequate (r = 0.85 to 0.97) as is the test-retest reliability (r's between 0.83 and 0.96) except for the employment section (r = 0.71). Validity for the ADAD is adequate for the majority of the sections (r's between 0.43 and 0.67), but not for the health and social sections (r's below 0.51 and 0.52 respectively in most studies).

c. Comprehensive Adolescent Severity Inventory (CASI; Meyers, 1995)

The Comprehensive Adolescent Severity Inventory (CASI) is the newest of the instruments considered in this report. It is a semi-structured interview that examines functioning in ten life areas: health, stressful life events, education, drug and alcohol use, family relationships, peer relationships, legal issues, mental health, and use of free time (Meyers, 1995). Items that assess response bias are also included. The CASI takes a trained technician approximately 45-90 minutes to administer. Administration time depends on the severity of problems and whether all modules are included. There is a three tiered approach to scoring the CASI. The first level of CASI scoring provides indicators of the assets and liabilities of the adolescent. In the second level of scoring, problem status along a temporal continuum within each life area are provided. The third level of CASI scoring provides overall scores for each of the ten areas of functioning for the last year and the last month. CASI users can use all or any combination of these scoring procedures. The CASI also has questions which lend themselves to potential outcome measures. If the computer version of the CASI is used a summary report of the adolescent's functioning may be obtained which can be utilized in treatment planing.

There is no published data on the psychometric properties of the current version of the CASI. Reliability and validity data for the CASI is currently being collected in a three year National Institute of Drug Abuse funded evaluation. The CASI is a revised version of the Comprehensive Addiction Severity Index for Adolescents (CASI-A; Meyers, 1995). Revisions to the CASI-A were made to improve the overall reliability and validity of the instrument as well as to improve wording of some items (Meyers, 1995). The original standardization sample was 103 adolescents receiving treatment for substance use problems and/or psychiatric problems. The sample had an average age of 16, was mostly Caucasian (89%), and mostly male (55%). Internal consistency of the original CASI-A scales ranged from 0.48 for Family History to 0.80 for Drug and Alcohol Use Consequences. Preliminary evidence of reliability was promising. Once again it should be noted that evaluation of psychometric properties for the CASI is ongoing.

Summary

All of these potential instruments have been used in research with adolescents and have had their reliability and validity assessed to varying degrees. Each of these instruments have ways of accounting for the possibility that the data obtained by these instruments can be biased by how truthfully the youth responds. This is one of the most important factors to consider because youth will be aware that the results of the assessments will help determine whether they are referred to a CDDA program or are sent to detention. No research exists that definitively states that one instrument is superior to the others; research indicates that the aforementioned instruments all appear adequate and would serve the necessary purposes for treatment planning and treatment outcome evaluation.

As discussed previously in the section on diagnostic instruments it is preferable to use a semi-structured interview for data collection compared to a self-report or structured interview whenever possible. Of all these instruments the CASI provides the most detailed information and would allow evaluation of all the elements described in the effectiveness standards with regard to changes in adolescent behavior, except the recidivism measures. Information regarding recidivism, such as subsequent convictions, violations of terms of community supervision, and completion of restitution to victims will be obtained primarily from criminal histories obtained through the JUVIS and OBTS databases. Despite the lack of psychometric information available for the CASI it is recommended for use with CDDA youth because of its comprehensive nature and its semi-structured format.

Other areas of assessment

If more specific information on CDDA youth is required it will be necessary to include additional instruments and/or interviews. The project may want to assess motivation for treatment, in which case the Decisional Balance Inventory (Migneault, 1997) could be used. More information on mental health problems can be assessed with inventories such as the Brief Symptom Index (BSI; Derogatis, 1982). Specific instruments are available to assess problems in areas including physical health, school problems, social/peer relationships, family relationships/problems, delinquency/illegal behavior, and psychological/psychiatric problems. As with all areas of adolescent assessment, however, the reliability and validity of these instruments, for the most part, varies greatly.

Treatment fidelity can be assessed with the Treatment Services Review (TSR; McLellan, 1992). This instrument, which takes 5-10 minutes to administer, queries individuals in treatment about the number of significant contacts that they have had with counselors, physicians, nurses, and other treatment staff. The TSR can also be used to review services documented in clinical charts. It is recommended that, at least initially in the CDDA project, TSRs be administered to a random sample of adolescents during treatment. Clinical charts can then be reviewed to ensure that services being documented in clinical charts are actually being delivered to adolescents.

PREVIOUS

NEXT