Prior to the publication of the Diagnostic and Statistical Manual of Mental Disorders (DSM) first edition in 1952, there was no formal system of classification and diagnosis of substance use disorders in the United States. Substance use disorders have received increasing attention throughout the DSM’s next four major revisions, as is evident by the increasing proportion of the manual that is dedicated to this group of conditions. It is necessary to look at the diagnostic evolution of these disorders to understand the rationale behind the proposed changes for the fifth edition of the DSM.
Only one axis was used for the diagnosis of Substance Use Disorders in the first and second editions of the DSM. In the manual’s first edition, information regarding substance abuse was limited to only one page, and was listed under the heading of sociopathic personality disturbance. This suggests that at its inception, the DSM conceptualized substance use disorders as arising from specific personality features.
The DSM-III was the first edition to distinguish between substance abuse and dependence. In order to qualify for a diagnosis of dependence, the patient had to show evidence of physiological dependence as indicated by tolerance and withdrawal. Abuse was categorized by social and medical consequences of drug use. These major changes in diagnostic criteria were modeled from a theory of Alcohol Dependence Syndrome created by Edwards and Gross in 1976 (Martin et al, 2008). This model was generalized to all classes of drugs in the third and fourth editions of the DSM. The DSM-III was also the first edition to include multi-axial diagnoses, which led to the “bi-axial” concept of substance use disorders.
The DSM-IV would retain the distinction between abuse and dependence, but would add a hierarchical approach to diagnosis. This specified that substance abuse was to be considered a less severe diagnosis that would serve as a marker for the onset of substance dependence. Another major change in diagnostic criteria was that physiological withdrawal and tolerance were no longer required for a diagnosis of dependence. Abuse criteria was also expanded from two to four factors that were now categorized by negative legal and social consequences.
Several major changes to the diagnostic criteria of substance use disorders have been proposed for the DSM-V. The entire category of Substance-Related Disorders is to be renamed Addiction and Related Disorders. This is being proposed to signify the distinction between substance addictions and newly proposed disorders related to compulsive behaviors such as gambling. It has also been proposed that Dependence and Abuse diagnoses be subsumed under the heading of 'Substance Use Disorders'.
Another change that has been proposed for the next edition of the DSM is that the word dependence be limited to physiological dependence rather than being used as a label for addiction. While the terms dependence and addiction have been used interchangeably in the past, it is argued by many members of the American Psychological Association that the words have two very distinct definitions. In terms of formal medical lexicon, dependence is a normal and expected physiological response to the repeated use of drugs that act on the central nervous system which is distinctly different from compulsive drug seeking behavior common to addiction (Heit, 2009). This ambiguity often leads to problems of misdiagnosis and stigmatization of patients who require opioid medications for pain management. It has thus also been proposed that the DSM-V definitions of tolerance and withdrawal will include the specification that these criteria shall only be considered if substance use is not related to a prescribed medication that is taken as directed by a physician.
Empirical evidence from several studies has led to the recommendation that the diagnoses of Opioid Abuse and Opioid Dependence be subsumed under the heading of Opioid Use Disorder. The DSM-III and DSM-IV both assumed that abuse was a prodromal phase of dependence, but recent studies using latent-factor statistical techniques have shown that many people first meet dependence criteria before they qualify for a diagnosis of abuse. In fact, these studies examined the ‘problems with the law’ criterion of abuse, and found that the endorsement of this criterion was associated with the most extreme cases of substance-related problems (Wu et al, 2011). In general, the reliability of abuse criteria in many of these studies has been found to be much lower and variable than dependence criteria.
The current diagnostic distinctions between abuse and dependence have led to both ‘diagnostic orphans’ and ‘diagnostic imposters.’ The former refers to people who may meet two criteria for dependence but none for abuse. Since a person must meet three criteria for dependence or one criterion for abuse, these people are often left undiagnosed despite their severe substance-related problems. Since a diagnosis of abuse only requires one positive symptom the opposite situation may also occur. Diagnostic imposters are people who meet diagnostic criteria for abuse despite low levels of substance use problems. This is often seen with adolescents who meet the criteria of ‘significant problems with parents.’
Empirical evidence indicates that if a categorical structure does exist, the categories are most likely distinguished by severity rather than by groups of symptoms. This largest body of evidence comes from a 2010 study that analyzed the 11 total opiate use disorder criteria against a 2007 national survey on drug use and health that provided the largest sample of adult non-prescribed opioid users in the United States (WU et al, 2010). The results of this study indicated that the four abuse and seven dependence criteria were highly correlation, r=.98. More importantly, the factor analysis employed in this study identified two distinct groups of non-prescribed opioid users. The affected group (7%) had a statistically significant higher probability of endorsing the eleven Opioid Use Disorder criteria than the less affected group (93%) These findings suggest that users are likely to exist on a single continuum of severity. The inclusion of severity scales in the DSM-V would allow clinicians to make more specific treatment recommendations by identifying different risk groups for customized treatment options. For example, someone in the less severe range might benefit from motivational interviews, whereas someone on the opposite end of the spectrum would most likely require intensive inpatient treatment.
These issues must be resolved in the DSM V in order to ensure better clinical diagnosis, treatment outcomes, and related research emerge in the future. This is not only important for the United States as it is obvious that the DSM is used as a field manual throughout the world.