Screening for adolescent substance-related disorders using the SASSI-A2: implications for nonreporting youth.
According to several national surveys (Johnston, O'Malley, Bachman, & Schulenberg, 2008; Spoth, Randall, Trudeau, Shin, & Redmond, 2008), adolescent substance use continues to be documented at a rate that causes concern. In 2008, for example, 20% of youth reported using an illegal drug before finishing the eighth grade (Johnston et al., 2008). In 2006, 48.2% of high school seniors reported lifetime use of any kind of illicit drug, while 25.1% reported drinking alcohol on three or more occasions within the past 30 days (Spoth et al., 2008). Given the potential risk behaviors (e.g., aggressiveness, delinquency), developmental sequence of psychopathology (e.g., major depression), and adverse life outcomes (e.g., school dropout, criminal incarceration, acute medical problems) that drug and alcohol use presents, it is imperative that early interventions target youth at risk for such behavior before they manifest these problems in adulthood (Fergusson & Horwood, 1997; Kandel & Yamaguchi, 2002; Leslie, 2008; Merline, Jager, & Schulenberg, 2008; Vakalahi, 2001; Zucker, 2008).
To reduce the multiple risks associated with adolescent drug and alcohol use, it is necessary to accurately identify such use in beneficial ways (Winters, Stinchfield, Henly, & Schwartz, 1991). Historically, the mere use of alcohol or drugs by adolescents was classified as "substance abuse" based on societal and legal norms (Harrison, Fulkerson, & Beebe, 1998). According to Harrison et al. (1998), this approach proved to be too simplistic, as deviating from societal and legal norms in and of itself does not constitute a mental disorder. As an alternative form of assessment, Harrison et al. recommended new diagnostic criteria that include diagnostic threshold categories based on the total number of symptoms presented. The current diagnostic criteria for substance-related disorders (Diagnostic and Statistical Manual of Mental Disorders [4th ed., text rev.]; DSM-IV-TR; American Psychiatric Association, 2000) require the detection of specific symptoms and the presence of a required number of symptoms for diagnosis. Specifically, the DSM-IV-TR criteria require significant impairment in one of four possible symptoms for substance abuse, and three of seven possible symptoms for substance dependence disorder, within the same 12-month period. In their critical review of adolescent substance-abuse instruments, Leccese and Waldron (1994) observed that most measures are based on what youth self-report. Self-report continues to be the primary source of information for screening for substance use in adolescents (Schwartz & Smith, 2003). Self-report of substance use poses a potential issue of underdiagnosis, especially in nonclinical samples (e.g., samples obtained from school settings that are not in treatment or being evaluated for treatment) in which participants minimize or deny alcohol and/or drug use even if at risk for substance abuse. Such denial is possibly due to the endorsement of social desirability (Winters et al., 1991).
Out of all the available instruments currently used in practice, a strength of the Adolescent Substance Abuse Subtle Screening Inventory (SASSI-A2; Miller & Lazowski, 2001) is that it purports to measure adolescent substance abuse regardless of the respondents' willingness to be open and honest about that use (Feldstein & Miller, 2007; Rogers, Cashel, Johansen, Sewell, & Gonzales, 1997). Consequently, the SASSI-A2 may be useful for certain populations at high levels of risk, but that do not tend to report drug and alcohol use. Research literature on SASSI-A2 is limited.
A critical review of the extant research on the previous version of the instrument (SASSI-A; Miller & Lazowski, 2001) suggests that its ability to detect substance-related disorders in nonreporting adolescents is questionable. Rogers et al. (1997), for example, noted that the SASSI-A, although useful for accurately identifying 75% of reporting users, misclassified approximately 66% of nonusers. Rogers et al. further reported that the SASSI-A demonstrated less efficiency in classifying Hispanic participants than in classifying Caucasian participants, thus possibly indicating an ethnic or cultural bias. Bauman, Merta, and Steiner (1999) concluded that although the SASSI-A is able to differentiate between clinical and nonclinical at-risk groups, the validity of the SASSI-A for adolescent substance abuse screening was questionable, because it "classified significantly more individuals as chemically dependent than did the clinicians" (p. 68). Sweet and Saules (2003) noted that while the SASSI-A face validity scales demonstrated validity, the other scales of the SASSI-A revealed poor construct validity. Additionally, Stein et al. (2005) noted that SASSI-A's construct validity was supported for use with juveniles in a correctional setting; however, their results indicated that the SASSI-A was best for detecting alcohol consumption, when participants scored three or above in the Face Valid Alcohol (FVA) Scale, but not useful for detecting drug use. After studying the available research on the SASSI, Feldstein and Miller (2007) concluded that SASSI's subtle scales did not add incremental validity over and above that which was gained from the face valid and obvious scales.
Virtually all contemporary theoretical models in the literature recognize that the risk for adolescent substance abuse is multifactorial, existing at the genetic, individual, familial, peer, school, and community levels of influence (Segal & Stewart, 1996; Vakalahi, 2001). When considering the total body of research concerning its pathogenesis, parental substance use, family dysfunction, exposure to abuse and/or violence during childhood, being male, novelty seeking, family history of alcoholism, and affiliation with substance-using peers increases the risk for youth (see Fergusson, Boden, & Horwood, 2008). Of this complex interaction, family influence has been shown to be the strongest risk or protective factor (Amey & Albrecht, 1998; Goebert et al., 2000). Age of onset also seems to be a strong predictor, that is, early onset increases the likelihood of using drugs (Meyer & Neale, 1992). On the other hand, a number of protective factors have also been investigated, including strong antialcohol/drug attitudes, involvement in extracurricular activities, high self-esteem, social support, and religious commitment (Anderson, Ramo, Schulte, Cummins, & Brown, 2007; Liepman, Calles, Kizilbash, Nazeer, & Sheikh, 2002; Maton & Zimmerman, 1992).
Given the confluence of multiple risk and protective factors at multiple units of analysis, the SASSI-A2 may be a valid instrument for clinical screening, assessment, and treatment planning, bacause its various subscales measure such factors and take them into account when predicting levels of risk for substance abuse. In this context, the main purpose of the current study was to examine if the SASSI-A2's diagnostic rule criteria were valid among a diverse sample of urban youth that was represented predominantly by African Americans. To date, there is a paucity of research on the SASSIA2 as it pertains to African American youth, as well as other youth (e.g., biracial/multiracial youth), with the minimal research available indicating that African American adolescents tend to present a substantially lower prevalence rate of drug and alcohol use when compared with White or Hispanic adolescents (Amey & Albrecht, 1998; Bolland, Bryant, Lian, McCallum, Vazsonyi, & Barth, 2007). The main empirical question explored in this study was how these natural advantages that the SASSI-A2 provides to clinicians may be applied to diverse urban youth who may not report, in general, any levels of drug or alcohol use. Unlike the clinically based populations with which the SASSI-A2 was developed, cross-validated, and eventually normed (i.e., youth treated in addictions centers, inpatient hospitals, outpatient health facilities, and juvenile correctional programs), our sample was obtained from a public urban high school, that is, within a nonclinical context.
One hundred thirty-seven high school-age students completed the SASSIA2. Seven participants did not complete the information for the face-valid alcohol and/or drug-related questions. The sample consisted of 38.7% (n = 53) male adolescents and 61.3% (n = 84) female adolescents. Regarding participants' racial/ethnic background, 52.6% (n = 72) were African American, 18.2% (n = 25) biethnic, 9.5% (n = 13) multiethnic, 6.6% (n = 9) Caucasian, 0.7% (n = 1) Hispanic, 0.7% (n = 1) other, and 11.7% (n = 16) did not report race/ethnicity. The participants' mean age was 15.72 years (SD = 1.27, range = 14-19 years).
The research proposal for this study was approved by the Human Subjects Review Board of the university and the research office of the high school district superintendent. The permission of school administrators and relevant teachers was also obtained. Students who returned a signed informed consent form from their parent(s) or legal guardian participated in the study. The students also read and signed an assent form. Data were collected from classes in language arts, English, and history. The participants completed a short demographic sheet and series of questionnaires during a single class period that were part of a broader project designed to address other lines of research in the areas of career development and positive youth development. They were given a small snack for participating.
The SASSI-A (Miller & Lazowski, 2001) was originally published in 1990 and was designed to screen adolescents between the ages of 12 and 18 years for the presence of a substance use disorder. The SASSI-A was revised and published as SASSI-A2 in 2001 with the same clinical uses (Miller, Renn, & Lazowski, 2001); the instrument is administered, scored, and interpreted in approximately 20 minutes (Bauman et al., 1999), similar to its previous version. The SASSI-A2 is composed of two pages. The first page contains face-valid questions regarding the Frequency of Alcohol (FVA scale) and other drug use (Face Valid Other Drugs [FVOD] scale). On the second page are 72 true-false questions that measure substance use symptoms, risk factors related to substance misuse, attitudes and beliefs associated with alcohol and drug use, and subtle items that seem irrelevant but have been associated with alcohol- and drug problems (Miller et al., 2001). The items on the true-false section are organized into 10 scales. Only seven of these scales are used in decision rules that determine if the adolescent has a high probability of a substance use disorder. Another scale determines risk for illegal behavior. Two additional scales are used to screen those in the low-probability category who may need further evaluation for a possible substance use disorder.
The scales of the SASSI-A2 are provided here with a brief explanation. The FVA scale includes 12 alcohol use frequency-related (i.e., never, once or twice, several times, repeatedly) questions. The FVOD scale includes 16 "other drug" use frequency-related questions. The six-item Family-Friends Risk (FRISK) scale was designed to measure the extent to which the adolescent is part of a social system that may enable substance misuse. The Attitudes (ATT) scale (10 items) is intended to measure attitudes and beliefs related to alcohol and drug use. The Symptoms (SYM) scale (nine items) is intended to measure consequences endured as a result of substance misuse and loss of control. The following five scales are considered subtle scales with items seemingly unrelated to alcohol and drug use. The Obvious Attributes (OAT) scale (11 items) is intended to measure the ability to acknowledge problems in a person's life. The Subtle Attributes (SAT) scale (12 items) is intended to measure detachment from feelings and lack of insight into problems. This scale is also intended to identify possible substance use disorders even if an adolescent is not reporting his or her alcohol and/or drug use. The Defensiveness (DEF) scale's 12 items are intended to measure ability to acknowledge personal limitations. A high score indicates faking good (i.e., an answer that conceals the evidence of a substance use disorder), whereas a low score may indicate a tendency to exhibit low self-esteem. The seven items of the Supplemental Addiction Measure (SAM) scale are intended to distinguish adolescents with and without substance abuse disorders. The Correctional (COR) scale is a 16-item scale, not used in the decision rules for substance use disorder, available to determine an adolescent's risk for illegal behavior. The last two scales are the Validity Check (VAL) scale, which has 11 items, and Secondary Classification Scale (SCS), which has 31 items. The SCS scale is intended to differentiate, during the screening process, between the likelihood of substance abuse versus substance dependence disorder (Miller et al., 2001).
There have been several changes and updates to the SASSI-A2 regarding its scales, norms, and psychometric properties. First, the existing scales were revised to coordinate with DSM-IV-TR diagnoses. Three scales (FRISK, ATT, SYM) and SCS scales were added. In addition, the RAP scale was replaced with the VAL scale (Miller & Lasowski, 2001). Second, the SASSI-A2 has six other questions intended to facilitate discussion with adolescents. These questions inquire about the frequency of current alcohol and drug use, age of onset of first alcohol and drug use, age of onset of regular alcohol and drug use, whether school grades are affected by alcohol and drug use, student status and current grade level, and whether the adolescent has been in trouble with the law. Third, the SASSI-A2 was normed on adolescents from both genders who represented diverse ethnic backgrounds (Miller & Lazowski, 2001), a significant portion of whom were Caucasian (i.e., 64%) and Hispanic (i.e., 20%). The Adolescent SASSI-A2 User's Guide (Miller et al., 2001) indicates that it has 95% sensitivity, 89% specificity, 98% positive predictive power, and 75% negative predictive power, with 11% false positive and 5% false negative rates. Fourth and finally, Miller and Lazowski (2001) also reported 2-week test-retest reliability coefficients between .81 and .92 and clinician diagnoses as the criteria used to validate the SASSI-A2.
Given the racial and ethnic composition of participants in the present study, an important limitation of the SASSI-A2's external validity is noteworthy. Only five African Americans were recruited in establishing the norms (N = 856); this represents 1% of the entire normative sample. Also, there was a lack of representation for biracial and/or multiracial groups among the SASSI-A2's norms, which constitute a rapidly growing population in the United States. It is of interest that 118 participants (13.8%) of the normed sample were classified as "other/unknown."
The means and standard deviations for the scales among the total sample and the African American subgroup, in comparison with the total sample as reported in the SASSI-A2 manual's norms, are provided in Table 1. Consistent with past research, the results indicated that both the total sample and the African American subgroup were two to three times less likely to report using alcohol or drugs than the normative sample. At the same time, the African American subgroup endorsed higher mean scores on the FRISK, ATT, SAT, DEF, and COR scales than the normative sample did for the SASSI-A2, regardless of gender. This pattern suggests that although African American urban youth are less likely to report using drugs and alcohol, they may be at similar (or even higher) levels of likelihood as the normative sample for being associated with a social support system of enablers, defensive about the consequences of substance use, detached from their feelings and lacking insight, avoiding the acknowledgement of personal limitations, and being associated with teenagers in a juvenile correctional environment. All these factors can significantly place adolescents at high levels of risk for substance abuse, even if they are not using regularly.
In terms of reliability analyses (see Table 2), the results indicated a contrast between the normative sample and this study's sample, including the African American subgroup. The internal consistencies for the FVA, FVOD, and SCS for the total sample, as well as African American youth, ranged from moderately acceptable to extremely high (i.e., .62 to .93). For the decision rules only procedure, which is based on responses across nine different scales, moderately acceptable consistency at .63 and .62 for the total sample and the African American sample, respectively, were present. However, the remaining scales did not show acceptable levels of internal consistency; in fact, many were extremely low, ranging from .07 to .59. These figures are in sharp contrast with the much higher reliabilities obtained from the normative sample, although no reliability estimates were reported in the manual for VAL and SCS. Given that most of the scales in the current study were not reliable, correlational analyses between multiple scales were not performed. Nonetheless, the decision rules criteria and other important scales regarding self-reported drug and alcohol use (e.g., FVA, FVOD) did show acceptable reliability, and thus could serve as a basis for investigating our central research question about assessing substance abuse risk among a nonreporting population that was in a nonclinical setting.
The probabilities of a substance-related disorder with each decision rule as a function of gender and ethnicity are summarized in Table 3. The low probability data sets were checked for validity using the "5 or more" indicator on the VAL scale and "16 or more" on the SCS scale. Data sets that met the validity check criteria are categorized and presented as questionable probability.
Using all nine decision rules, 39.41% (20 male adolescents, 34 female adolescents) screened for a high probability of a substance-related disorder, although 70.8% reported no current alcohol or drug use and 42.3% reported never taking a drink or drug. Another 21.89% (12 male adolescents, 18 female adolescents) qualified for further assessment for a probable substance-related disorder using the SASSI-A2 validity check.
A good study of the SASSI-A2 should provide criterion validity for each scale so that the information can be compared with other research on SASSI-A2 (Feldstein & Miller, 2007). In our study, a closer look at the different criteria that screen in our sample for a possible substance abuse disorder may provide a better understanding of how the SASSI-A2 screens ethnic minorities for such a disorder. We provide the screening results, rule by rule, to facilitate an understanding of the percentage at which rules were screened in our sample. Using only Decision Rule 1 (i.e., FVA and FVOD scales), 13.87% of our sample qualified for a high probability of a substance-related disorder. This percentage increased to 35.04% (an additional 10 male adolescents and 19 female adolescents) with the inclusion of the FRISK scale. Among the racial and ethnic groups, 29.17% African American, 55.56% Caucasian, 44% biethnic, and 53.85% multiethnic youth were screened for a high probability of a substance-related disorder. Moreover, Decision Rule 1 indicated that 9.72% African American, 22.22% Caucasian, 8% biethnic, and 23.08% multiethnic adolescents were at a high probability of experiencing a substance-related disorder; in turn, Decision Rule 2 screened-in an additional 15.28% African American, 22.22% Caucasian, 32% biethnic, and 23.08% multiethnic youth as having a high probability for the disorder. Finally, the validity check classified an additional 27.78% African American, 33.33% Caucasian, 8% biethnic, and 23.08% multiethnic youth as needing further assessment for a high probability of experiencing a substance-related disorder.
Responses to the discussion questions related to frequency of current use, age of first use, onset of regular use, effect on grades, and encounters with the legal system were analyzed. The choice answers were coded as 1 = none, 2 = once a month, 3 = 1 to 3 times a month, 4 = once a week, 5 = twice a week, and 6 = more than twice a week. The mean of current use was 1.53 (SD = 1.08), indicating that, on average, youth were using between once a month to not using at all. The answers for first use were coded as 1 = never tried, 2 = under age 12, 3 = age 12, 4 = age 13, and so on up to age 18. The mean age of first use was 3.26 (SD = 2.40), indicating that, on average, the youth had first used when they were between 12 and 13 years old. Onset of regular use was coded in a similar way to age at first use as described previously. The mean age of onset was 1.64 (SD = 1.64). Six (2.2%) participants reported that their grades had suffered due to alcohol and/or drug use; 48 (35%) reported no such effect on their grades. In addition, 29 participants (21.2%) reported that they had been in trouble with the law and 105 (76.6%) reported no such encounter. Overall, these results confirm the notion that this particular population of adolescents is less likely to report current use and less likely to report any kind of consequences pertaining to use. The results also lend further support to the results obtained from the FVA and FVOD scales.
The SASSI-A2 screened slightly higher than one third of our sample as qualifying for a substance-related disorder, and slightly lower than one quarter of other participants as requiring further assessment. However, most of these students were screened in by decision rules that included scales other than the face valid scales.
This discrepancy between screening for a possible substance-related disorder and the nonreporting of such use supports the notion that African American youth are more likely to nonreport than other groups (Amey & Albrecht, 1998; Bolland et al., 2007). The inconsistencies between the reporting on the FVA and FVOD scales and the discussion questions on onset and frequency of regular use are congruent with the aforementioned hypothesis.
It is of interest that our findings also indicated gender differences in the probability of a substance-related disorder. That is, both Decision Rule 1 (i.e., FVA or FVOD [greater than or equal to] 12) and Decision Rule 2 (i.e., FRISK [greater than or equal to] 5) classified more female adolescents as having a high probability of substance-related disorder than did male adolescents. Among the African Americans, male adolescents had substantially higher levels of drug use than did female adolescents (albeit, still relatively low in terms of averages), whereas the male adolescents had significantly lower levels of alcohol use. In contrast to traditional scholarship, the findings are inconsistent with the assumption that male adolescents are at higher risk than female adolescents (Fergusson et al., 2008). The potential explanations for this discrepancy, at least among African American youth, are difficult to ascertain at this time. However, some researchers (Van Etten & Anthony, 2001; Van Etten, Neumarks, & Anthony, 1999) have suggested that the gender difference in drug and/or alcohol use is due to its higher accessibility to male adolescents; when chances were similar for female adolescents, there appeared to be no gender difference in consumption. In addition, Nishimura, Hishinuma, Else, Goebert, and Andrade (2005) found higher reporting of drug use by female adolescents than by male adolescents in their Hawaiian sample. The combination of the previous findings suggests that the gender differences in alcohol and/or drugs may be changing, depending on availability. Future research should seek to further verify or explore such gender differences.
With respect to the reliability estimates found in the study, it would be premature at this point to conclude that such findings are representative for all urban youth in nonclinical settings, including African Americans. Yet, the findings are suggestive because they indicate that this particular sample, when responding to the dichotomous items (true-false) on the second page of the instrument, did not respond in a consistent manner for the majority of the scales. Why, then, would there be relatively higher reliability estimates for the ATT and SCS scales?
While speculative, it is possible that the ATT scale was comparably more reliable, in part because of the repetitive content of its items; more specifically, all 10 items on the ATT have the phrase "alcohol and/or drug," which is not reflective for any other scale. This general familiar terminology may help decrease sources of error and increase the perceived face validity of the scale. Youth may not become as confused or random in their responses when compared with answering the items of other scales. In the case of the SCS scale, it had a relatively more impressive internal consistency because of the sheer number of items that it contains (31 items); moreover, the SCS scale contains three items that also appear in the ATT. Indeed, none of the other scales used any of the items on the ATT.
It is interesting to note that among the African Americans, the reliability of the SCS scale was substantially higher than that for the total sample (.73 versus .62, respectively); this may coincide with the fact that the African American adolescents had higher means on the COR when compared with the normative mean as well as the total sample. It could be that even among nonclinical samples, especially African American youth in urban areas, peer-based influences are just as influential (if not more influential) than clinical samples included among the instrument norms because many of these youth may still be likely to associate with juvenile offenders. This speculation seems to be supported by the fact that among the normative sample for the SASSI-A2, 17% reported being in trouble with the law. By contrast, 21% of the youth in our sample reported legal trouble.
In the current study, we specified a 6-month time frame when responding to the SASSI-A2 items; that is, we focused the participants on the past 6 months of their life. According to Miller et al. (2001), using the 6-month time frame increases the rate of false negatives; thus, perhaps even more of the youth in this sample were undetected for substance abuse risk by the decision rules. Another limitation pertained to research design is that we did not administer another substance abuse screening instrument or method along with the SASSI-A2. Thus, it was not possible to corroborate the interpretations from the SASSI-A2. A limitation related to our findings is that existence of group differences does not mean that the instrument's results are biased for or against one of the groups. Future research will need to integrate the SASSI-A2 results with other formal and informal sources of diagnostic criteria and data to shed light on this study's implications, which remain tentative. Results from future research will also serve to bolster or weaken the overall construct validity of the instrument.
Despite these limitations, the current study helps to advance empirically the psychometric and clinical utility of the SASSI-A2 among one of its most underrepresented populations while providing initial data for future research concerning populations that do not have norms (i.e., biracial and multiracial or ethnic youth). What is unique about the current study is that, in general, such a population tends to be a nonreporting when compared with the group used to establish norms and to validate the instrument. Understanding the biracial and multiracial youth's response as well as other ethnic youth's response to the SASSI-A2 is important as the population of the United States tends to move toward a more ethnically mixed composition (Miville, Constantine, Baysden, & So-Lloyd, 2005). Future research replicating studies on African American youth, biracial and multiracial youth, as well as other minority youth groups and SASSI-A2's screening for possibility of a substance abuse disorder are needed to establish construct validity and clinical significance of this instrument for ethnic minority youth. Ultimately, the advantages of using the SASSI-A2 to evaluate substance use among adolescents will rest on its applications to all segments of youth in a variety of clinical and nonclinical settings.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Arlington, VA: Author.
Amey, C. H., & Albrecht, S. L. (1998). Race and ethnic differences in adolescent drug use: The impact of family structure and the quantity and quality of parental interaction. Journal of Drug Issues, 28, 283-298.
Anderson, K. G., Ramo, D. E., Schulte, M. T., Cummins, K., & Brown, S. A. (2007). Substance use treatment outcomes for youth: Integrating personal and environmental predictors. Drug and Alcohol Dependence, 88, 42-48.
Bauman, S., Merta, R., & Steiner, R. (1999). Further validation of the adolescent form of the SASSI. Journal of Child and Adolescent Substance Abuse, 9, 51-71.
Bolland, J. M., Bryant, C. M., Lian, B. E., McCallum, D. M., Vazsonyi, A. T., & Barth, J. M. (2007). Development and risk behavior among African American, Caucasian, and mixed-race adolescents living in high poverty inner-city neighborhoods. American Journal of Community Psychology, 40, 230-249.
Feldstein, S. W., & Miller, W. R. (2007). Does subtle screening for substance abuse work? A review of the Substance Abuse Subtle Screening Inventory (SASSI). Addiction, 102, 41-50.
Fergusson, D. M., Boden, J. M., & Horwood, L. J. (2008). The developmental antecedents of illicit drug use: Evidence from a 25-year longitudinal study. Drug and Alcohol Dependence, 96, 165-177.
Fergusson, D. M., & Horwood, L. J. (1997). Early onset cannabis use and psychosocial adjustment in young adults. Addiction, 92, 279-296.
Goebert, D., Nahulu, L., Hishinuma, E., Bell, C., Yuen, N., Carlton, B., ... Johnson. (2000). Cumulative effect of family environment on psychiatric symptomatology among multiethnic adolescents. Journal of Adolescent Health, 27, 34-42.
Harrison, P. A., Fulkerson, J. A., & Beebe, T. J. (1998). DSM-IV substance use disorder criteria for adolescents: A critical examination based on a statewide school survey. American Journal of Psychiatry, 155, 486-492.
Johnston, L. D., O'Malley, P. M., Bachman, J. G., & Schulenberg, J. E. (2008). Monitoring the future national survey results on drug use, 1975-2007. Volume I: Secondary school students (NIH Publication No. 08-6418A). Bethesda, MD: National Institute on Drug Abuse.
Kandel, D. B., & Yamaguchi, K. (2002). Stages of drug involvement in the U.S. population. In D. B. Kandel (Ed.), Stages and pathways of drug involvement: Examining the gateway hypothesis (pp. 65-89). New York, NY: Cambridge University Press.
Leccese, M., & Waldron, H. B. (1994). Assessing adolescent substance use: A critique of current measurement instruments. Journal of Substance Abuse Treatment, 11, 553-563.
Leslie, K. (2008). Youth substance use and abuse: Challenges and strategies for identification and intervention. Canadian Medical Association Journal, 178, 145-148.
Liepman, M. R., Calles, J. L., Kizilbash, L., Nazeer, A., & Sheik, S. (2002). Genetic and nongenetic factors influencing substance use by adolescents. Adolescent Medicine State of the Art Reviews, 13, 375-401.
Maton, K. I., & Zimmerman, M. A. (1992). Psychosocial predictors of substance use among urban Black male adolescents. Drugs and Society, 6, 79-113.
Meyer, J. M., & Neale, M. C. (1992). The relationship between age at first drug use and teenage drug use liability. Behavior Genetics, 22, 197-213.
Merline, A., Jager, J., & Schulenberg, J. E. (2008). Adolescent risk factors for adult alcohol use and abuse: Stability and change of predictive value across early and middle adulthood. Addiction, 103, 84-99.
Miller, F. G., & Lazowski, L. E. (2001). The adolescent SASSI-A2 manual: Identifying substance use disorders. Springville, IN: SASSI Institute.
Miller, F. G., Renn, W. R., & Lazowski, L. E. (2001). Adolescent SASSI-A2 user's guide. Bloomington, IN: Baugh Enterprises.
Miville, M. L., Constantine, M. G., Baysden, M. F., & So-Lloyd, G. (2005). Chameleon changes: An exploration of racial identity themes of multiracial people. Journal of Counseling Psychology, 52, 507-516.
Nishimura, S. T., Hishinuma, E. S., Else, T. R. N., Goebert, D. A., & Andrade, N. N. (2005). Ethnicity and adolescent substance use. Cultural Diversity and Ethnic Minority Psychology, 11, 239-258.
Rogers, R., Cashel, M. L., Johansen, J., Sewell, K. W., & Gonzalez, C. (1997). Evaluation of adolescent offenders with substance abuse: Validation of the SASSI with conduct-disordered youth. Criminal Justice and Behavior, 24, 114-128.
Schwartz, R. C., & Smith, S. D. (2003). Screening and assessing adolescent substance abuse: A primer for counselors. Journal of Addictions & Offender Counseling, 24, 23-34.
Segal, B. M., & Stewart, J. C. (1996). Substance use and abuse in adolescence: An overview. Child Psychiatry and Human Development, 26, 193-210.
Spoth, T., Randall, G., Trudeau, L., Shin, C., & Redmond, C. (2008). Substance use outcomes 5.5 years past baseline for partnership-based, family-school preventative interventions. Drug and Alcohol Dependence, 96, 57-68.
Stein, L. A. R., Lebeau-Craven, R., Martin, R., Colby, S. M., Barnett, N. P., Golembeske, C., & Penn, J. V. (2005). Use of the adolescent SASSI in a juvenile correctional setting. Assessment, 12, 385-394.
Sweet, R. I., & Saules, K. K. (2003). Validity of the Substance Abuse Subtle Screening Inventory-Adolescent version (SASSI-A). Journal of Substance Abuse Treatment, 24, 331-340.
Vakalahi, H. (2001). Adolescent substance use and family-based risk and protective factors: A literature review. Journal of Drug Education, 31, 29-46.
Van Etten, M. L., & Anthony, J. C. (2001). Male-female differences in transitions from first drug opportunity to first use: Searching for subgroup variation by age, race, region, and urban status. Journal of Women's Health & Gender-Based Medicine, 10, 797-804.
Van Etten, M. L., Neumarks, Y. D., & Anthony, J. C. (1999). Male-female differences in the earliest stages of drug involvement. Addiction, 94, 1277-1437.
Winters, K. C., Stinchfield, R. D., Henly, G. A., & Schwartz, R. H. (1991). Validity of adolescent self-report of alcohol and other drug involvement. International Journal of the Addictions, 25, 1379-1395.
Zucker, R. A. (2008). Anticipating problem alcohol use developmentally from childhood into middle adulthood: What have we learned? Addiction, 102, 100-108.
Dilani M. Perera-Diltz and Justin C. Perry, Department of Counseling, Administration, Supervision, and Adult Learning, Cleveland State University. Correspondence concerning this article should be addressed to Dilani M. Perera-Diltz, Department of Counseling, Administration, Supervision, and Adult Learning, Cleveland State University, 2121 Euclid Avenue, JH 282, Cleveland, OH 44115 (e-mail: firstname.lastname@example.org).
TABLE 1 Comparison of Means for the Scales for Total Sample, African American Subgroup, and the Adolescent Substance Abuse Subtle Screening Inventory (SASSI-A2) Norming Sample Total Sample African American Subgroup Men Women Men Women (n = 53) (n = 84) (n = 28) (n = 44) Scale M SD M SD M SD M SD FVA 0.74 1.65 1.10 2.19 0.39 1.30 0.88 1.82 FVOD 2.26 5.98 0.94 2.71 2.46 7.08 0.55 1.88 FRISK 3.28 1.75 3.37 1.96 3.14 1.58 3.07 1.91 ATT 3.77 1.91 3.19 1.94 3.68 1.87 3.14 1.95 SYM 1.51 1.45 1.40 1.68 1.46 1.50 1.14 1.19 OAT 5.32 2.70 5.08 2.48 4.89 2.54 4.55 2.63 SAT 3.64 1.78 3.23 1.92 3.71 1.68 2.93 1.91 DEF 5.43 2.61 5.48 2.34 5.25 2.49 5.68 2.38 SAM 2.11 1.44 2.56 1.79 1.93 1.33 2.23 1.27 COR 8.26 2.41 8.11 2.52 8.61 2.13 8.55 2.28 VAL 3.43 2.28 3.87 1.88 3.61 2.28 4.00 1.79 SCS 9.59 5.04 9.11 5.01 9.11 5.00 8.46 4.69 SASSI-A2 Sample Men Women (n = 336) (n = 449) Scale M SD M SD FVA 4.20 5.70 3.60 5.60 FVOD 5.00 8.70 4.00 7.80 FRISK 2.60 2.00 2.60 2.10 ATT 3.60 2.50 3.10 2.30 SYM 2.30 2.30 2.00 2.10 OAT 5.80 2.70 4.90 2.80 SAT 3.40 2.20 2.80 2.10 DEF 5.00 2.40 4.90 2.60 SAM 2.00 1.60 1.70 1.50 COR 7.00 2.40 6.80 2.70 VAL -- -- -- -- SCS -- -- -- -- Note. SASSI data were taken from Miller, F. G., & Lazowski, L. E. (2001). The adolescent SASSI-A2 manual: Identifying substance use disorders (p. 57). Springville, IN: SASSI Institute. Reprinted with permission. FVA = Frequency of Alcohol Use scale; FVOD = Frequency of Alcohol and Other Drug Use scale; FRISK = Family-Friends Risk scale; ATT = Attitudes scale; SYM = Symptoms scale; OAT = Obvious Attributes scale; SAT = Subtle Attributes scale; DEF = Defensiveness scale; SAM = Supplemental Addiction Measure scale; COR = Correctional scale; VAL = Validity Check scale; SCS = Secondary Classification scale. TABLE 2 Comparison of Reliability Coefficient Alpha SASSI Coefficient Alpha Scale (a) Coefficient Frequency of Alcohol Use .75 .91 Frequency of Alcohol and Other Drug Use .91 .95 Family-Friends Risk .12 .67 Attitudes .58 .76 Symptoms .24 .82 Obvious Attributes .07 .72 Subtle Attributes .18 .63 Defensiveness .12 .64 Supplemental Addiction Measure .21 .66 Correctional .22 .61 Validity Check .18 -- Secondary Classification .62 -- Total .63 .75 Note. The comparative Substance Abuse Subtle Screening Inventory (SASSI) alpha coefficients were taken from Miller, F. G., & Lazowski, L. E. (2001). The adolescent SASSI-A2 manual: Identifying substance use disorders (p. 29). Springville, IN: SASSI Institute. Reprinted with permission. (a) Kuder-Richardson estimates. TABLE 3 Probablility for a Substance Abuse Disorder by Gender and Race/Ethnicity With Each of the Adolescent Substance Abuse Subtle Screening Inventory (SASSI-A2) Decision Rules SASSI-A2 Rule Variable 1 2 3 4 5 6 Gender High probability 19 48 48 48 49 49 Men 7 17 17 17 18 18 Women 12 31 31 31 31 31 Low probability 113 84 84 84 83 83 Men 44 34 34 34 33 33 Women 69 50 50 50 50 50 QLP -- -- -- -- -- -- Men -- -- -- -- -- -- Women -- -- -- -- -- -- Ethnicity High probability 12 48 48 48 49 49 AA 7 18 18 18 19 19 HA 1 1 1 1 1 1 CA 2 4 4 4 4 4 Biethnic 2 10 10 10 10 10 Multiethnic 3 6 6 6 6 6 Other 0 0 0 0 0 0 Missing 4 9 9 9 9 9 Low probability 113 84 84 84 83 82 AA 61 50 50 50 49 48 HA 0 0 0 0 0 0 CA 7 5 5 5 5 5 Biethnic 23 15 15 15 15 15 Multiethnic 9 6 6 6 6 6 Other 1 1 1 1 1 1 Missing 12 7 7 7 7 7 QLP -- -- -- -- -- -- AA -- -- -- -- -- -- HA -- -- -- -- -- -- CA -- -- -- -- -- -- Biethnic -- -- -- -- -- -- Multiethnic -- -- -- -- -- -- Other -- -- -- -- -- -- Missing -- -- -- -- -- -- SASSI-A2 Rule Variable 7 8 9 VAL SCS Gender High probability 51 53 54 54 54 Men 20 20 20 20 20 Women 32 34 34 34 34 Low probability 81 79 78 48 47 Men 32 32 31 19 18 Women 49 47 47 29 29 QLP -- -- -- 30 31 Men -- -- -- 12 13 Women -- -- -- 18 18 Ethnicity High probability 51 53 54 54 54 AA 21 21 21 21 21 HA 1 1 1 1 1 CA 4 4 5 5 5 Biethnic 11 11 11 11 11 Multiethnic 6 7 7 7 7 Other 0 0 0 0 0 Missing 9 9 9 9 9 Low probability 80 78 78 48 47 AA 47 46 46 27 26 HA 0 0 0 0 0 CA 5 5 4 1 1 Biethnic 14 14 14 12 12 Multiethnic 6 5 5 2 2 Other 1 1 1 1 1 Missing 7 7 7 5 5 QLP -- -- -- 30 31 AA -- -- -- 20 21 HA -- -- -- 0 0 CA -- -- -- 3 3 Biethnic -- -- -- 2 2 Multiethnic -- -- -- 3 3 Other -- -- -- 0 0 Missing -- -- -- 2 2 Note. Rule 1 = Frequency of Alcohol Use scale (FVA) or Frequency of Alcohol and Other Drug Use scale (FVOD) [greater than or equal to] 12; Rule 2 = Family-Friends Risk scale (FRISK) [greater than or equal to] 5; Rule 3 = Symptoms scale (SYM) [greater than or equal to] 5; Rule 4 = Subtle Attributes scale (SAT) [greater than or equal to] 9; Rule 5 = Obvious Attributes scale (OAT) [greater than or equal to] 4 and Defensiveness scale (DEF) [greater than or equal to] 10; Rule 6 = OAT [greater than or equal to] 7, SAT [greater than or equal to] 6, DEF [greater than or equal to] 2, and Supplemental Addiction Measure scale (SAM) [greater than or equal to] 4; Rule 7 = FVA or FVOD [greater than or equal to] 7, FRISK or Attitudes scale or SYM [greater than or equal to] 3, and OAT [greater than or equal to] 5; Rule 8 = FVA or FVOD [greater than or equal to] 5, OAT [greater than or equal to] 4, and DEF [greater than or equal to] 7; Rule 9 = FVA or FVOD [greater than or equal to] 5, SAT [greater than or equal to] 3, DEF [greater than or equal to] 4, and SAM [greater than or equal to] 3; VAL = Validity Check scale (VAL [greater than or equal to] 5); SCS = Secondary Classification scale (SCS [greater than or equal to] 16); QLP = questionable low probability; AA = African American; HA = Hispanic American; CA = Caucasian American.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Substance Abuse Subtle Screening Inventory|
|Author:||Perera-Diltz, Dilani M.; Perry, Justin C.|
|Publication:||Journal of Addictions & Offender Counseling|
|Date:||Apr 1, 2011|
|Previous Article:||Emotional skills-building curriculum.|
|Next Article:||Adolescent substance abuse treatment: organizational change and quality of care.|