Attention Research Update
"Helping parents, professionals and educators stay informed about new research on ADHD"
David Rabiner, Ph.D. Associate Research Professor, Duke University
Support for Attention Research Update is provided by SOAR
Is ADHD Overdiagnosed?
Is ADHD overdiagnosed? Despite widespread concerns that
this occurs, a study that specifically addresses this issue has not been
conducted. Thus, although it is well established that many children
with ADHD are never identified or treated, the extent to which children are
incorrectly diagnosed with ADHD is not known.
There are several reasons to be concerned about overdiagnosis. First, it may lead children to be inappropriately treated with stimulant medication when they do not need it. Second, it may contribute to children not receiving treatment that would better address another condition they actually have. Third, it could contribute to increased health care costs for society as a whole. Issues related to possible stigma are also important.
The DSM-IV-TR sets out clear diagnostic criteria for ADHD. In addition to a minimum number of inattentive and/or hyperactive-impulsive symptoms, children must show impairment from these symptoms in at least two settings, the symptoms must cause clinically significant impairment in social or academic functioning, they must have been associated with some impairment before age 7, and they must not be better explained by another mental disorder such as a mood or anxiety disorder. For a more complete presentation of the diagnostic criteria, go to http://www.helpforadd.com/criteria-for-add/
Unfortunately, some clinicians who diagnose ADHD in children may not carefully follow the diagnostic guidelines. Instead, clinicians' may focus on the presence of particular symptoms that they believe are central to the disorder, and, when these are present, fail to consider whether all the necessary additional criteria are met.
It is possible that this partially explains why many more boys are treated for ADHD than girls. Specifically, boys with ADHD are more likely than girls with ADHD to display disruptive behavior. If clinicians regard disruptive, impulsive, and hyperactive behavior as to the disorder it could contribute to boys being overdiagnosed and girls being underdiagnosed. That is because when such behavior is present, clinicians may be less attentive to necessary diagnostic criteria that are not met. And, when it is not evident, clinicians may disregard or at least underemphasize the presence of other ADHD symptoms when making their diagnostic decision.
A study published recently in the Journal of Consulting and Clinical Psychology presents a very interesting look at this issue [Bruchmuller et. al., (2011). Is ADHD overdiagnosed in accord with diagnostic criteria? Overdiagnosis and influence of client gender on diagnosis. Journal of Consulting and Clinical Psychology, 80, 128-138. The study was conducted in Germany and involved over 470 psychiatrists, psychologists, and social workers who were licensed to practice child and youth psychotherapy. These clinicians were representative of all such mental health professionals licensed to work with children in Germany.
The design of the study was both clever and straightforward. The authors created 2 series of 4 different case vignettes - one series depicting a school-aged boy and an identical series describing a girl. In one vignette for both genders, the information provided was fully consistent with ADHD, Combined Type. That is, not only were a sufficient number of specific symptoms described, but other information in the vignette made it clear that the symptoms were impairing the child in at least 2 settings, were creating impairment judged to be clinically significant, had emerged prior to age 7, and that alternative explanations for the child's symptoms could be ruled out.
In the other 3 vignettes, the required diagnostic criteria were not present. In one of the vignettes, symptoms were described as present only in one setting (school) and as having only appeared when the child was 9. In another vignette, not only were only 3 total symptoms described, but these were reported to be present in only one setting and not to have appeared before age 9. In the final vignette, the behaviors described fulfilled the diagnostic criteria for Generalized Anxiety Disorder, which includes symptoms of restlessness, nervousness, and concentration difficulties that can be mistaken for ADHD. Thus, the children described in these vignettes should never have been diagnosed with ADHD according to current diagnostic criteria.
These vignettes were pretested using 14 expert diagnosticians instructed to carefully apply complete diagnostic criteria to determine whether the child qualified for an ADHD diagnosis. All clinicians diagnosed ADHD for the vignette where full criteria were met and none diagnosed ADHD to any of the remaining vignettes.
One vignette was selected at random to be mailed to each of 1000 mental health professionals across the county. They were instructed to read the vignette carefully and to determine what, if any diagnosis, should be assigned to the child described. Of the thousand participants invited to participate, responses were provided by over 470. These responses were categorized as ADHD diagnosis or no ADHD diagnosis. The latter category included any diagnosis other than ADHD, statements indicating that insufficient information was available to make any diagnosis, no diagnosis, and suspected ADHD.
For Vignette 1, where the child described met full diagnostic criteria for ADHD, approximately 79% of the therapists diagnosed ADHD. Nearly 10% stated they did not have enough information and just over 4% indicated 'suspected ADHD'. The remaining 7% of clinicians assigned a diagnosis other than ADHD, most often an adjustment disorder. These cases represent false negatives, i.e., assigning a diagnosis other than ADHD when ADHD should have been diagnosed.
For the vignettes where criteria for ADHD were not met, nearly 17% diagnosed ADHD. This represents the false positive rate, i.e., assigning a diagnosis of ADHD when it did not apply. Clinicians were thus more than twice as likely to diagnose ADHD when they should not have than to assign a different diagnosis when they should have assigned ADHD, i.e., a false positive rate of 17% compared to a false negative rate of 7%. And, when an ADHD diagnosis was assigned, clinicians were far more likely to indicate that they would recommend medication treatment. Another 6% indicated 'suspected ADHD'. Fifty-seven percent made another diagnosis and just over 10% indicated they would make no diagnosis at all.
Diagnostic accuracy rates in relation to gender yielded very interesting findings. For the vignette where ADHD criteria were met, clinicians were equally likely to diagnose ADHD regardless of whether the vignette described a boy or a girl.
However, for the remaining 3 vignettes, clinicians were significantly more likely to diagnose ADHD when the child described was a boy. In fact, when averaged across the 3 vignettes, they were over twice as likely to incorrectly diagnose ADHD in boys than in girls, i.e., 23% vs. 11%.
What's more, these differences depended on the gender of the clinicians. Female clinicians did not differ in the rate of false positive and false negative diagnoses according to whether the child described was a boy or girl. For male clinicians, however, the false positive rate clearly depended on the child's gender, and was incorrectly assigned by 39% of the clinicians diagnosing boys vs. only 13% of clinicians diagnosing girls. Thus, male clinicians diagnosed ADHD nearly 40% of the time in boys who did not meet diagnostic criteria. This is strikingly high.
The authors also examined factors other than gender of the clinician that were associated with the overdiagnosis of ADHD, including years of experience, reported familiarity with the DSM diagnostic criteria, theoretical orientation (e.g., cognitive behavioral vs. psychodynamic), and professional occupation (i.e., psychiatrist, psychologist, or social worker). None of these factors were found to be significant predictors of the diagnoses that clinicians assigned.
Summary and Implications
Results from this study suggest that ADHD may be over identified by clinicians in that roughly 17% diagnosed ADHD for a child where full diagnostic criteria were not met. In contrast, when all ADHD diagnostic criteria were present, a diagnosis other than ADHD was assigned only 7% of the time. The misdiagnosis of ADHD is likely to lead some children to be placed on medication inappropriately, as medication was far more likely to be a recommended treatment when the diagnosis of ADHD was assigned.
Also important was the finding that boys were more likely to receive an ADHD diagnosis than girls even when the symptoms described were identical. And, male clinicians were far more likely than female clinicians to misdiagnose ADHD in boys. In fact, nearly 40% male clinicians incorrectly assigned a diagnosis of ADHD when the male child described did not meet diagnostic criteria.
Another interesting result was findings pertaining to the diagnosis of boys and girls when ADHD diagnostic criteria were met. Recall that for this vignette, clinicians were equally likely to diagnose ADHD regardless of whether the child described was male or female. In conjunction with the above, these suggest that the potential overdiagnosis of males may be a greater problem then the underdiagnosis of females.
This study does not specifically examine why many clinicians diagnosed ADHD incorrectly. The authors suggest that since each vignette described some ADHD symptoms, clinicians may have been basing their decision on the presence of several prominent symptoms rather than carefully determining whether all necessary diagnostic criteria were present. This does not necessarily indicate a problem with the diagnostic criteria per se, but rather in how even experienced clinicians apply them.
All studies have limitations and the authors are careful to acknowledge limitations of their study. First, they note that results obtained with clinicians in Germany may not generalize to other countries like the US. Although I am not aware of any reason to believe findings with US clinicians would be substantially different, it would be important to conduct a similar study here.
One could also question the validity of diagnostic decisions that are based on written case vignettes compared with real-world clinical settings. In particular, the vignette methodology does not enable clinicians to gather additional information to confirm or disconfirm his or her decision. However, even when diagnosing based on case vignettes, clinicians should still have applied the diagnostic criteria. And, conducting a study of this issue in real clinical settings would have its own set of challenges.
One other limitation worth noting is that the vignette depicting a child who met ADHD diagnostic criteria was limited to the combined type of the disorder, i.e., both inattentive and hyperactive-impulsive symptoms were present. And, in the vignettes where ADHD should not have been diagnosed, some inattentive and hyperactive-impulsive behaviors were described. To complement the interesting results obtained, future work should examine how the findings may or may not change when children with the inattentive and hyperactive-impulsive subtypes are described. It would be interesting to learn whether the tendency towards overdiagnosis of ADHD varies by subtype and whether this varies by gender for the different subtypes.
It is also important to emphasize that the findings reported here do not negate the fact that many children who would qualify for an ADHD diagnosis, and could certainly benefit from treatment and school-based services, are never identified or treated. Thus, results from this study not withstanding, the underdiagnosis of ADHD remains a serious problem and it is important not to lose sight of that.
In summary, this interesting study demonstrates that not all clinicians follow DSM-IV criteria "...requirements to base their diagnosis on a thorough evaluation of relevant diagnostic criteria." Preliminary evidence suggests this may especially be true for male clinicians evaluating male children. Diagnosing children incorrectly can potentially lead to a number of adverse consequences and this research will hopefully contribute to raising awareness of this important issue.