The term “Intellectual Disability (ID) equivalence” refers to accommodations that are made by legal and other governmental entities when they provide services, supports or protective arrangements to people who—because of brain impairment—function as if they have ID but fail to qualify for the ID label which is needed for access to many programs because their IQ scores are a few points too high. Fetal Alcohol Spectrum Disorder (FASD) is a logical candidate for such an accommodation as (a) it (like ID) involves brain impairment, (b) people with FASD have adaptive deficits and support needs that are identical to those with ID, and (c) while many people with FASD do qualify as having ID, the majority do not, because full-scale IQ scores are typically too high.

ID-equivalence accommodations are an attempt to free the human services field from the straight-jacket that has been imposed by over-reliance on full-scale IQ ceiling scores, which functioned as a gate-keeper and barrier for developmental services eligibility. In this chapter, using specific examples, we review some of the forms that these ID-equivalence accommodations have taken, point out problems with those forms, and suggest improvements for implementing this concept. First, we provide a brief historical overview of the role of intelligence in intellectual disability and the growing belief that an IQ score provides an inadequate basis for determining whether someone may or may not have that disorder.

Intellectual Disability, under various names and most of them now offensive, has been around for centuries, as seen in references to it in Egyptian, Greek, Roman and other ancient documents (Scheerenberger 1983). Until the advent of intelligence testing in the early twentieth century, individuals with ID were identified by how they functioned in the everyday world or what has come to be termed “adaptive behavior”. Specifically, people with ID were differentiated from the general population by the perception that they lacked ability to survive on their own. With the invention of the IQ statistic, diagnostic emphasis shifted from everyday functioning as seen by others in one’s environment to “intelligence” as measured on a one or two hour test of mainly academic problems, administered by a psychologist who typically did not know the child or adult being tested.

The original Binet-Simon scale, on which most subsequent tests were modeled, took progressively more advanced items from the educational curricula. Thus, as was argued by Anastasi (1983), there was essentially no difference between a measure of aptitude (IQ) and a measure of academic achievement, as seen in the extremely high correlations among the two types of instruments and extremely good predictive validity between IQ and academic performance. When there was divergence between aptitude and achievement scores, as in former but now largely abandoned operational definitions of “learning disabilities,” these often reflected differences in test reliability as well as the fact that achievement profiles were typically differentiated in most people, including those with brain impairments. The question thus becomes: “how appropriate is it to place central reliance on a measure essentially of academic potential as a gateway to receiving developmental services, subsidies and entitlements?” The answer, it seems to us and many others, is “not much.”

The IQ metric became overwhelmingly popular in North America in the second quarter of the twentieth century, as it proved useful to eugenicists such as Lewis Terman (author of the Stanford-Binet, a translation and extension of the original French Binet-Simon) in promoting their racist political agenda (Blumenfeld 2011). That agenda included shutting down immigration to the United States from Southern and Eastern Europe, placing individuals with below-average IQ in gender-segregated institutions before subjecting them to involuntary sterilization, and otherwise preventing those termed “morons” (many of whom today would likely not be considered to have ID) and “imbeciles” (people who today would be considered to have mild ID) from transmitting their “diseased germ plasma.” The implementation of that agenda depended on being able to persuade legislators and the public through quasi-scientific or outright fraudulent studies such as Henry Goddard’s (1912) book on the “Kallikak” family, that low-IQ people posed a threat to society and to the continued superiority of the “white race.” Much evidence that Goddard faked his Kallikak data can be found in Smith (1985).

The basic idea being promoted at the time was that IQ represented a precise snapshot of the brain that was concrete, immutable, inherited, and incorporative of all that we view as “intelligence.” Now of course we understand that there are discrete cognitive capacities such as “social intelligence” that are not well captured by an IQ score. We also recognize that some cognitive measures such as “executive functioning” are better indices of intelligence than IQ, and that cognitive profiles across batteries of IQ and other tests are more useful than a single index score (McGrew and Flanagan 1998). It is also accepted that there are environmental as well as other biological (e.g., brain injury) contributors to poor intellectual functioning (Ceci and Williams 1997), and that instruments which purport to measure IQ are not equally reliable or valid and are affected by well-known psychometric variables, such as experience with prior tests, obsolescence of test norms, problems with test construction and statistics, and normal variation in performance that is random or due to examiner incompetence or bias (Greenspan and Olley 2015).

The idea of using IQ ceilings to demarcate a dividing line between impairment and “normality” came to the fore a little over half a century ago when the American Association on Mental Deficiency (later, the American Association on Mental Retardation and today, the American Association on Intellectual and Developmental Disability) published a diagnostic manual. In this manual, they established a three-prong definition of mental deficiency (later, “mental retardation” and today “intellectual disability”) as a condition in which the first prong, intellectual impairment, was reflected in an IQ score that fell at least one standard deviation (SD) below the population mean (a standard score equal to or less than 85, or the 17th percentile of the population distribution). The conventional belief, not particularly based on research, was that the incidence of ID was 3 % of the population. Therefore, a standard that took in the bottom 17 % of the population was obviously too high as it created many “false positives” of individuals labeled ID who should not have been. This problem was exacerbated by a failure for over a decade to use the second diagnostic prong of adaptive behavior, the intended purpose of which was to bring the actual incidence down to 3 %. To correct for this problem, the criterion for prong two was changed three decades ago from one standard deviation (IQ = 85, 17th percentile) below the mean to two standard deviations (IQ = 70, 2nd percentile) below the mean (Greenspan and Switzky 2006). However, the minus 2 SD standard was just as arbitrary as the minus 1 SD standard had been two decades earlier. No apparent rationale was provided for using the minus 2 SD cut-score other than the superficial elegance of using the statistical convention of standard deviation units. Just as the earlier use of the minus 1 SD criterion was too easy a hurdle to clear, this new reliance on a minus 2 SD criterion was too difficult a hurdle, with the result being an excessive number of “false negatives” of people who deserved ID eligibility but were wrongly denied it. Most of the definitional reforms essayed over the past several decades were motivated by an attempt to rectify the problem of false negatives caused by setting the qualifying IQ score too low.

To date, four solutions have been proposed to overcome the problem of false negatives. The first attempted fix in the 1980s (Grossman 1983) involved encouraging diagnostic evaluators and agencies to take into account the confidence interval of the IQ test (5 points at the 95th percent of confidence) when making ID diagnostic determinations. When this admonition was largely ignored, the second attempted fix, enacted in the 1990s (Luckasson et al. (1992), was to take the suggested five-point confidence interval and make it the new ceiling standard, later modified (Lucksson et al. 2002) as the IQ range “70–75.” More recently, this second fix has involved using 75 as the cutoff point (Schalock et al. 2002), which is also what DSM-5 adopted. Part of this fix was to insert the word “approximately” before the number 70 or, more typically, the term “approximately minus two standard deviations below the mean,” with “approximately” meaning that one should take into account the unreliability or standard error of the IQ statistic. When changing the ceiling was typically ignored, a third fix was to encourage clinicians to change the number itself by correcting for norm obsolescence, a phenomenon known as the “Flynn effect” (Gresham and Reschly 2011). The Flynn correction process involves subtracting 0.3 IQ points per year of elapsed time between the date of norming and date of test administration to correct for the fact that norms are toughened by 3 years per decade of norm obsolescence to adjust for changes in population performance in the interim. This practice has become commonplace when determining eligibility exemptions in the US under the Atkins v Virginia standard but has not typically been used for school, residential or other less catastrophic purposes. The fourth fix is the one with the most promise for eventually solving the false negative problem and the most relevant for the ID-equivalence issue. It involves approaching the first diagnostic prong of “intellectual functioning” as a broad construct that is tapped by various cognitive measures of which IQ is but one data point. This is the position underlying the ID section in DSM-5 (American Psychiatric Association 2013), which actually states that neuropsychological tests, particularly of “executive functioning,” are typically more useful diagnostically than an IQ score. This provision reflects a step away from the “disability” (arbitrary, numbers-based) view of ID to one that views the condition as a brain-based “disorder” determined clinically rather than psychometrically. In such a conceptualization, the key is to look at the individual as a whole person and not just through the narrow lens of a single numeric score. This position will be explored more fully in the concluding section of this paper, but first we will look at how various government entities have put in place ID-Equivalence provisions intended to get around our slavish and inappropriate reliance on arbitrarily-instituted full-scale IQ ceilings.

Various attempts have been made over the years to address the inequitable obstacles to ID service eligibility posed by the continuing rigid reliance on full-scale IQ cutoffs. People with FASD are among those most victimized by the current practice and therefore stand the most to gain from efforts to develop more flexible frameworks. In the following pages, we describe some of these ID-Equivalency approaches, and point out both their strengths and limitations.

Neal Fahlman is a young man in British Columbia, Canada, of First-Nations ethnicity who was adopted at age five weeks. As a child he had three diagnoses: FASD, ID and what today would be termed ASD (Autism Spectrum Disorder), in addition to ADHD (a common co-morbid problem for people with FASD). As a child and pre-adolescent, Neal qualified for developmental disability (DD) services as his full-scale IQ was below the minus two standard deviation threshold used in BC by Community Living British Columbia (CLBC, a Crown Corporation) to determine eligibility for residential DD services. He also met the other two criteria for a diagnosis of ID: onset in the developmental (pre-age-18) period and significant deficits in adaptive functioning. Starting at age 15, Neal was funded by CLBC for living in foster homes and eventually in his own small home due to behavior outbursts, with one-on-one daily supports but with continued involvement by his adoptive parents. The program was costly ($77,000 per year) but less expensive than institutional alternatives (such as prison), given Neal’s very poor judgment and impulse control problems (e.g., lashing out at others).

When Neal reached age 19, his eligibility for CLBC services needed to be re-established. However, his newly-obtained IQ score of 79 was now a few points above the approximately minus two standard deviations level, which was cited as the basis for denial of services even though his adaptive behavior scores were still in the moderate (minus three standard deviations) range. In many other states or provinces, once someone is in the system and begins receiving services, eligibility remains intact, regardless of any change in IQ. Neal’s family sued CLBC and a panel of judges ruled that the use of a specific IQ cut-off to determine continued eligibility, which had never been discussed in the agency’s authorizing legislative history, was arbitrary and unfair. The agency was ordered to continue serving him and to devise a more equitable, and less IQ-driven, formula for making eligibility determinations for other applicants.

After considering a framework with broader applicability, the BC Ministry of Community Living Services (the government agency over CLBC) came up with a formula that appears to have been very narrowly tailored to people exactly like Neal Fahlman. Essentially, they established two pathways to residential services: (a) the traditional one, for people who qualified for a diagnosis of ID, which required an IQ below 70 along with adaptive deficits and onset in childhood; or (b) an alternative pathway, only available to people who—like Neal—had ASD or FASD. In addition to having onset in the developmental period, successful applicants needed to have deficits in adaptive and intellectual functioning, but IQ scores could be above the traditional 70–75 ceiling. However, to ensure that applicants were sufficiently deserving of services, prong two (adaptive functioning) required a score that fell at least three standard deviations below the population mean on a standardized rating measure. This happened to be the precise profile characterizing Neal Fahlman.

There are three problems with the ID-Equivalence solution in British Columbia that was devised in response to the Neal Fahlman lawsuit. The first problem was that the BC “fix” continued to be based on artificial and arbitrary numeric criteria, i.e., the numbers were adjusted to make the judges in the Fahlman case happy. The underlying issue raised by the court—namely the need to look at the whole individual—was not really addressed at all.

A second problem with the BC solution was that adaptive deficiency was defined as minus three standard deviation units, which was far too stringent a requirement as it included only a fraction of the first percentile of the population. This was also grossly inconsistent with the usual prong two criterion, which at minus two standard deviation units (at or below the second percentile) was more in line with the population for whom ID-equivalence relief was being sought. Our guess, which has been confirmed by conversations with experts and applicants in BC, is that adaptive behavior informants such as parents and even evaluators such as psychologists are then motivated or perhaps forced under the BC solution to exaggerate the severity of an applicant’s deficits. There is obviously something wrong with such a state of affairs, as it is inequitable to reward those savvy or dishonest enough to exaggerate, while punishing those too unsophisticated or honest to do so.

The third problem with the BC solution is that the ID-equivalency pathway was made available only to people with one of the two disorders—ASD and FASD–which applied specifically to Mr. Fahlman. What about people with the dozens if not hundreds of other brain-based developmental syndromes who have the same problem of significant adaptive needs but whose IQs straddled the minus two standard deviation demarcation line? For example, do individuals with Prader-Willi syndrome or Dandy-Walker malformation, where there is an ID-equivalent need for services but where IQs sometimes fall above 70, have to file their own lawsuits when they are denied services? A more equitable solution would have been to use general language, such as “persons with brain-based neurodevelopmental disorders” rather than “persons with autism or FASD.” Consideration of equity applies here as well, as it is unfair to give relief to very politically influential advocacy groups (such as parents of autistic children) and deny it to equally deserving clients (such as individuals with, say, Dandy-Walker syndrome, where IQs also fall on both sides of the 70–75 barrier) whose disorders are less known and thus lack comparable visibility or influence.

Just as FASD is granted ID-Equivalency in British Columbia (as long as the minus three standard deviation criterion on prong two is met or faked), other specific syndromes have been specified in various jurisdictions for ID-Equivalency. A difference is that for these disorders, eligibility is sometimes met automatically if the syndrome diagnosis is established without any, let alone a more severe, adaptive behavior deficit finding. Perhaps the most prevalent of these exemption examples includes Prader-Willi syndrome (PWS), a rare (1:10,000 to 1:25,000 live births) developmental disorder caused by a deletion or partial expression of several genes on chromosome 15. Among the symptoms of PWS are cognitive deficits typically, but not always, involving IQ below minus two standard deviations, significant adaptive deficits, and a compulsive need to eat constantly often resulting in morbid obesity.

A statute in Connecticut gives automatic ID-Equivalency status exclusively to applicants for ID services who have PWS, regardless of IQ. According to a 2006 website description of the Connecticut Department of Mental Retardation (since renamed the Department of Developmental Services): “An application for eligibility determination may be made by… any person who is a resident of Connecticut at the time application is made, or by someone on the person’s behalf, and who is, appears to be, or believes him/herself to be a person with mental retardation, as defined in Connecticut General Statutes 1–1 g or Prader-Willi Syndrome…” Later, the eligibility criteria are spelled out thusly: “…The process by which the Department examines information relative to an applicant for Department to determine if the applicant meets the statutory criteria for mental retardation or Prader-Willi Syndrome,” and PWS is described as “a neurobehavioral genetic disorder that has been diagnosed by a physician utilizing medically appropriate criteria.” Such a diagnosis, which until the 1990s was based mainly on clinical criteria (e.g., hypotonia, hypogonadism, inability to achieve food satiation), is now derived from extremely reliable DNA-based genetic testing methods.

The relevant aspect of the Connecticut exemption is that applicants with PWS do not have to meet any other criteria for eligibility such as level of adaptive behavior impairment. Apparently it is just assumed that an individual with PWS, a spectrum disorder with varying levels of severity, has significant enough service needs to merit DD services whether or not his/her IQ score falls below 70–75. Several other states also list PWS in their eligibility criteria but embed it within a slightly broader list of neurodevelopmental disorders and also specify that an individual still has to meet prong two, the adaptive functioning deficits. An example is Wisconsin, which specifies that developmental services are for people who have “…a disability attributable to brain injury, cerebral palsy, epilepsy, autism, Prader-Willi syndrome, intellectual disability, or another neurological condition closely related to an intellectual disability or requiring treatment similar to that required for individuals with an intellectual disability, which has continued or can be expected to continue indefinitely and constitutes a substantial handicap to the afflicted individual….” Similarly, Massachusetts affords ID-Equivalence status to people with “…cerebral palsy, epilepsy, autism, Prader-Willi or other condition other than mental illness or an emotional disturbance; closely related to mental retardation because the condition results in impairment of general intellectual functional or adaptive behavior similar to those with mental retardation; manifests before 22 years of age; [is] likely to continue indefinitely; [and] results in substantial limitations in three or more of major life activities: self-care, understanding and use of language, learning, mobility, self-direction, [and] capacity for independent living.”

According to a PWS researcher with first-hand knowledge (Dykens 1996), the legislation that established automatic ID-Equivalence for applicants with PWS in Connecticut occurred as a result of successful legislative lobbying of the state legislature by PWS parents and advocates. Not long after this legislation was enacted, a request was made for similar special treatment (i.e., ID-Equivalence) by parents of children with autism which is also a spectrum disorder straddling both sides of the 70–75 IQ divide but with virtually all diagnosed individuals exhibiting severe adaptive deficits and service needs. However, this request was turned down for three related reasons: (a) autism is diagnosed clinically without absolute certainty of a biological (DNA) test as in PWS; (b) the diagnosis of autism is made too freely (which reportedly is why DSM-5 eliminated Asperger disorder) and consequently includes many individuals unlikely to ever need disability services; and (c) even if an autism diagnosis was reliable and valid, the numbers of affected individuals would be enormous (perhaps as high as 1 %) in comparison to PWS; thus granting automatic ID-Equivalency to all autistic individuals would likely have severe fiscal consequences for any state or provincial human services budget. Unfortunately, ID definitions and diagnoses are in part driven by political and economic considerations, much as we might wish that were not the case.

By implication, therefore, the political initiative of PWS advocates on behalf of automatic ID-Equivalency status was more likely to be successful because diagnosis of PWS was highly reliable, virtually all affected individuals needed and deserved DD services, and—last but not least—the prevalence rate for PWS was low enough that such a scheme was financially viable. Still, one could make the same inequity argument against singling out PWS in Connecticut as we made against singling out FASD (and Neal Fahlman’s other diagnosis, autism) in British Columbia. Moreover, the case against PWS could be made even more strongly, as the lawsuit-driven solution in BC still required significant evidence of adaptive deficits. In fact, this was so excessive as to almost ensure fraud, while such (or any) evidence of adaptive deficits was not a requirement in Connecticut for PWS eligibility. Looking at this history and convoluted bases for “fixes” to the eligibility criteria, it seems questionable to us that a single genetic disorder is singled out for special ID-Equivalency treatment when there are so many other syndromes (Patau and Edward, to name but two) that are equally deserving. It is clear then that the broader problem goes beyond the naming of a specific ID-qualifying disorder and involves the somewhat outdated emphasis on underlying specific medical categorization itself.

A major way in which government entities increase the population eligible for ID services beyond those with IQ scores at or below 70–75 is to use the broader term “Developmental Disabilities” (Administration for Community Living 2013; Disability Law Center, undated; National Council on Disability 2012; Zaharia and Moseley 2008). This term owes its origins to pioneering Kennedy-era legislation in the United States which authorized various government-funded disability-related human rights, research, training and facility-building initiatives (Gettings 2011). The term is now widely used in many state and provincial eligibility statutes. The original legislation—the “Mental Retardation Facilities and Community Mental Health Centers Construction Act of 1963”—defined the population addressed by these programs as people with “mental retardation.” Seven years later, the law—renamed the “Developmental Disabilities Services and Facilities Construction Amendments of 1970”– was revised, with the term “Developmental Disabilities” (DD) substituted for “mental retardation.” However, DD was still defined categorically as in this expanded list: “mental retardation, epilepsy, cerebral palsy, and other neurological conditions originating before the age of 18.” In 1975, the legislation was revised again, and DD was again defined categorically to include mental retardation plus these conditions “closely related to mental retardation”: cerebral palsy, epilepsy, autism and dyslexia, again with a pre-18 age of onset, which were expected to continue indefinitely and that constitute a substantial handicap.” The term “other neurological conditions” was dropped for some reason. In our opinion that was a mistake as such a loophole along with qualifying language such as “producing service needs similar to those needed by people with ID”, has since been added in many places to keep IQ-Equivalency from being unfairly limited to people in only a few, and somewhat arbitrarily chosen, diagnostic categories.