DSM-5 has just been published—not a happy moment in the history of psychiatry or for me personally. It risks turning diagnostic inflation into hyperinflation—further cheapening the currency of psychiatric diagnosis and unleashing a wave of new false epidemics. The economic equivalent would be printing up loads of new money when prices are already rising way too fast. DSM-5 is a cautionary tale of soaring ambition, poor execution, and a closed process.
Allen Frances, Former Chair of DSM-IV Task Force
It has been almost a year since DSM-5 was released in May 2013 by the American Psychiatric Association (APA). Critics have lined up to outline significant concerns about the consequences that will result from its use. Among many issues, DSM-5 has been lambasted for its overpathologizing of normal, fluctuating patterns of development, aging trends, social responses, societal fads, and general behavioral patterns. The controversy that has ensued has brought about significant questions about the future of DSM. Agencies even in my home state have begun to look at the possibility of discarding its use altogether in favor of ICD classification, even though the implementation of ICD-10 continues to be uncertain. Medical systems such as that which I am employed continue to hang in balance about what will occur. Much of this confusion has been well-documented.
But what have not been well-documented are the minute, at times peculiar, changes that are present in the specific verbiage of the conditions themselves as detailed in DSM-5. For interests of both thoroughness and brevity, I will focus on two conditions that pertain largely, but not solely, to children.
Let’s start with ADHD. In the DSM-IV-Text Revision (DSM-IV-TR), a diagnosis of ADHD required that a few specific criteria be met. One, all individuals (whether adult or children) must have had six (or more) symptoms of either inattention or hyperactivity-impulsivity. Symptoms must have been present before the age of seven. Some impairment must have been present in two or more settings. And, there must have been clear evidence of clinically significant impairment in social, academic, or occupational functioning.
In DSM-5, a diagnosis of ADHD requires the following. For youth ages 16 and younger, six (or more) symptoms must be present; only five symptoms are required for those 17 and older. Symptoms must be present prior to the age of twelve. Symptoms must be present in two or more settings (with no clear indication of impairment in these settings). And, there is clear evidence that the symptoms interfere with, or reduce the quality of, social, academic, or occupational functioning.
Upon initial reflection regarding DSM-5 changes, what is readily discernible is that the diagnosis of ADHD in adults, and those evaluated at a later age, is easier to make based on fewer symptoms being required and later age of onset. But closer analysis also notes the change from “impairment” in DSM-IV to “present” and “interfere[nce]” in DSM-5. Impairment (by dictionary.com standards) is the “state of being diminished, damaged, or weakened…” whereas interference (based on the root word interfere) is “to come into opposition with one thing or another…” Power outages, detours, and illnesses often interfere with our daily lives. Thought of in another way, a 2-year-old’s difficulty in sitting for an extended period of time often interferes (as it should be, given his or her age) with doing certain activities. But, the question remains, does “interference”, or “presence”, or even “reduction of quality,” constitute the diagnosis of a disorder? It seems that with a faculty as central to our being as attention, it ought to be impairing to constitute a diagnosis. Regardless of opinion, it is clear that as with other noted changes, these subtle alterations in terminology renders a diagnosis of ADHD easier to give.
Beyond these curious changes, though, there are a couple of other comparisons that deserve illumination. Although the ADHD diagnosis no longer uses the word “impairment”, other diagnoses, including those often given to children (e.g., autism spectrum), still retain this term and do not substitute the idea of “interference.” It is peculiar that changes to one condition would allow for less severity while others retain a more stringent approach. The ability to focus and attend remains as basic to our functioning as do social communication or emotional regulation skills. Thereby, it seems to have as much potential for impacting an individual as any other personal faculty that could be compromised.
Finally, in wrapping up questions regarding ADHD, there is the question regarding the Not Otherwise Specified (NOS) category. In DSM-IV-TR, ADHD-NOS was used for “prominent symptoms of inattention or hyperactivity/impulsivity” that did not meet full criteria for any category of ADHD. In DSM-5, the NOS category has been divided in two ways, and this new division is seen in many other DSM conditions. There is “Other Specified” ADHD and “Unspecified” ADHD. Like the previous NOS category, both are used for symptoms characteristic of ADHD, but which do not meet full criteria. The primary distinguishing factor between the new category is that in “Other Specified,” the “clinician chooses to communicate the specific reason that the presentation does not meet the criteria for ADHD or any other specific neurodevelopmental disorder. In “Unspecified”, the clinician chooses not to specify the reason that criteria are not met. In communication with a colleague about the APA’s unofficial response regarding this question, the response (underline added) was as follows:
The APA would really prefer clinicians to use the other specified category, because it conveys clinical information about patients that do not neatly fit into one of the criterion sets but who, from a functional/severity standpoint, might as well (i.e., similar likely prognosis, treatment recommendations, etc.)…. the other specified diagnosis offers a way to reduce rates of NOS category (which is basically the same as unspecified) while maintaining the basic structure of [that] system.
I must admit I don’t understand this justification. The old NOS term never mentioned anything about specifying or not specifying reasons for its use, and I as a clinician can think of no reason why I would ever “choose not to specify” why I made any diagnosis I gave, Unspecified or otherwise. As a clinician, I also worry that this approach further mystifies what we do in regards to evaluating children and families as if there were a reason we needed to withhold justification behind our diagnostic impressions.
Moving on to the topic of Learning Disorders (LD), significant questions also remain in this domain. In DSM-IV-TR, a learning disorder (whether in reading, mathematics, or written expression) was diagnosed when achievement/ability was “…substantially below that expected given the person’s chronological age, measured intelligence, and age-appropriate education.” Therefore, the diagnosis required a standardized measure of both achievement and intellectual skills. Historically, challenges have always been present with this diagnosis in some ways. Issues regarding cultural sensitivity of IQ tests, overlap between achievement and intellectual skills, and difficulty in assessing environmental/societal factors (that may have had a significant impact on a child’s performance in academics) have resulted in diagnostic challenges. But, in general, the idea with DSM-IV was that an individual with an LD was performing significantly below their general reasoning/problem-solving abilities in a specific area (as with phonetic decoding and phonemic awareness skills) which necessitated further intervention. This distinction remains important as all but a small percentage of children with early reading difficulties (in the presence of average reasoning and cognitive skills) can be remediated through more intensive educational strategies if identified early in elementary school.
In DSM-5, a diagnosis of an LD now occurs if “the affected academic skills are substantially and quantifiably below those expected for the individual’s chronological age.” In addition, “for individuals ages 17 years and older, a documented history of impairing learning difficulties may be substituted for the standardized assessment.” Immediately, two things should be noted. One, the new provisions no longer require the use of a measure of intelligence as comparison to a person’s achievement/ability performance. I will address this further in the following paragraph. Two, for older teens and adults, no actual evaluation is needed to diagnosis the condition, just a “documented history.” In addition to concerns about what is meant by “documented history,” it implies that once you have an LD at this age, you will always have an LD even though plenty of standardized measures exist to still examine this in real time. In general, research would generally support the idea that intellectual and achievement skills do not typically shift after the late teens and beyond. But, with the dramatic differences that environmental/societal factors can have on educational attainment, it begs the question about whether many still have the opportunity to learn new skills—in essence, overcoming a previous diagnosis of a “learning disorder.” Many certainly do, as I detailed before about George Dawson and his newfound literacy at the age of 98.
Back to the topic of removing intelligence tests, some would advocate for this approach given previously mentioned issues that have blighted their use. It would also more likely allocate services to those who did not receive a diagnosis in the past (e.g., due to lower levels of intelligence). However, an assessment of an individual’s patterns of intellectual strengths and weakness through administration of an intelligence test still remains one of the more effective ways to work with parents and teachers in differentiating instruction so that a student can succeed. Interestingly, though, DSM-5 actually does not do away with the use of intelligence tests after all. In differentiating diagnoses, it states that an LD must be distinguished “from general learning difficulties associated with intellectual disability [previously Mental Retardation in DSM-IV], because the learning difficulties occur in the presence of normal levels of intellectual functioning (i.e., IQ score of at least 70±5).” First of all, “normal” is a misnomer as anyone reading this would assume that normal means a child should be able to function reasonably well with other same-aged peers. But an IQ score of 65 (70-5) is at the 1st percentile, and a score of 75 is at the 5th percentile. Children in this range (and even above) have difficulties keeping pace with same-aged peers unless significant, additional assistance is given. Average (and above) in statistical terms include standard scores 90 and greater. Clearly “normal” and “average” are not synonymous here. Strangely, too, it appears that DSM-5 itself even agrees with this inconsistency albeit unintentionally. Borderline Intellectual Functioning (BIF) was previously given in DSM-IV-TR as V-code for individuals with an IQ of 71-84, which now DSM-5 calls “normal” earlier in the manual. A V-code is an “other condition that may be a focus of clinical attention.” In DSM-5, BIF remains as a V-code, but now simply provides the vague statement to define what it is without any statistical parameters. It states “This category can be used when an individual’s borderline intellectual functioning is the focus of clinical attention or has an impact on the individual’s treatment or prognosis.” But what exactly is borderline intellectual functioning?
Regardless of opinion or attempted clarification, once again the new criteria do make it easier to diagnose an LD as with ADHD. As noted earlier, some may argue that these were the children that slipped through the cracks before, especially those with lower than average intelligence. They may be right to some extent. But the price paid will be more diagnoses given for what many argue are associated with a larger pattern of strengths and weaknesses that exist in the population.
As whole, these changes illustrate two themes in regards to DSM-5. One, in making diagnosing conditions easier, it continues to bring serious questions about what constitutes a disorder versus a pattern of strengths and weaknesses inherent to the human condition. Two, DSM-5 clearly gives clinicians much more leeway in deciding what a specific disorder ultimately is. It risks diluting things even further for those of us who work in facilities that require DSM codes to be billed and also for the general public in understanding what all this means. So, it leaves me with a lot of questions as I know many others have, but two in particular that the APA, as all advocacy organizations, must continually ask of themselves. What will the consequences (intended and unintended) of these changes be? And do these changes reflect the best interests of the people they serve?