12 ADHD: Special Considerations and Tools
Key takeaways for this chapter…
- Because ADHD is quite common, school psychologists should expect to encounter many students already diagnosed (as well as those who are candidates for school-based ADHD diagnosis)
- School districts have an obligation to identify students with ADHD so that they receive special education services or 504 plans
- A physician is not generally required to establish an ADHD diagnosis
- When a referral question concerns ADHD, or when ADHD emerges as a hypothesis, background information, observation, student interview and rating scales are typically warranted
- The HR approach and, for some school psychologists, use of probability nomograms are well-suited to ADHD considerations
- When school psychologists encounter a student with a condition highly comorbid with ADHD (e.g., SLD, Tourette syndrome), it can be argued that they are obliged to screen for ADHD
Cases/vignettes in this chapter include…
- Laskisha Washington and her student, Shannon Sales
Few situations provoke more bewilderment and professional anxiety than the prospect of identifying a student with ADHD. Consider the situation confronting Lakisha Washington, a school psychologist. Lakisha has been practicing in elementary schools for 10 years, and she takes her professional responsibilities with profound seriousness. She has been awarded an EdS degree, is a certified school psychologist in her home state, a member of the National Association of School Psychologists, and she holds NCSP designation. She is a consistent participant in continuing education programs and keeps abreast of practice trends by networking with school psychology colleagues.
While conducting a learning disability evaluation, Lakisha observed first-hand several symptoms of behavioral dysregulation in a fourth-grade, 10-year-old student named Shannon Sales. These included repeated problems staying seated, extreme restlessness, excessive talking, interrupting class with unsolicited comments, answering teachers’ questions without being called on, intruding into classmates’ conversations, and extreme fidgetiness. Review of school records showed mounting problems with work incompletion but only a slight dip in report card marks, most notably in reading. Inattention, only a minor problem in kindergarten and first grade, had reached troublesome levels during second grade. Reading problems intensified through third grade as did difficulty routinely completing assignments. She also lost necessary material, and she recurrently appeared disorganized and forgetful.
Broadband rating scales, which Lakisha’s school district’s policy and her own professionalism mandated as part of every learning disability evaluation, pointed toward clinical-range elevations on an ADHD index. What’s more, the scores were comparably elevated on rating scales completed by both Shannon’s teacher and her parents. When she administered individual achievement tests to Shannon, Lakisha found scant evidence of genuine academic skill deficiencies; repeated curriculum-based measures found the same thing. Not surprisingly, Lakisha had strong suspicion that much of Shannon’s school problems might arise from previously undiagnosed ADHD.
Lakisha’s school district has a crucial principle bearing on this case: “Our school psychologists classify, they do not diagnose.” All district school psychologists grasped that making an ADHD diagnosis is especially taboo. Lakisha concluded that notwithstanding the distinct prospect of ADHD that she was not free to pursue that “diagnosis.” Equally troublesome, even the chance to “classify” seemed constrained. This was because no obvious match existed with any special education category (i.e., those specified under IDEA). Simply put, Lakisha at this point doubted her ability to satisfy criteria for either SLD or ED. Although the IDEA category of Other Health Impairment (OHI) might apply to this student, Lakisha’s home state mandates a medical certification for all OHI cases. Obviously, Lakisha is not a physician (Lakisha’s school district employed no physicians to make such a diagnosis).
Like many U. S. school psychologists, Lakisha senses a way forward, albeit an imperfect one. Lakisha noted that this student received routine care from a well-regarded community pediatrician. Thus, Lakisha’s Child Study Team sought her involvement, hoping for confirmation or disconfirmation of ADHD. Indeed, Shannon’s parents had spoken highly of this pediatrician’s collaborative outlook and her vast medical knowledge, making her input seem all the more straightforward. Without ever mentioning the prospect of ADHD per se, Lakisha floated the idea of involving the pediatrician. Surprisingly, both mother and father seemed puzzled. They said, “If you’re thinking about an attention or hyperactivity problem that might be causing trouble at school, then we prefer you make a judgment yourself. We’re not interested in involving our pediatrician now.”
Before considering Lakisha’s best course of action, let’s take a look at some background facts regarding ADHD diagnosis in schools. Then we will consider several professional practice considerations as well as techniques available for those working in schools. After doing so, charting a course of action for Lakisha will make more sense.
ADHD, a Few General Considerations
According to DSM-5, the yardstick most used by mental health diagnosticians, ADHD requires noteworthy symptoms of either inattention, hyperactivity-impulsivity, or both. ADHD is present only if there is cross-situational impairment and childhood onset. If these facts are even slightly unfamiliar to you, read DSM-5 (American Psychiatric Association, 2013) pages 59 to 66 before going further in this chapter. This chapter offers a few salient points about ADHD, but these are no substitute for reading the relevant pages in DSM-5.
ADHD is a Neurodevelopmental Disorder
This means that ADHD’s overt manifestations result from presumed brain dysfunction. Although children with ADHD are quite heterogeneous, research appears to have established some basic understanding. Foremost is hypothesized dysfunction in the prefrontal cortex; frontal dysfunction appears as failure of an interrelated set of capabilities referred to as executive function. This means that youngsters with ADHD suffer from poor behavioral inhibition. Because they are poor at inhibiting impulses, youth with ADHD suffer impairments (or at least delays) in development of various executive functions: planning, regulation of emotion, working memory, recapitulation of past events, and internalization of rules. Russell Barkley (2012) in particular argues that these executive failures are the core of ADHD. According to this conceptualization of ADHD, conspicuous symptoms of impulse control and inattention are the mere outward expressions of executive dysfunction. This formulation, however, applies only to individuals with ADHD predominately hyperactive-impulsive type and ADHD combined type.
Critically, ADHD predominately inattentive type is envisioned to arise from different brain dysfunction and to appear as qualitatively different overt behavior. The behavior of children with inattentive type is characterized by daydreaming and sluggish cognitive tempo. They can be summarized as “impersistent-distractible-forgetful” (Smith, Barkley & Shapiro, 2007). Students with predominately inattentive type, however, lack the driven, intrusive, overactive, and impulsive behavior that often characterizes their ADHD hyperactive-impulsive counterparts.
The “developmental” aspect of the neurodevelopmental designation concerns onset and appearance. There is an onset (< 12 years) requirement. Moreover, ADHD is recognized in part by symptoms inconsistent with expectations for one’s age. Accordingly, age-specific standards are needed to help judge if problems controlling impulses and focusing attention are inconsistent with age and global developmental level. There is no secret that 12-year-olds pay better attention than three-year-olds. Equally important, the older group, on average, is superior at formulating plans, processing feedback, holding material in working memory, forestalling unwise actions, and slowing the expression of intense emotions. This means that standards for the behavior and focus of an older group cannot be simple-mindedly imposed on a younger group. Accordingly, everything from classroom observations to parents’ commentaries, to rating scale raw scores must be weigh against the backdrop of a child’s age.
With reflection, this is obvious. Less obvious is the context of each child’s cognitive maturity, his “mental age,” not just his “chronological age.” Twelve-year-olds with high full scale IQs, for example, on average might be expected to appear much like older classmates across a number of neurocognitive dimensions, including impulse control, planning, and sustained attention. In contrast, 12-year-olds with lower full scale IQs, on average, might be expected to appear on these same dimensions much like younger classmates. Researcher affirmed the role of IQ this way, “A moderate-to-strong relationship between general intelligence and neurocognitive functions, such as attention, working memory, and inhibition, has already been established in the literature” (Munkvold, Manger & Lundervold, 2014, p. 120). This line of thinking means that considering students’ developmental level can be important in appraising symptoms of inattention or hyperactivity-impulsivity.
Situational Demands are Particularly Relevant
Any overt behavior is partially about the student and partially about her environment (i.e., the immediate situation in which she finds herself). It’s generally recognized that all students (those with ADHD as well as those without) focus better on interesting activities (and activities they select for themselves) and worse on tedious and effort-intense tasks (and activities imposed by others). There is nothing about the definition of ADHD to make this untrue. Thus, when a parent (often a father) says, “I don’t think my son could have ADHD because he attends so well to video games” nothing has been learned to rule out ADHD. Video games are attention grabbers for many, many people. And in contrast, nearly everyone experiences some sense of boredom confronting a thick stack of uninteresting worksheets. Diagnosticians need to understand that it is perfectly plausible for a student to appear devoid of attention problems in one setting only to appear replete with problems in another setting. The former are often low-attentional-demand situations, the latter are often high-attentional-demand situations.
It is high-attentional demand situations that spread out the distribution of attentional skills. This fact matters when school psychologists observe a student in class. If you want to capture (or rule out) attention problems, watch the student in a setting where it is difficult to pay attention. Compare his success with classmates in the same demanding setting. Empirical data speak to this point. In a meta-analysis Kofler, Raiker, Sarver, Wells and Soto (2016) confirmed that core ADHD symptoms (e.g., hyperactivity) are not ubiquitous among affected children. Rather, they appear differentially depending on executive and working memory demands and level of environmental interestingness.
Expect Discrepancies among Informants
Distinctive situation-to-situation demands that impact ADHD symptomology may be why raters agree with one another only modestly. What is agreement like? Consider Kennerley and colleagues (2018) who found parent-teacher correlations for ADHD symptoms ranged for among symptoms of Hyperactivity-Impulsivity from .56 to .17. For symptoms of Inattention correlations ranged from .41 to -.03. Overall, parent-teacher ratings were more concordant for Hyperactivity-Impulsivity symptoms than for Inattention symptoms. You are cautioned against assuming that discrepant score indicate that one rater is correct and the other incorrect. Perhaps one rater was in a setting amenable to witnessing one set of symptoms whereas another rater was in a setting lending itself to seeing other symptoms. In light of considerations like these, Smith, Barkley and Shapiro (2007, p. 65) offer the following regarding ADHD diagnosis: “…we recommend that symptoms reported by one source should be tallied; then, the number of additional symptoms identified by the other source should be added to the tally, totaling the number of different items endorsed across both sources.”
A practice like this might work well in clinics where assessment is centered around DSM-5 and its symptom list. This approach, however, at least in the early stages of the assessment process, might not fit with school psychologists who prefer teacher and parent rating scales. But shouldn’t an approach like adding symptoms from several sources be considered when multiple informants’ ratings are a prime source of information? Perhaps so. Indeed, a longitudinal study of ADHD diagnosis and outcomes seems to verify empirically the wisdom of the assertion made above. Researchers (Shemmassian & Lee, 2016) examined several options for simultaneous consideration of parent ratings and teacher ratings of ADHD. Specifically, they used four algorithms for applying parent ratings of ADHD and teacher ratings of ADHD for predicting gold standard diagnoses as well as academic and social impairment. For our purposes, just two of their methods matter: one entitled the “and algorithm” and the other entitled the “or algorithm.” As you might suspect, in the former approach elevations in both parent and teacher ratings were required to judge ADHD as present. In the latter just one or the other being elevated (not both) was required. The “or algorithm” produced a better balance of sensitivity and specificity regarding accurate diagnoses, plus better prediction of future impairment. The “and algorithm” had poor sensitivity (i.e., just .25); it missed many cases of true ADHD. To be clear, trusting a high parent ADHD rating or a high teacher rating (but not requiring both) worked out best. Thus, school psychologists should be cautious about setting excessive demands for concurrent parent and teacher rating elevations when they are considering ADHD. Don’t reject ADHD as a credible working hypothesis based on a single informant’s non-clinical range score.
Comorbidities are Common
Many students with ADHD (perhaps 2/3) have one or more additional diagnosable mental health conditions. Among the most common are behavior problems (oppositional defiant disorder and conduct disorder), mood disorders, anxiety disorders, SLD, tics and Tourette syndrome (Masi & Gignac, 2015). This is consequential because youth with genuine ADHD may present with problems that seem more akin to the comorbidity than to ADHD. A moody, non-compliant or anxious student may turn out to have ADHD. There are also patterns of comorbidity based on a child’s ADHD type. For example, those students with the predominately inattentive type are more apt to experience anxiety and depressive problems; those with predominately hyperactive/impulsivity type are more apt to present with behavior problems (Barkley, 2013).
The Prospect that ADHD is Outgrown, Empirical Research
Occasionally parents and teachers believe that ADHD represents nothing more worrisome than a circumscribed inability to focus. ADHD is sometimes envisioned to be a minor problem, a trivial one in the overall scheme of factors influencing academic and life success. According to the it’s-just-minor-attention-problem line of thinking, simply tweaking demands to attend or incentivizing sustained attention should go a long way toward making the condition inconsequential (or barely consequential). What’s more, sometimes one hears that ADHD is actually beneficial. It is reasoned that impulsiveness and dysregulation promote spontaneity and risk-taking. After all, aren’t many successful entrepreneurs and scientists like this? Well, what do the data say? How do these lines of thinking hold up to logical and empiricism?
First, consider that across the school years ADHD is associated with a whole host of problems. A meta-analysis of academic functioning summarizing 74 studies found ADHD/non-ADHD differences averaged .71 standard deviations. This means that, on average, students with ADHD score approximately 10.6 standard score units (assuming a standard deviation of 15) worse than their non-ADHD counterparts (Frazier et al., 2007). Reading performance fared worst. A longitudinal study of more than 2,000 primary grade students demonstrated that ADHD symptoms were correlated with lower reading levels and less growth in both decoding and reading comprehension (Ehm et al. 2016). At the other end of age distribution, high schoolers with ADHD were found to experience lower GPAs, higher rates of course failure, and more remedial class placements that counterparts without ADHD (Kent et al., 2011). Similarly, high school students with ADHD ,compared to classmates without ADHD, have been found to be four times more likely to repeat a grade or drop out before graduation (Fried et al., 2016). Interestingly, the drop out statistic held true even after IQ scores and the presence of learning disabilities were accounted for.
Next consider adult indices. Although not all children with ADHD suffer problems that persist into adulthood, a disproportionate number do. This includes adult problems that extend to work, management of personal finances, interpersonal relationships and even responsible driving motor vehicles (Barkley & Fischer, 2011). Klein and colleagues (2012) found elevated risk of adulthood incarceration, substance abuse, and psychiatric diagnoses in the ADHD cohort diagnosed as children. Among young adult females with childhood ADHD, Babinski, Pelham, Molina, Gnagy et al. (2011) and Babinski, Pelham, Molina, Waschbusch et al. (2011) documented impairments in social, occupational, as well as some educational dimensions.
Likewise, it is illogically to assume that ADHD drives success based on scattered anecdotes. When highly successful adults reveal that they have ADHD diagnoses, it’s tempting to surmise that one thing (ADHD) caused the other (success). But if one stops and thinks about it, success is caused by a complex interaction among things like cognitive ability, opportunity, personal drive, social connections plus a big dose of luck. Perhaps for the high achiever with ADHD, the ADHD did not matter—it was insufficient to constrain them. Perhaps their ADHD even sometimes helped a little, such as by promoting risk taking. But it’s a huge logical stretch to assert that ADHD is an advantage. If you want to know if ADHD actually proves advantageous, look at success among 1,000 adults with ADHD and 1,000 without, not anecdotes.
Big Picture
Why does this matter? Although all diagnoses, including, ADHD can be stigmatizing, they can also prove empowering. Understanding, predicting and controlling (improving) behavior is psychology’s prime goal, as argued for since before WW II (Allport, 1940). But too often school psychologists assess a student with obvious symptoms of ADHD and turn a blind eye to the entire notion of ADHD. Either explicitly or implicitly many seem to view ADHD as beyond the borders of their school-based practice. Their boss may have told them not to do so (as likely happened with Lakisha).
How can stakeholders, however, understand the action of an affected student without recognition of his ADHD? How does one make sense of unwanted comments blurted out in class? How does one understand misplaced or forgotten homework assignments? How does one anticipate everyday problems that are likely to appear during adolescence and adulthood of a child who clearly has ADHD but has never been notified of this critical (potentially life altering) fact? Most importantly, how does a school psychologist assure proper educational services (including those entitled under Section 504) and access to empirically documented treatments specific to ADHD? In other words, how can science (and the nomothetic approach) be brought to bear to help this human? The answer is that they cannot. In turn, how does a school psychologist fully satisfy her ethical obligations? The NASP Principles of Professional Ethics says, school psychologists are to “improve the quality of life for students” (NASP, 2020, p. 39). Specifically, how does one meet the obligation to foster improvement when valuable understanding never occurs and worthwhile treatment options are never considered because of a refusal to address the obvious?
Practice Considerations
Who Should Identify ADHD?
Just who is empowered to make ADHD identifications? That question remains largely unresolved and subject to local professional culture, state level regulations, and historical precedent. Nonetheless, various national organizations have spoken. For example, the National Institute of Health (NIH), which is obviously a physician-centric institution, simply calls for a licensed professional. They do not stipulate that the professional is a physician. It is interesting that NIH lists the following as able to establish ADHD’s presence:
- physicians
- psychologists
- social workers
so long as they possess experience in ADHD. For those who might think that pediatricians are especially skilled ADHD diagnosticians, consider this additional quotation from NIH (2015)….. “some pediatricians can assess the child themselves, but many will refer the family to a mental health specialist with experience in childhood brain disorders such as ADHD.” These statements seem compatible with those of the U. S. Centers for Disease Control: “The diagnosis can be made by a mental health professional, like a psychologist or psychiatrist, or by a primary care provider, like a pediatrician.”
https://www.cdc.gov/ncbddd/adhd/diagnosis.html
Critically, because licensing is handled at the state level, local laws may determine who has the right to make ADHD diagnoses. That said, research suggests that school personnel assign ADHD diagnosis in a scant 2.8% of cases. In contrast, more than 50% of the time diagnoses are assigned by primary care physicians, such as pediatricians or family practitioners (see Table 12.1). Non-school-based psychologists (e.g., licensed psychologists in private offices) make diagnostic school-age ADHD determinations in about one in seven cases (see Visser et al., 2015 for more details www.cdc.gov/nchs/data/nhsr/nhsr081.pdf). No wonder Lakisha seems hesitant.
But what about the actual skills school psychologists (like Lakisha) compared to those of typical pediatricians. Based in schools, school psychologists can interview teachers, access facts spanning years of a student’s life (cumulative files), and observe first-hand a student’s actions, including her ability to focus and exert self-control in a setting with ecological validity. None of this is possible in the sterile confines of a physician’s office. A second advantage enjoyed by school psychologists is their relative distribution of time. School psychologists routinely spend large blocks of time assessing children (e.g., several entire mornings on a single student). In contrast, many pediatricians see patients for 15-minute time blocks, with additional time allocated for complex cases or those requiring extended data collection (e.g., ADHD diagnosis). It is implausible that a typical pediatrician would invest as much time in an ADHD evaluation as a typical school psychologist. A third advantage is the focus of professional training. School psychologists earning EdS degrees, for example, spend their entire 3 years of post-graduate training on development, measurement, assessment, behavioral observation, and interviewing. Pediatricians spend a decided minority of their training time on this suite of tasks. It makes sense. Pediatricians primarily must learn about life-threatening illnesses that must never escape detection as well as routine (high-frequency, low-intensity) presenting problems that consume most of their office time. These include ear infections, colds, rashes, diarrhea, immunizations and, occasionally, questions about ADHD. Thus, it’s no surprise when the NASP position statement on Students with Attention Deficit Hyperactivity Disorder says the following, “School psychologists are trained in child psychopathology and behavior assessment practices and have the expertise to evaluate whether students are presenting with ADHD” (NASP, 2018, p. 1).
Are School Psychologists Obliged to Identify ADHD?
To echo Chapter 10, Local Educational Agencies (LEAs), have an obligation to identify all kinds of conditions. The obligation, however, rests with the LEA, not necessarily any one of its specific employees (including a school psychologist). Among the elements of the Individuals with Disabilities Education Act (IDEA) is a provision entitled “Child Find,” which charges all LEA’s to find students with disabilities. Regarding IDEA, this means any of the listed disabilities (e.g., SLD, ED). When this act’s earlier version was up for reauthorization, Congress contemplated changes and thus solicited public input. Many parents and professionals proposed that ADHD be added to the IDEA list of handicapping conditions as a stand-alone category (i.e., assume the same status as SLD or Hearing Impairment). In fact, in 1990 Congress seriously considered adding ADHD to its list of “children with disabilities” (see Wodrich & Davila, 1994, for details). The position of the U.S. Department of Education dating from at least 1992 made it clear that students with ADHD needed to be identified because many of them meet criteria for SLD or ED. Once OHI was added to the listed disabilities in 1993, LEAs’ obligation to identify student with ADHD became even more compelling. This was because ADHD became listed as a candidate medical condition (just like asthma or diabetes) for OHI designation. Indeed, the limited empirical evidence that exists suggests that the OHI category is the common special education route to create eligibility for students with ADHD. Critically, however, IDEA’s Child Find proviso is mandatory, not optional. Furthermore, it is broadly inclusive, not exclusive. Vis-à-vis ADHD, school districts have an “affirmative obligation” to evaluate any child suspected of having LD, ED, or OHI. There is no opt out, such as when a student with ADHD might be eligible under OHI but a school district declines to evaluate her for eligibility. Think back to Lakisha’s situation.
Obviously, issues of entitlement (e.g., under IDEA and Section 504) arise when ADHD is considered at school. School psychologists are quite capable of shouldering the LEAs obligation to identify students with ADHD. But individual school psychologists will need to consider their own professional preparation, their personal skill set, authority (based on state-level rules and regulations), and finally directives from their supervisors. They also should know some facts, including the current position of the federal government regarding ADHD in schools (see Illustration 12.1).
Illustration 12.1 Important Points for the United States Office of Civil Rights.
Much was clarified by a detailed “Dear Colleague Letter” of July 2016. It was authored by Catherine Llamon, United States Assistant Secretary for Civil Rights. This letter made several important points, including highlighting a second (non-special education) obligation for LEAs—the civil rights protection afforded under Section 504 of the American’s with Disabilities Act (which you read about in Chapter 10). It’s easy for school psychologists to grasp the Section 504-ADHD link; many students with ADHD suffer impaired classroom functioning and warrant reasonable accommodations. In reality, however, Section 504 plans seem to be common when parents make a fuss and bring their own documenting paperwork (such as from the student’s pediatrician). Otherwise, they are rare. Critically, The Dear Colleague letter warned LEAs that the obligation to identify and document did not reside with parents, it resides with them. Some of the key points of that letter follow www.ed.gov/about/offices/list/ocr/letters/colleague-2016:
- “A school district must evaluate students who are suspected of having any kind of disability in all specific or all related areas of education need, even if the students do not fit into one suspected disability category or into multiple disability categories.” (p. 18).
- “It is the district’s obligation to evaluate; it cannot shift the burden of that cost or obligation onto the parent” (p. 19). This is relevant to Lakisha’s situation because the time, effort, and expense for OHI evaluations cannot be transferred to parents any more than they could for SLD evaluations.
- … “concentrating”…..is [itself] a major life activity (as required for Section 504 designation). In other words, once the presence of ADHD is established so too has the 504 requirement for impact on a major life activity.
- “OCR will presume, unless there is evidence to the contrary, that a student with a diagnosis of ADHD is substantially limited in one or more major life activities.” (p. 10)
- “Mitigating measures [medications, coping strategies] shall not be considered in determining whether an individual has a disability.” (p. 5)
- The requirement to provide a Free Appropriate Public Education (FAPE) can be met by an IEP (just like for students with any type of special education designation). However, other methods for outlining how a “free appropriate public education” (FAPE) are permitted. To be clear, even as far back as the 1990s, the U.S. Office of Special Education and Rehabilitation indicated that Section 504 accommodations can take the form of regular or special education services (also see Wodrich & Davila, 1994).
- Plans for students with ADHD often fail at the implementation level, which violates the FAPE requirement. Thus, it is not satisfactory to merely create an accommodation plan and then file it away.
- “If the district suspects that a student has a disability….it would be a violation of Section 504 to delay the evaluation in order to first implement an intervention….” (p. 17).
- “It is important that school districts appropriately train their teachers and staff to identify academic and behavioral challenges that may be due to a disability so a student is referred for an evaluation under Section 504” (p. 17).
- “There is nothing in Section 504 that requires a medical assessment as a precondition to the district’s determination that the student has a disability” (p. 23).
ADHD Practice Guidelines and Lakisha’s Situation
How should children with ADHD be identified? Which assessment techniques should be followed? How should things actually be put together? Fortunately, at least two sets of practice guidelines exist–one from the American Academy of Pediatrics and the other from NASP. Both are presented in Table 11.1 The two are surprisingly similar, both advocating use of multiple procedures, consideration of comorbidity, and application (or at least cognizance of) DSM-5 criteria. The school psychology guidelines also correspond closely to the elements you read about in Chapter 2. If we assume that Lakisha decides to conduct an assessment for ADHD, then she could access the standard school psychologist tool kit and simply enter the resulting information in a HR Worksheet. Many times, however, school psychologists start with no clear working hypothesis (ADHD or anything else). Instead, as you have seen, they adopt an opened-end process like the HR approach. But specialized, narrowed consideration of ADHD is a different matter.
When ADHD becomes a prime hypothesis, school psychologists are obviously dealing with a quite specific classification decision (and this process begs for a strongly nomothetic mindset). Thus, ADHD-focused assessments are ripe for possible use of probability nomograms for school psychologist who like to see the probability of ADHD emerge as they work. Consider the position advocated one-quarter of a century ago in an article published in the Journal of School Psychology. “The probability of obtaining a correct diagnosis of children in the general population with behavioral rating scales is affected by two factors: (a) the sensitivity and specificity of the instrument, and (b) the prevalence of ADHD in the population” (Reid & Maag, 1994, p. 342). These authors argue for exactly the Bayesian (probability nomogram) approach to harness the various assessment tools proposed in the NASP practice guidelines. Even if you never use a probability nomogram, knowing how well various tools work to identify ADHD is invaluable. That’s what we turn to now. At the end of the chapter, we revisit Lakisha and her case.
Table 12.1 Guidance from Professional Organizations |
|
American Academy of Pediatrics (2011) |
National Association of School Psychologists (2018) |
The primary care physician should:
|
Generally, includes multi-informant, multi-setting procedures and recommends the following:
|
Base Rate Considerations
The ultimate probability of ADHD hinges to a surprising extent on the base rate of ADHD before any assessment information is even collected. This was detailed in Chapter 2. A few supernaturally motivated school psychologists might collect their own base rate information. It is interesting (and potentially instructive) that in some settings the verified ADHD base rate is quite high. For example, back in 1990, the base rate was a mammoth .65 for children referred to a clinic for attention problems, high activity levels, or poor school grades (Aylward, Verhulst & Bell, 1990). And, at least one expert (Russell Barkley) has speculated about base rates where you practice, “About half of the children referred to child mental health settings have ADHD. We expect similarly high base rates of ADHD are found in populations referred to school psychologists and similar professionals” (Smith, Barkley & Shapiro, 2007, p. 68, underlining added). This value may seem high to some readers. A more conservative estimate might be between the population prevalence rate (.06) and the estimate by the authors above (.50), perhaps .25 as a rough base rate estimate. Again, collecting your own information would be the best if you find yourself dealing recurrently with questions about ADHD. That said, guessing at base rate is arguably better than completely ignoring the topic.
Information from Background
ADHD is heritable. Logically then, having a biological relative with ADHD raises a student’s own risk of ADHD. Consider family history and its ability to provide an estimate of a risk, expressed as a ratio (see Chapter 2 for details on calculating risk). When a sibling is affected with ADHD, the risk goes up dramatically. For example, in one study (Chen et al. 2008) the risk ratio was found to be 9.0 (12.7/1.4) when researchers used stringent diagnostic criteria concerning ADHD combined type. In another study (Faraone et al., 2008), the ratio was found to be 4.0 (20.0/5.0) when a broad definition of ADHD was applied. As you would anticipate, either of these values entered into the middle column of a nomogram would boost the pre-test probability upward.
Risk of ADHD can be gleaned from other background information. Sometimes, for example, school psychologists encounter a student with a prior diagnosis known to predict (raise the risk) of ADHD. In actual practice, diagnoses of comorbid conditions may appear in the paperwork submitted by parents, such as in an evaluation conducted by an outside professional. This is exemplified by the presence of Tourette syndrome, which denotes an estimated risk ratio for ADHD of 5.3. You saw this information already in Chapter 4 (Figure 4.12). Similar findings appear regarding children with diagnosed language disorders. The risk of having ADHD is boosted for children with language disorders about five times based on two studies, both of which are now dated (Beitchman et al., 1986; Reilly, Cunningham, Richards, Elbard & Mahoney, 1999). Like risk ratios associated with family history, information about conditions that predict a risk of co-existing ADHD can be entered into the middle column of an initial nomogram (or permit completion of a second, third, or subsequent nomogram to track changing ADHD probabilities). You are urged to consider these factors whether or not you use nomograms.
Broadband Rating scales and ADHD
You are already generally familiar with broadband rating scales from Chapter 5. But regarding ADHD, it is their potential to accurately classify that counts most. As you have seen, diagnostic likelihood ratios (DLRs) can be used to help classify, as well as to provide a sense about which tools work best (have the largest effect size for proper ADHD identification). DLRs are sometimes found (conveniently) in the test manual or (inconveniently) in post-publication research reports (i.e., published journal articles).
Behavior Assessment System for Children-Third Edition (BASC-3)
What about the extremely popular BASC-3 and ADHD identification? Unfortunately, DLRs do not appear in the manual (and there is no obvious report of sensitivity and specificity to permit a school psychologist to make her own calculations of DLRs). You saw this already in Chapter 5, when general features of the BASC-3 were reviewed. What’s more, recent research that does report ADHD-related sensitivity and specificity values may not have direct practice applications because of the complex algorithm the researchers used to classify participants (Zhou, Reynolds, Zhu, Kamphaus & Zhang, 2018). Nonetheless, school psychologists will likely consider both the BASC-3’s Attention Problems and Hyperactivity scales, even if they are unsure of their diagnostic utility. In addition, the BASC-3 offers an ADHD Probability Index, which you learned in Chapter 5 to represent a typical specialized index (comprised of highly discriminating items). The ADHD Probability Index is reported in T-scores; there is no obvious method of generating a “probability” from this so-called probability index. And, as you saw in Chapter 5, the manual fails to provide evidence of statistically significant differences between clinical and control groups on its scales concerning ADHD diagnosis (i.e., Attention Problems, Hyperactivity, ADHD Probability Index). Consequently, BASC-3 scores related to the prospect of ADHD will often be considered by school psychologists (after all this is by far the most popular broadband behavior rating scale; Benson et al., 2019). Caution, however, seems in order given the scant empirical information substantiating their classification validity.
Conner Comprehensive Behavior Rating Scale (CBRS)
The CBRS manual offers its readers the most helpful information regarding classification–sensitivity and specificity. This information comes from a study of three clinical groups of youth: those with ADHD Combined, those with ADHD Hyperactive-Impulsive type only, and those with ADHD Inattentive type only when they are compared with unaffected youth. These three groups (compared to controls) served to examine the accuracy of two predictors that school psychologists might use in the field: (1.) the Conner CBRS DSM-IV-TR symptom scale for ADHD Inattentive and (2.) the Conner CBRS DSM-IV-TR symptom scale for ADHD Hyperactive-Impulsive. The results are found in Table 12.2. This implies that diagnosticians could plug the values found in Table 12.2 directly into one or more nomograms to gain updated probability estimates that their student might have ADHD. For example, if both a teacher’s and a parent’s rating were elevated on the Conners CBRS DSM-IV-TR Hyperactive-Impulsive scale, then two DLRs could be applied to the middle columns of nomograms. Let’s assume that the diagnosis being contemplated was Combined type ADHD. In this scenario with the two elevations just mentioned, a first nomogram could be calculated with the teacher-associated DLR of 3.83. Then another nomogram could be calculated with the parent-associated DLR of 4.11. The pre-test probability, hence, would have received two upward nudges based on Conners CBRS scores. This all seems great. But, as seen earlier with the Conners CBRS, the manual is missing some critical information. That is, the Conners manual fails to share the exact cut-score used to produce the values found in Table 12.2. Consequently, one would be forced to guess about where to draw the line. Presumably, any DSM-IV-TR ADHD symptoms scores above T-score > 70 should be safe regarding application of the DLRspositive from Table 12.3, but one cannot be entirely certain. Prior comments notwithstanding, note that these are small effect sizes.
Table 12.2 Effect Sizes and DLR Values for the Conners Comprehensive Behavior Rating Scale concerning ADHD |
||||
Source of Rating | Effect size associated with positive score | Diagnostic Likelihood Ratio Positive | Effect size associated with negative score | Diagnostic Likelihood Ratio Negative |
Teacher | ||||
Conners Inattentive (combined) | Small | 3.65 | Small | .24 |
Conners Hyperactive-Impulsive (combined) | Small | 3.83 | Small | .38 |
Conners Inattentive (specialized)◊ | Small | 3.04 | Small | .42 |
Conners Hyperactive-Impulsive (specialized)∗ | Small | 2.94 | Minimal | .57 |
Parent | ||||
Conners Inattentive (combined) | Small | 4.81 | Small | .31 |
Conners Hyperactive-Impulsive (combined) | Small | 4.11 | Small | . 32 |
Conners Inattentive (specialized)◊ | Small | 3.35 | Small | .41 |
Conners Hyperactive-Impulsive (specialized)∗ | Small | 3.00 | Minimal | .53 |
Source: Conners CBRS manual (Conners, 2010). Combined comparisons concern all clinical groups without disaggregation. Specialized comparisons concern disaggregated data for subgroups as follows: ◊Comparison with ADHD Inattentive group only; ∗Comparison with ADHD Hyperactive-Impulsive group only |
ASEBA Child Behavior Checklist (CBLC) and Teacher’s Report Form.
As you saw in Chapter 5, the manual (Achenbach & Rescorla 2001) for the CBCL and TRF include diagnostic utility statistics that are already calculated and ready for your use. These are reported as odds ratios (rather than DLRs) as seen in Table 12.3. Critically, however, the values concern predicting membership in a generic clinical sample (of referred youth) compared to a non-clinical sample (non-referred youth). The values do not concern children who met criteria for a gold standard diagnosis of ADHD. They should, thus, be taken with a grain of salt.
Table 12.3 Odds ratios for ASEBA CBLC and TRF |
||
Child Behavior Checklist (CBCL) |
Teacher’s Report Form (TRF) |
|
Empirically-based scales | ||
Attention Problems | 12 | 7 |
(Inattention) | N/A | 5 |
(Hyperactivity-Impulsivity) | N/A | 5 |
DSM-oriented scales | ||
ADH problems | 10 | 6 |
(Inattention) | N/A | 6 |
(Hyperactivity-Impulsivity) | N/A | 5 |
Source: Test manual (Achenbach & Rescorla, 2001) |
Fortunately, potentially more user-friendly post-publication diagnostic utility data also exist for the CBCL. As seen in Table 12.4. Raiker and colleagues (2017) studied the classification success of the CBCL’s “Attention Problems” scores. They discovered that Attention Problems was indeed most effective (i.e., compared to other scores, such as an Externalizing composite) in classifying youth as having or not having ADHD. This was true for both older and younger children. In fact, this article’s user-friendly aspects include depictions of nomograms with ratios applied to hypothetical cases; it also includes DLRs? and demonstrates their use. As you can see from Table 12.4, high cut-score values give progressively more favorable ratios. It is likely that cut-scores of T-score = 64 or T-score = 70 would be selected by some diagnosticians for most ADHD-related decisions.
Table 12.4 Effect Sizes and DLR Values for the CBLC Concerning ADHD |
||||
Source of Rating | Effect size associated with positive score | Diagnostic Likelihood Ratio Positive | Effect size associated with negative score (T< 64) | Diagnostic Likelihood Ratio Negative |
Teacher with younger group (5 -11 years) | ||||
T > 64 | Minimal | 1.06 | Minimal | .88 |
T > 70 | Minimal | 1.31 | Minimal | .88 |
Teacher with older group (12-18 years) | ||||
T > 64 | Minimal | 1.21 | Minimal | .73 |
T > 70 | Minimal | 1.67 | Minimal | .73 |
Parent with younger group (5 – 11 years) | ||||
T > 64 | Minimal | 1.86 | Small | .23 |
T > 70 | Minimal | 1.97 | Small | .23 |
Parent with older group (12 – 18 years) | ||||
T > 64 | Minimal | 1.41 | Small | .34 |
T > 70 | Small | 2.22 | Small | .34 |
Source: Raiker et al., 2017 |
Narrowband Rating Scales and ADHD
Narrowband scales might be used in the presence of an ADHD-working hypothesis or when an explicit ADHD referral question has been present from the start of a case. Just two examples of narrowband scales are provided here, one concerning ADHD expressly and another the related construct of executive function.
ADHD-5 Rating Scale
You already heard about the ADHD-5 Rating Scale (DuPaul, Power, Anastopoulos & Reid, 2016) in Chapter 2. It is a teacher-completed and/or parent-completed 18-item rating scale, which concerns ADHD only. More specifically, 9 items on the ADHD-5 Rating Scale’s Inattention dimension match DSM-5’s 9 symptoms for ADHD predominately inattentive presentation. In parallel, 9 additional items correspond precisely to the 9 symptoms of DSM-5’s ADHD predominately hyperactive/impulsive presentation. Thus, the ADHD-5 Rating scale enjoys exquisite content validity. To aid practitioners, optimum cut-scores are available when both teacher- and parent-completed forms are used singularly or in concert. The latter, for example, might involve the simultaneous occurrence of teacher ratings > 85th percentile and parent ratings > 90th percentile. When these cut scores are used, for example, a DLR of 9.6positive can be determined. This DLR can be entered into the center column of a nomogram or simply recognized to reflect a moderate effect size in helping to confirm ADHD’s presence. By the way, similar information about sensitivity and specificity is provided in the manual for many other cut-scores.
Behavior Rating Inventory of Executive Function-2nd edition (BRIEF-2)
Barkley’s theory places executive function (EF) front and center in ADHD conceptualizations. Thus, even if psychometric measures of EF, which you will see below, are skipped, might not convenient narrowband EF rating scale options work? This is intuitively appealing. Indeed, the BRIEF-2 (Gioia, Isquith, Guy & Kenworthy, 2015) has now become popular, with approximately half of school psychologists sometimes using it (Benson et al., 2019). The BRIEF-2 is for children 5-18 years, with its 86 items producing a Global Executive Composite and two other tiers of scores (see Table 12.5). There are also validity scales. Parent, teacher, and self-rating options exist. This popularity, however, is not solely for its potential role regarding ADHD. Executive dysfunction such as documented on the BRIEF-2 might co-occur with problems as diverse as intellectual disability (Shishido, Mahone & Jaconson, 2020), autism (Freeman, Locke, Rotheram-Fuller & Mandell, 2017), adolescent substance abuse (Clark et al., 2017) or might stand alone as a source of problems not fitting any condition appearing in DSM-5 (e.g., Baron, 2016). Regarding ADHD, the updated BRIEF-2 includes base rate tables, sensitivity and specificity values, likelihood ratios, as well as PPV, and NPV.
Table 12.5 BRIEF-2 Organization of Scales |
|
Global Executive Composite
|
|
↓Additional finer-grained scores appear below↓ | |
Behavior Regulation Index
|
|
Sub-index name |
Potential informants for that index |
Inhibit | Parent, teacher, self |
Self-monitor | Parent, teacher, self |
Emotional Regulation Index
|
|
Shift | Parent, teacher, self |
Emotional control | Parent, teacher, self |
Cognitive Regulation Index
|
|
Initiate | Parent, teacher |
Task completion | Self |
Working memory | Parent, teacher, self |
Plan/organize | Parent, teacher, self |
Task monitor | Parent, teacher |
Organization of materials | Parent, teacher |
Three published studies provide data to help determine how well the BRIEF-2 (or the original BRIEF) assists in ADHD classifications. In the first study, Reddy, Hale and Brodzinsky (2011) found that two indices, Behavioral Regulation and Metacognition from the parent version of the BRIEF demonstrated high rates of correct classification (82%). Theoretically inhibition is most important and indeed Behavioral Regulation had the best discriminant function scores. In a second study, conducted in Switzerland, Drechsler, Logoz, Walitza and Steinhauser (2018) found only the BRIEF Inhibit scale was statistically different between the two groups. What’s more the effect size was small (it accounted for 8 percent of the variance). No diagnostic utility statistics were provided. A third much larger U.S. study (Jacobson, Pritchard, Koriakin, Jones & Mahone, 2020) confirmed evidence of adequate specificity but poorer sensitivity. Among children with ADHD whose presentation included hyperactivity/impulsivity, BRIEF-2 Behavioral Regulation and Emotional Regulation scores were most elevated. But when inattention was present, BRIEF-2 Cognitive Regulation scores were apt to be elevated.
These studies are interesting, but they fail to prove incremental validity. In other words, might BRIEF-2 scores measure much the same characteristics already captured by more straightforward ADHD measures (e.g., ADHD-5 Rating Scale), thus constraining much additional information coming from BRIEF-2 scores? Some recent texts (e.g., Frick, Barry & Kamphaus, 2020) omit use of EF rating scales such as the BRIEF-2 from their procedures involved in ADHD assessment. Readers would necessarily need to spend time with the BRIEF-2 manual before applying procedures related to these statistics in their practice while recognizing EF dysfunction and ADHD are related but distinct.
Continuous Performance Tests
In part because the diagnosis of ADHD is often contentious, finding a credible child-completed test may circumvent stakeholders’ concern about the subjectivity of rating scales. “Continuous performance tests,” or CPTs might fit the bill. The idea is simple. Ask an individual to pay attention to something boring and have him respond over several minutes. Usually, this involves watching (or listening) to a steady stream of stimuli and responding when a target stimulus appears. Scores reflect things like the youngster’s accuracy and speed of responding. Complementary variables, such as the rate of false alarm responses, might also be calculated. There are several commercially available CPT options. Although school psychologists may not themselves use CPTs, school-based practitioners are nonetheless likely to see CPT scores brought by parents.
Gordon Diagnostic System (GDS)
One of the earliest popular CPTs is the Gordon Diagnostic System, which is a clutter-free, indestructible, special-purpose device (not mere software to be added to a general-purpose computer). The front of the GDS is depicted in the following link http://www.gsi-add.com/gordondiagnosticsystem.htm. There are several available tasks, one option is the Vigilance (detecting and responding to a target stimulus across a 9-minute trial), which the target and distractor numbers appearing one a small screen. Another GDS task is entitled Delay (self-paced responding with feedback across an 8-minute trial). Some limited evidence that GDS scores might assist in diagnostic decisions appeared in the 1990s. For example, a study by Barkley and Grodzinsky (1994) that positive scores might be valuable, but there are problems with application because the high base rate of ADHD used. A larger Canadian study (Reilly, Cunningham, Richards, Elbard & Mahoney, 1999) failed to offer much support for use of the GDS. Even with many analyses tapping all of the GDS tasks and assigning various cut-scores, only modest DLRs appeared (no DLRpositive better than 1.73; no DLRnegative better than .41).
Tests of Variables of Attention (TOVA)
The TOVA, another CPT, works with an existing computer https://www.tovatest.com. It offers a procedure much like the GDS Vigilance task seen above—providing indices of commission errors, as well as speed and consistency of responding. An Israeli study suggests quite limited diagnostic utility. Working in an ADHD clinic, researchers found good sensitivity but poor specificity (Zelnik, Bennett-Black, Miari, Goez, & Fattal-Valevski, 2012). Using this data, positive DLRpositive of 1.16 (minimal effect size) can be calculated. But if entered into the center column of a nomogram, for example, would do little to change the prior probability of ADHD.
Conners Continuous Performance Test-Third Edition (CPT-3)
Like the GDS and TOVA, the CPT-3 provides an attention task delivered via a computer. Although there is also an auditory version (Conners Auditory Test of Attention, CATA), stimuli appearing visually in the standard CPT-3 seem to garner greater use. The CPT-3 is suited to children 8 years and older. It provides 16 scores subsumed under the following four broad dimensions: Inattentiveness, Impulsivity, Sustained Attention and Vigilance. The Pearson website, where the CPT-3 is marketed even though it was devised by MHS, a heading entitled Discriminative Validity offers effect size statistics rather than diagnostic utility statistics. Specifically, differences between children with and without ADHD resulted in Cohen’s ds from .10 to .49., and these were found to be statistically significant. https://www.pearsonclinical.com.au/products/view/548 [retrieved September 25, 2020]. You already saw in Chapter 2, however, the limitation of information like this (i.e., statistical significance, effect size indicators) concerning classification validity. There was also a section entitled Incremental Validity on the same website, where there is information indicating that the CPT-3 provides a slight boost to the rate of correct classification compared to merely using the Conners rating scale alone.
Cumulatively (i.e., across GDS, TOVA, and CPT-3) these findings may help explain less-than-universal enthusiasm for including EF and ADHD identification. Consider the following, “Thus, although there is considerable theoretical support for deficits in EF accounting for much of the impairment in individuals with ADHD, there are serious limitations regarding its definition and measurement that limit the potential clinical utility of EF” (Owens, Evans & Margherio, 2020, p. 115).
Other Cognitive Measures (WISC-5 and CAS)
Maybe familiar tools could hint at ADHD’s presence. On the WISC-5s, for example, it appears that some children struggle to hold material in working memory on the Digit Span or Letter-Number sequence subtests, or that they fail to work persistently on subtests like Coding or Symbol Search. Might not relatively low scores on attention-laden tasks like these suggest ADHD’s presence? And, in fact, children with ADHD have been shown to score low on certain subtests. One study, for example found tantalizing patterns—23% of children with ADHD had an intra-individual profile characterized by low scores on Digit Span and Arithmetic, whereas this was true for no children without ADHD in the same study (Mayes, Calhoun & Crowell, 1998). But with a much closer look, Devena and Watkins (2012), seemed to dash this prospect. When comparisons were made between scores on predominately cognitive WISC-IV subtests (i.e., the General Ability Index, GAI) with those on subtests requiring relatively more focus and persistence (i.e., the Cognitive Proficiency Index., [CPI] comprising Working Memory and Processing Speed subtests) the latter proved lower for across samples of children with ADHD (as well as other diagnoses) relative to representative (non-clinical) peers. But when it came to classification, the rate of true positives and true negatives meant that there was only modest evidence to support classification validity. This prompted the following: “Thus, using the GAI-CPI cognitive profile to distinguish children with ADHD is less accurate than the methods already used by clinicians and considered best practice for identifying children with ADHD” (Devena & Watkins, 2012, p. 150).
The Cognitive Assessment System (Naglieri & Das, 1997) contains two factor scores that would seem to be largely unrelated to ADHD (Successive and Simultaneous Processing) as well as two factors potentially linked to ADHD (Attention and Planning). As one might intuit, the former turned out to be less impacted by ADHD than the latter in clinical vs. control studies (Naglieri, Salter & Edwards, 2004). And as you saw in Chapter 2, credible diagnostic utility statistics can sometimes appear when tools like the CAS are used to compare children with and without ADHD (e.g., DLRpositive = 3.20, DLRnegative = .27; Canivez & Gaboury, 2016). As impressive as these demonstrations of classification validity appear to be, it is unlikely that tools like the CAS have an applied role in ADHD diagnosis (e.g., administering an entire CAS [or CAS-2] to get scores for ADHD seems inefficient).
Fixed Test Batteries for ADHD Diagnosis
Sometimes a complete test battery is configured expressly to assist diagnosticians interested in ADHD diagnosis. The batteries may comprise various components that are designed to complement each other. One is mentioned here because it has appeared in the school psychology literature.
One battery is the Pediatric Attention Disorders Diagnostic Screener (PADDs), which is described by Keiser and Reddy (2013). The PADDS includes the following elements together:
- Parent and teacher behavior rating scales
- Diagnostic interview of parent
- Computer-based measures of executive functioning
- Nomogram-based analysis
These four elements seem to require about 45 minutes of professionals’ time. Critically for practitioners, it appears that available software for the PADDS calculates DLRs and then applies these values to help make judgments about individual children (age 6-12 years). Clinical use of the PADDS is demonstrated in the school psychology literature (i.e., the Journal of Applied School Psychology), including nomograms calculated for individual students.
Maybe in the future, practice will become more specialized. School psychologists, for example, might handle almost all tasks at their site, but specialized questions like ADHD (or autism, or traumatic brain injury) might be triaged to a more specialized colleague. In this scenario, more finely tailored procedures, like the PADDS might be welcomed. Without district-wide redistribution of services, however, the specialized nature of ADHD-specific assessments might prove fraught with practical difficulty, legally/administratively perilously and time consuming. We will return to Lakisha Washington’s case a bit latter, but implications for her may be already obvious.
Documenting Impairment
Symptoms are just part of the picture. Many of the tools you have seen so far can confirm ADHD symptomology, but that is not the same as confirming ADHD as a “disorder.” In fact, DSM-5 mandates evidence of impairment, and so does sound clinical practice. As one professor put it to her students, “It’s a free country. Everyone is permitted to express whatever symptoms they choose until (unless) they start to represent a problem to others or a problem to themselves. Only then does a diagnosis become a real possibility.” As you might suspect, researchers have documented that some students experience clinical level symptoms (on either the inattention or the hyperactivity/impulsivity dimension) without accompanying evidence of impairment (Power, Watkins, Anastopoulous, Reid, Lambert & DuPaul, 2017).
In light of findings like this, how might one assure that an impairment is actually present? One possibility is to use the ADHD-5 Rating scale. It provides items to address six dimensions of impairment. These are:
- family/teacher relationships
- peer relationships
- academics
- behavior problems
- homework
- self-esteem
In other words, the ADHD-5 Rating Scale assesses symptoms (including the ability to calculate DLRs based on symptomology), as you already saw. Plus, it can provide multiple indicators of impaired functioning. Of course, many school psychologists are already comfortable assessing impairment via broadband scales like the adaptive scales on the BASC-3. Or, they may simply interview parents or teachers. Impairment determinations using the ADHD-5 Rating Scale have an advantage, however. These determinations relate to ADHD symptoms, a feature lacking in other non-specialized impairment scales found in the BASC-3 and Conners CBRS.
Words You Might Barrow
Among the great challenges of school psychology practice is how to write about one’s conclusions clearly, accurately, and concisely. Generically, this topic gets its own entire chapter (Chapter 13). But ADHD is something of a special challenge. Table 11.2 offers some hints that you might wish to consider. As with all forms of professional communication, you are encouraged to listen to the words and examine the writing of others.
Table 12.6 Wording Help for Positive (affirming) ADHD Conclusion |
||
Test only was used (not recommended) |
HR approach was used |
Probability nomograms were used |
“Chad appears to have ADHD. This was concluded because he has high rating scale scores from teacher and parents.” | “Chad appears to have ADHD. This was concluded by looking at many sources of information (background facts, observation, general rating scales and scales developed expressly to measure ADHD).” | “Chad appears to have ADHD. This was concluded by carefully weighing (numerically) several sources of information. These included background facts, observation, general rating scales and scales developed expressly to measure ADHD. In part, Chad’s extreme probability index leads his team to this conclusion.” |
Optional addition if DSM-5 was used |
Optional addition if DSM-5 was used |
Optional addition if DSM-5 was used |
“We also examined the most-recognized criteria for ADHD, entitled DSM-5. A point-by-point analysis suggests that Chad satisfies these criteria.” | “We also examined the most-recognized criteria for ADHD, entitled DSM-5. A point-by-point analysis suggests that Chad satisfies these criteria.” | “We also examined the most-recognized criteria for ADHD, entitled DSM-5. A point-by-point analysis suggests that Chad satisfies these criteria.” |
Case: A Straightforward ADHD Referral question—Shannon Sales (Lakisha’s case)
Lakisha has become persuaded to conduct an ADHD evaluation for Shannon. After securing necessary parental consent and childhood assent, she starts with background information. This helps her eliminate some possibilities right from the outset. Shannon’s records suggest that she is a healthy girl free of any chronic medical problems that might mimic ADHD. Furthermore, she has passed hearing and vision screenings. Shannon lives in an intact family where both parents work and where there are ample financial resources. Parents deny stresses in the home that might contribute to their daughter’s behavior.
Other information begins to point toward ADHD as one reasonable hypothesis. Parents concur with Shannon’s classroom teacher regarding some aspects of the girl’s behavior. At home, they see their daughter as overactive, sometimes careless, given to mood changes, easily frustrated, but generally cooperative. She reportedly requires considerable external structure, such as in the morning, to assure that she is ready to be out the door on time. Without prompting, for instance, she may forget to brush her teeth or put homework in its designated spot in her backpack.
Family history is reportedly negative regarding developmental, emotional, or behavioral concerns. Turning to Shannon’s personal history, her mother did indicate the presence of one condition. Repetitive throat clearing emerged about six months ago, and when it intensified Shannon was taken to her pediatrician. A subsequent visit with a child neurologist confirmed the presence of Tourette syndrome.
Turning to broadband rating scales, the Conners CBRS was completed by Shannon’s teacher and by her parents. Both the Conners Clinical Index and the DSM-IV-TR ADHD Hyperactive-Impulsive symptoms scale were elevated above a recognized cut-point (i.e., T-score > 70). This held true for both sets of informants. There was also endorsement of impairment across settings on the Conners CBRS. No other Conners CBRS DSM-IV-TR scales, including those related to depression, were elevated. This information helped Lakisha because she was considering ADHD and depression as diagnostic possibilities.
As the assessment process moved forward, Lakisha observed Shannon twice in her classroom and once on the playground. The classroom observations were conducted using the Behavioral Observation of Students System (BOSS, Shapiro, 2010a). The playground observation resulted in notes and anecdotal comments but no structured observations. Lakisha found Shannon to be aggressive and overactive on the playground. She was not seen as moody or unhappy, however. Interestingly, BOSS percentages were unremarkable. Specifically, during 48 intervals on one classroom visit and during 48 intervals on another classroom visit Shannon’s percentage of Active Engagement Time, Passive Engagement Time, Off-task Motor, Off-task Verbal, and Off-task Passive were almost identical to representative classmates observed on the same occasions.
When Lakisha interviewed Shannon, she disavowed depressed mood, worry, obsessions and compulsions, and noteworthy stress at home or school. She also denied a history of trauma. Nonetheless, she confirmed discouragement and diminished self-esteem, especially in class. She had observed, for instance, her own growing difficulty completing work. Simultaneously, she had begun to sense her teacher’s frustration with her. Favorably, Shannon presented as a friendly and ostensibly cooperative young lady free of any remarkable physical features. Lakisha noted several instances of throat clearing (tic) but Shannon seemed either unaware of these occurrences or to simply disregard them. There was nothing unusual or peculiar about Shannon’s social relatedness or her thought processes.
All of this information could be summarized various ways. Two are: (1.) via the HR Worksheet and (2.) repeated probability nomograms (see Appendix D). Similar conclusions arise from each process. Critically, both approaches arguably accomplish similar missions—they require a deliberate, point-by-point consideration of assessment data. That is, they constraint the fast, intuition-loving Automatic System while they compel use of its more rational, analytical counterpart, the Reflective System. Diagnostic errors should consequently shrink. That said, with the use of nomograms there always remains a question about just what final post-test probability is sufficiently elevated to conclude that a particular condition is present. Shannon’s final post-test ADHD probability of around .90 certainly seems quite high. Unfortunately, there are no established rules of thumb. Still, the following general guidelines are offered regarding a final post-test probability by Keiser and Reddy (2013):
- .90 diagnostic of ADHD
- .80 – .90 suggestive of ADHD
- <.80 not diagnostic of ADHD
No matter how Shannon’s information is considered, ADHD looks to be the most likely conclusion (with comorbid Tourette’s syndrome). Still, Lakisha would probably choose to nail down her high stakes conclusion by turning to the ADHD criteria found in DSM-5 (see Table 12.7, right column). The argument here is that probability nomograms can help create a final quantitative index but that a formal ADHD diagnosis depends on satisfying all elements of textbook criteria. The sole definitive source of those criteria is found in DSM-5. Only after their confirmation might she decide to write into her report a conclusion that ADHD is present. Accordingly, Lakisha might create a simple checklist of DSM-5’s ADHD criteria and verify every item on her list.
The outcome of the case turned out to be no surprise for Lakisha Washington. A meeting to review evaluation findings included both parents, the school district’s coordinator for Section 504 services, Shannon’s teacher, and Lakisha. The coordinator was present in anticipation of potential implementation of a 504 accommodation plan. As Lakisha shared her findings, it became apparent that parents did not want such a plan. Instead, after acknowledging Lakisha’s ADHD conclusion, they requested several actions: (1.) sharing basic ADHD information with Shannon’s teacher, (2.) a follow-up meeting between Lakisha and Shannon, (3.) a copy of the school’s report sent to Shannon’s pediatrician.
Regarding the first request, Shannon’s parents indicated that they hoped basic information about ADHD would help Shannon’s teacher understand the actions of their daughter. In the words of Shannon’s mother, “the first step to compassion is understanding.” Regarding the second request, Shannon’s parents anticipated that an explanation of her condition could help diminish Shannon’s intermittent bewilderment over her own actions. Regarding the third request, parents anticipated that Shannon’s pediatrician would consider a medication trial. This physician had already told parents that the standard treatment for ADHD is pharmaceuticals coupled with behavior management. Parents were intent on a pediatrician-to-behavioral therapist referral to assist them. Shannon’s father made it clear that the prospect of a 504 plan might be revisited later. Lakisha was relieved that the process turned out so well.
Table 12.7 Shannon’s Summarized Regarding Probability and Tasks Yet to be Accomplished |
|
What is already learned about Shannon’s probability of ADHD by repeated nomograms |
Yet to be completed to establish the presence of ADHD |
|
Address with DSM-5 each of the following. Pages 59-66 in DSM-5 elaborate these points:
|
Cultural and Linguistic Considerations, Lakisha and Shannon
You saw earlier the importance of cultural and linguistic considerations regarding all manner of referral questions. This includes deliberate cognizance of parents’ language use and their cultural perspective when collecting background information (Chapter 4) and awareness of language preferences (and potential use of cultural norms when using broadband scales; Chapter 5). These matters should not be disregarded when a referral question concerns ADHD or when ADHD emerges as a working hypothesis. But there’s more. Research shows that although ADHD diagnoses over time grew among all US school children, noteworthy racial disparities also characterize trends. In other words, students from minority groups may not receive ADHD diagnoses (and treatment) at a rate comparable to their European-American counterparts (Morgan, Staff, Hillemeier, Farkas & Maczuga, 2013). A recent, large scale study using several representative samples provides details (Visser et al., 2015). Among fifth graders, 16% of European-American children were diagnosed with ADHD compared to 9% of African-American and just 4% of Latino children. By 10th grade the values for the same three respective groups were 19%, 10% and 4% (Visser et al).
Lakisha and Shannon’s case (with a middle class, African-American, English-only-speaking school psychologist and a middle class, African-American, English-only-speaking family) would seem to impose few cultural and linguistic stumbling blocks. There was, for example, no ethnic incongruence (see Chapter 4 for an explanation). Plus, Shannon’s family possesses resources to access primary medical care (and a pediatrician’s consideration of ADHD) if they had chosen to do so. But that is not always true. What if Lakisha were dealing with an immigrant family from Honduras lacking financial means, access to health care, and basic transportation? Moreover, what if this same family was neither aware of the educational importance of addressing ADHD nor disposed to advocate for an assessment even if such awareness were present? The same might be true for families who are neither African-American nor Hispanic but who are poor.
Although minority membership or SES status may be associated with limited access to health care, it should not constrain access to committed and professional school psychologists. Those school psychologists may help assure essential opportunity to have ADHD identified in schools. Consider the clarion call made by the researchers mentioned above when they published in a pediatric, not a school psychology, journal: “Our findings provide additional support to calls for increasing solicitations by pediatricians, school psychologists, teachers, and other clinicians of concern by minority parents for their children’s learning and behavior, ensuring sensitivity to differing cultural values about disability during well-child visits and referrals, and encouraging symptom recognition and help-seeking behaviors by these parents, and suggest that these efforts by clinicians should be made throughout minority children’s early life course.” (Morgan et al., 2013, p. 91). This statement seems to beg for school psychologists to act assertively “in the best interest of the student.” Those school psychologists may help assure equal opportunity to have ADHD identified in schools. Recognizing that LEA’s, not parents, are obliged to identify all students with ADHD is arguably a first step.
Summary
Controversy and professional insecurity seem especially common when school psychologists confront the prospect of making an assessment that involves ADHD. Critically, it is now clear that school districts have an obligation to locate and assist students with ADHD. School psychologists are able to assist in this process even though most ADHD identifications now occur out of school. School psychologists’ standard assessment techniques of accessing background information, observing, interviewing, and using multiple rating scales work well when there are questions about ADHD. A structured process such as using the HR Worksheet or probability nomograms can aid the process. Final confirmation by using DSM-5 criteria is a practice compatible with the position of various professional organizations, including the National Association of School Psychologists.