Tuesday, December 17, 2013

Comment on NY Times article

On December 15th 2013 Alan Schwartz of the NY Times published an extensive piece about "Selling ADHD."  The article in part was based on several interviews with me. Already at least one other interviewee has complained that a quote of his was "taken out of context." This of course is a well-known stratagem of someone who dislikes the implication ascribed to him or her by the reporter. Now some of my close friends have asked whether I really said the things ascribed to me.

Let me respond by saying that the reporter in this case got my quotes exactly right. After a long career in ADHD research and clinical practice I was aware of the importance of what I was about to say and its significance in the field. Moreover, I was impressed with the research and methods being used by Mr. Schwartz.  He began by telling me, "We don't want opinions, we want documents,"
not the "he-said-she-said approach." He made it clear that although he respected my opinion because of my role in the sometimes controversial arena of ADHD diagnosis and treatment, he needed hard evidence, not hearsay.

Moreover, as we talked in several interviews,  I became aware of how extensively Mr. Schwartz had already spoken to key leaders in the field and read many substantive research papers dealing with the prevalence of ADHD and the statistics regarding the explosion of stimulant drug prescriptions. I learned he was unbiased, balanced, and not out to demonise ADHD, Big Pharma, Doctors, or anyone else. Indeed, during that time I became aware that Mr. Schwartz received an award from the American Statistical Society.

However, we did not agree on everything. I was inclined to see the current situation and the opinions of participants in the debate in a historical context involving many factors, whereas Mr. Schwartz was aware of the present status , but limited by the number of words available to him in print. I even suggested that the topic deserved a book to fully understand all of the interconnected influences.  So I could say that everythingin this article is true, but not necessarily the whole truth as I choose to see it.

Monday, September 9, 2013

Prevalence of ADHD from APSARD talk Sept. 29, 2013

Ever since it was minimal brain damage, or hyperactivity disorder, minimal brain dysfunction, or ADD, or DSM-IV Attention Deficit Hyperactivity Disorder (ADHD), people have been curious to know how many of them there are. Early local studies placed the numbers somewhere between 2% and 20%.
This remarkable disparity obviously raises the question of who defines what a “case” is, the methods used  to arrive at the numbers, and the relationship between the “diagnosis” and the treatments that result. Until recently all studies were estimates made from clinical samples or local regional studies without regard to the factors of  age limits, ethnicity, demographics or gender. Sometimes it is the numbers relating to use of stimulant drugs and their misuse or abuse that prompts the question of how many ADHD there are. So in this presentation I will focus on all three questions: The nature of the “diagnosis”, the impact of diagnosis on the numbers, and the role of pharmacotherapy and other factors in determining the numbers.

Bell was a famous
Scottish lecturer at the medical school of the University of Edinburgh in the 19
th century. Bell emphasized importance of close observation in making a diagnosis. To illustrate this, he would often pick a stranger and, by observing him, deduce  his occupation and recent activities, These skills caused him to be considered a pioneer in forensic science.

Doctor Sir Arthur Conan Doyle was the most famous student of Bell, passing on Bell’s wisdom of close observation in diagnosis to his creation, Sherlock Holmes, and himself as Holmes’ companion, the bumbling Dr. Watson. A constant reminder of Holmes to Watson was, “You see Watson, but you do not observe,” implying that seeing without close observation led to mistaken conclusions. 

An apocryphal story illustrates the key point: Holmes & Watson are on a camping trip, and lying on their sleeping bags and peering at the sky, Holmes asks, “What do you see, Watson?” Watson replies, Well Holmes, I see the milky way, and countless stars and…” Holmes interrupts and points out, “You fool Watson, someone has stolen the tent! YOU SEE BUT YOU DO NOT OBSERVE!”

Justice Potter Stewart made this observation when in the midst of a Supreme Court debate on the definition of pornography. Apparently seeing without further thought also infects the legal profession!
Early prevalence estimates suffered from generalizing from clinical samples without attention to the demographic, gender, ethnicity, and diagnostic rules.Many of the studies used local school districts as the unit of observation and seldom distinguished between teacher and parent definitions or verified physician diagnosis.

Many of the studies used “hyperkinesis” as the defining syndrome marker. Easily seen, but a characteristic of many conditions that would require more exact observation to sort out comorbities or other diagnoses. Nevertheless, one sees here such a diverse range of prevalence estimates that it casts doubt upon the validity of all.

An exception to most of the other studies was a very comprehensive approach by  Bosco and Robin in Central Michigan in 1980. They used verified physician diagnoses, and data from both parents and teachers, paying careful attention to demographic factors.
They received data from 9293 teachers and 7248 parents. Rates of 3.16% and 3.38% “represent the highest possible prevalence rates obtainable from our data…” The VERIFIED prevalence for physician diagnosis was only 2.92%. Only 1.79% had been treated with stimulants in the previous 5 years. This report deserves more creditability than all of the extreme numbers. Bosco, J. J., & Robin, S. S. (Eds.). The Hyperactive Child and Stimulant Drugs. Chicago: University of Chicago Press, 1977.

J Am Acad Child Adolesc Psychiatry. 2000 Aug;39(8):975-84; discussion 984-94.
Stimulant treatment for children: a community perspective.
Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, Durham, NC 27710, USA.
To examine the use of prescribed stimulants in relation to research diagnoses of attention-deficit hyperactivity disorder (ADHD) in a community sample of children. METHOD:
Data from 4 annual waves of interviews with 9- to 16-year-olds from the Great Smoky Mountains Study were analyzed. RESULTS:
Over a 4-year period, almost three quarters of children with an unequivocal diagnosis of ADHD received stimulant medications. However, girls and older children with ADHD were less likely to receive such treatment. Most children with impairing ADHD symptoms not meeting full criteria for DSM-III-R ADHD did not receive stimulant treatment. Stimulant treatment in this group was significantly related to the level of symptoms reported by parents and teachers and was much more common in individuals who met criteria for oppositional defiant disorder. The majority of individuals who received stimulants were never reported by their parents to have any impairing ADHD symptoms. They did have higher levels of nonimpairing parent-reported ADHD symptoms, higher levels of teacher-reported ADHD symptoms, and interviewer-observed ADHD behaviors, but these typically fell far below the threshold for a DSM-III-R diagnosis of ADHD.CONCLUSIONS:In this area of the Great Smoky Mountains, stimulant treatment was being used in ways substantially inconsistent with current diagnostic guidelines
"Stimulant treatmentwas much more common in individuals who met criteria for oppositional defiant disorder. The majority of individuals who received stimulants were never reported by their parents to have any impairing ADHD symptoms [which] typically fell far below threshold for a DSMiii-R diagnosis of ADHD."

By cleverly intermixing questions about prescriptions for ADHD medicines with questions about doctors diagnosing ADHD, telephone interviews of parents produces a wide range of “cases” for determining prevalence.
To be sure, this method produces some curious data. For example, why is the prevalence so low in the southwest? And what about Illinois? We might understand Texas, where the death rate from executions may be snuffing out potential ADHD candidates. But what of North Carolina, with the highest rate near 15 percent?
And no, it is not because I happen to live there! It happens to be one of the lowest states for using the Conners’ scales, with cultural lag possibly accounting for North Carolinians fondness for the Achenbach Scales (Just kidding). Actually, research recently showed that the explosion of "diagnoses" in NC is caused by educational policies which shifted ADHD from regular classes to avoid the reimbursement penalties of new requirements imposed by end of year testing.

Children 3-17
5 million children (9% of this age group) have ADHD
Boys (12%) twice as likely as girls to have ADHD

Incidence of ADHD diagnoses increased an average of 3% annually between 1997-2006

In 17 regional or multisite studies utilizing DSM3 or DSM4 prevalence rates ranged between 2% to 26%
Data obtained from The National Health and Nutrition Examination Survey (NHANES), an annual multistage probability sample survey of non-institutionalized US population
Data from 2001 to 2004 obtained on 3082 children ages 8-15 using DISC IV by telephone.
8.7% met DSM-IV criteria for ADHD in the year prior to the survey = 2.4 million children
An additional 3.3% did not meet DSM-IV criteria but had both a parent-reported prior diagnosis of ADHD and treated with stimulants at some time during the past 12 months
Rates for boys were higher than girls (11.8%) vs 5.4%).
Of those meeting DSM-IV criteria, only 48% reported receiving a diagnosis by a health professional in the prior year
38.8% who met criteria reportedly received medication at some time in the prior year.
Froelich et al choose to interpret the 3.3% who were medicated but not meeting DSM criteria as those correctly diagnosed and successfully treated and thus no longer showing symptoms. But one could equally argue that they NEVER met ADHD criteria and were given stimulants incorrectly in children who were in fact miss-diagnosed.

Note that their estimates of that group plus the 8.7% reaches the phenomenal level of 15% prevalence, just as the upper bound estimated by the CDC.

The consensus conference was modeled (it might appear) on the example of the Spanish Inquisition searching for witches. One might imagine that here I am Pleading the case for ADHD as a reliable diagnostic category. I said, in the 15 minutes I had to present, “Formal diagnostic criteria for the disorder underwent rapid changes as new syntheses and accumulation of data from field trials took place. The fact that the concept of ADHD has evolved with changing evidence should be taken as a strength, not as a sign of unreliability or vague conceptualization. Comprehensive review of the evidence regarding diagnosis and treatment carried out by independent expert medical reviewers concludes that diagnostic criteria for ADHD are based on extensive empirical research and, if applied appropriately, lead to the diagnosis of a syndrome with high interrater reliability, good face validity, and high predictability of course and medication responsiveness (Goldman, Genel, Bezman, et al., 1998)."

However, I qualified my judgment by adding,  "Important areas of our knowledge about ADHD remain to be clarified. Developments in cognitive neuroscience point to the multidimensional nature of both attentional processes and activity level, yet these concepts are poorly operationalized by current symptomatic criteria. Neuropsychological studies demonstrate a clear heterogeneity in samples of ADHD defined solely by symptomatic criteria (Conners, 1997). Doubts have been raised about the current nosological subtyping and the possibility that inattention and hyperactivity-impulsivity reflect separate disease entities (Barkley, 1998b). Current diagnostic criteria require that symptoms be more frequent and severe than are typically observed in individuals at a comparable level of development (American Psychiatric Association, 1994), but marked variations in the application of this rule lead to serious underdiagnosis or overdiagnosis, resulting in excesses or deficiencies of pharmacologic treatments (Angold, Costello, 1998).

The embarassment of riches from neuroimaging studies reflects a poor understanding of any specificity for the neural basis of ADHD. The high levels of comorbidity of ADHD with oppositional, conduct, and mood disorders also call into question the specificity of the definition of the disease and whether current criteria are sufficient to allow further understanding of the neurobiology of the syndrome."

The list of speakers at the conference attests to the wide range of expertise from many different disciplines regarding the reliability and validity of ADHD. What a knowledgeable group; every body who knows anything about ADHD in 1998! This formidable group had to submit highly structured and detailed chapters in advance to the group of prominent but non-involved judges whose job was to determine the validity and reliability of diagnosis of ADHD and the value of its treatments.

An impartial group of scientists and scholars served as a judging panel based on the oral presentations and extensive documentation provided to them by the speakers

Mark Vonnegut, son of Kurt Vonnegut, is a very interesting and funny man. A pediatrician, his memoir details his days of severe psychosis and hospitalizations.
Despite this extraordinary handicap he managed to become a pediatrician and spokesman for the plight of the mentally ill.

When the Director of NIMH at the consensus conference called upon the eminent panel for comments about what they had heard (and presumably read), Mark was the only one with the courage to speak up. He inadvertently supplied the rubric shared by the vast body of providers and critics about ADHD. Once again, seeing without observing. Had the eminent panel actually read all of the detailed documents? Was he the only one?? Probably not.

Compilation of studies is another method for establishing prevalence. Unlike the CDC and Froelich’s data on a national representative sample of telephone surveys of a parent, these data appear to give much more modest prevalence estimates.

These data show the rapid increase in both diagnosis and  treatments with stimulant drugs over a 10 year period.

The increase over a 5 year span shows that the greatest changes are the increases in diagnosis of young adults
Though humorous, this cartoon depicts a common misconception that BIG PHARMA is responsible for the increase in prescriptions for psychiatric drugs. Overlooked is the fact that they simply behave like any American corporation whose job is to make money, and sell their drugs to those whom doctors write a prescription. Vigilance in their practices is called for like oversight of any large corporation but mental health professionals who prescribe are the gate that opens a flood of pharmaceutical sales by those who presumably have made a valid "diagnosis." A presumption that needs to be called into question by the impossible estimates of ADHD prevalence.

The seriousness of the flood of stimulant medications is shown by the data from emergency rooms between 2005 and 2010. Visits involving ADHD and stimulants increased from 13,379 to 31,244 visits over the 10 year span.

The problems of diagnosis of ADHD are compounded with adults for many reasons. The need for a sound clinical history is obviously absolutely essential.

Although parents are not necessarily available for gaining critical developmental information, the contribution of significant others who know the patient well is often practicable and important for verifying the subjective report by the patient.

Depressed Dopamine Activity in Caudate and Preliminary Evidence of Limbic Involvement in Adults With Attention-Deficit/Hyperactivity Disorder.
Nora D. Volkow, MD; Gene-Jack Wang, MD; Jeffrey Newcorn, MD; Frank Telang, MD; Mary V. Solanto, PhD; Joanna S. Fowler, PhD; Jean Logan, PhD; Yeming Ma, PhD; Kurt Schulz, PhD; Kith Pradhan, MS; Christopher Wong, MS; James M. Swanson, PhD
Arch Gen Psychiatry. 2007;64(8):932-940. doi:10.1001/archpsyc.64.8.932.
Regression slopes between changes in dopamine (DA) in caudate and in putamen and scores on Conners Adult ADHD Rating Scales (CAARS) section E (DSM-IV symptoms of inattention) in subjects with attention-deficit/hyperactivity disorder (ADHD). Correlations correspond for left caudate (r = −0.49, P < .04), right caudate (r = −0.56, P < .02), left putamen (r = −0.61, P < .008), and right putamen (r = −0.71, P < .001).
Studies like this give some assurance that ADHD symptoms in adults have predictable functional anatomic bases. 

Careful look at the CAARS “Inattention” factor shows that the symptoms are primarily those of Executive Dysfunction.

Executive dysfunction is a characteristic of many psychiatric disorders, emphasizing the importance of thorough diagnostic workup for establishing that the client has ADHD and not one of the many alternatives

This powerful statement by Alan Frances, the prime moving force in the development of DSM-IV, may be seen on a video on YouTube.

Dr. Frances points out how the new criteria for DSM-5 have opened the doors for a huge increase in the number of normal behaviors which will now be interpreted as mental illnesses. He uses data from his own experience of losing his wife to illustrate how the new criteria for depression mistake normal bereavement for a mental disorder.

Data show already that the expected increase has affected "diagnoses" of ADHD.