Investigating the appropriateness of current alcohol-use disorder criteria for adolescents


  • The same alcohol-use disorder criteria are currently used for both adults and adolescents, despite concerns about the appropriateness of these criteria for adolescents.
  • New findings suggest that "tolerance" and "time spent" using/obtaining/recovering from alcohol may be over-diagnosed in adolescents.

North American clinicians generally use alcohol-use disorder (AUD) criteria as defined in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). Concerns exist, however, about the appropriateness of these criteria for adolescents. For the first time, a study in the May issue of Alcoholism: Clinical & Experimental Research uses a single representative sample of the U.S. population to examine the effects of age, gender, race/ethnicity, and drinking status on the prevalence of DSM-IV diagnostic criteria among both adolescents and adults.

"DSM-IV criteria were developed with clinical adult populations," said Thomas C. Harford, a senior research analyst with CSR, Incorporated and first author of the study. "However, when compared to adults, drinking among adolescents is relatively infrequent and drinking histories tend to have shorter durations. Consequently, many symptoms such as withdrawal and alcohol-related medical complications are not typically experienced by adolescents."

"Recent research with adolescent clinical and community samples has identified important limitations of DSM-IV criteria when applied to adolescents," added Christopher Martin, associate professor of psychiatry and psychology at the University of Pittsburgh School of Medicine. "These include, one, that the dependence symptoms of tolerance, 'much time spent' using, and using more/longer than intended appear to often be over diagnosed in adolescent studies; two, that the ways in which these dependence criteria are measured has a huge downstream impact on the estimated prevalence of abuse and dependence diagnoses; and three, that certain AUD criteria may have different meanings when applied to adults and adolescents.

Harford noted that the majority of national surveys are limited to respondents 18 years and older. "There are a few national surveys which include both adolescents and adults, but sample sizes as well as the DSM-IV AUD assessments for adolescents are limited. This study draws upon a recent national survey with large sample sizes for both adolescents and adults and includes symptoms relevant for each of the DSM-IV AUD criteria. The large sample sizes provide better accuracy for estimating criteria, especially for those with low prevalence."

Harford and his colleagues used data from the 2001 National Household Survey on Drug Abuse. Of the 55,561 survey subjects, 33,576 (60.5%) reported alcohol use in the previous year and also provided information on DSM-IV AUD criteria. DSM-IV AUD criteria were assessed by questions related to specific symptoms that had occurred during the previous 12 months.

Results indicate that the most prevalent criteria of DSM-IV alcohol dependence were "tolerance" and "time spent obtaining alcohol, drinking, or getting over its effects." The most prevalent criterion of DSM-IV alcohol abuse was "hazardous use." The high prevalence of "tolerance," "time spent," and "hazardous use" among the adolescent sample is consistent with findings from other adolescent studies.

"Similar findings have been reported in clinical studies," said Harford, "and the age-specific distributions for these criteria conform to the age-specific distributions of alcohol use - increasing use with increasing age up to ages 21-22, followed by lower use with increasing among respondents aged 23 years and older. The lower proportion of adolescents endorsing these criteria, when compared to younger adults, is consistent with the higher proportions of non-drinkers among adolescent populations."

Martin concurred. "Similar to what has been suggested in research using only adolescents, this study found that, relative to other criteria, the dependence criteria of tolerance and time spent using were larger among adolescents and young adults compared to other age groups. For example, a great deal of time spent using, obtaining or recovering from alcohol may reflect difficulties in obtaining the substance rather than compulsive patterns of drinking. These data suggest that these two symptoms may be over-diagnosed in adolescents, which would lead to higher rates of dependence diagnoses in these age groups."

Results also indicate that variation in the wording of specific symptom items may have a strong influence on the diagnosis of abuse and dependence.

"In the present study, two items for 'tolerance' yielded similar proportions of positive response, but these proportions were shown to vary by age group," said Harford. "Higher proportions of younger respondents endorsed the item 'drank more for same effect,' and higher proportions of older respondents endorsed 'less effect from same amount.' Interpretation of this finding is not obvious. It may reflect age differences in the history of alcohol exposures for younger and older respondents. We don't know if reversal of the ordering of the two items would have lowered the prevalence estimates for tolerance among adolescents. The point to be made here is that variations in questions yield differences in the proportions of responses, which in turn, influence the estimates for criteria and overall estimates for AUD."

"These findings are important because valid diagnostic criteria are critical for advances in research, treatment and prevention," said Martin. "This study suggests ways in which criteria and the way they are measured could be improved to better serve the adolescent age group. Who is and is not given an AUD diagnosis among adolescents can help determine insurance coverage for treatment for a large number of youth."

Harford said that he will continue to examine, in greater detail, age-specific changes in DSM-IV diagnostic categories. He noted, however, that it is important to draw from both epidemiological and clinical studies. "General population or 'epidemiological' surveys are important because they provide estimates of symptoms and diagnoses in the general population," he said. "Clinical studies provide more detailed symptom assessments. Yet, because clinical studies draw upon adolescents in treatment settings in which the symptoms are well developed and severe, epidemiological studies provide broader perspective among healthy and more representative populations. Both approaches are complementary."

Source: Eurekalert & others

Last reviewed: By John M. Grohol, Psy.D. on 21 Feb 2009
    Published on All rights reserved.



The time when you need to do something is when no one else is willing to do it, when people are saying it can't be done.
-- Mary Frances Berry