The Epidemic of Inadequate Adult Autism Assessment
- Chris Dabbs
- May 6
- 15 min read
Since 2022, I have had a steadily increasing number of verbally-fluent adult clients seek autism assessment. In fact, this has occurred with such regularity that I was able to, with support from my clinical agency, build and complete an entire post-doctoral residence around providing adult autism and ADHD assessment. A little about me: I’m a psychology professor, a researcher focusing on cool things like autism and religion, and a psychological diagnostician. I am also autistic; I was clinically diagnosed at age 29, after having been self-identified (self-diagnosed) for some years prior. I spend my time writing papers about autism and religion, lurking on #AutRes on BlueSky, and volunteering with an autistic non-profit. I care deeply about autism, the autistic community, and psychology.
Many of my clients over the past few years are working professionals, high-masking autistics, in their mid-20s to mid-30s, and demographically lean female and White. This matches current demographic research into adult autism diagnostics which suggest a 450% diagnostic rate increase from 2011 - 2022 in the 26-to-34-year-old demographic slice. Due to changes in autistic diagnostic criteria and definitions, greater awareness of heterogeneous autism presentations, increased diagnostic tool validity and sensitivity, and decreasing stigma, what was once a niche diagnostic area (autism in adults) is now an extremely high-demand field. I cannot express how absolutely jacked I am that adults are seeking, and receiving, autism diagnoses that are helping them understand themselves and making their worlds more accessible. However, now that this field of adult autism assessment has been opened, there are obvious foundational cracks becoming apparent. These foundational issues are beginning to threaten the integrity of testing and assessment structure as a whole, and are being quickly widened by, from my vantage, three core issues: definitional problems in testing and assessment, concerns with provider qualifications, and quality control issues.
While often conflated, testing and assessment are two distinct, yet complementary, processes that require different skill sets–their interrelation has created some definitional confusion across and within disciplines that we will circle back to later. Here I provide a couple of popular definitions, starting with those from the American Educational Research Association, American Psychological Association, and National Council on Measurement in Education who co-authored the joint Standards for Educational and Psychological Testing (hereafter, Standards). The Standards define psychological assessment as the “examination of psychological functioning that involves collecting, evaluating, and integrating test results and collateral information, and reporting information about an individual” (p. 222) and psychological testing as “the use of tests or inventories to assess particular psychological characteristics of an individual” (p. 222). However, even the Standards have some circularity to their definitions, as can be seen in the definition of testing: “the use of tests or inventories to assess…”
Hogan (2018) attempts to delineate testing into six core functions represented by his definition: “a standardized process or device that yields information about a sample of behavior or cognitive processes in a quantified manner” (p. 29). Here, then, testing is defined by six components: 1) standardization, 2) a process or device, 3) information yield, 4) a sample, 5) regarding behavior or cognitive processes, and 6) quantified. Testing, then, is the discrete process that results in quantifiable data. This is distinct from assessment which, according to the Standards, is a broader term that involves the integration of test data, historical information (e.g., educational, social, medical histories), collateral information, and behavioral observations. To put it simply: testing provides raw quantitative data while assessment is the process of making meaning of that data in qualitative context.
It is clear that testing and assessment are different occupational activities requiring different skill sets to perform adequately. Based on my professional experiences in the field, primary skills required for testing include technical proficiency, an understanding of psychometrics, and specialized instrument knowledge. First, technical proficiency relates to the standardized administration of tests (when required) such that “examiners [are] careful to adhere to the same administration and scoring procedures that were used during the standardization of the test” (Lichtenberger & Kaufman, p. 41). Small deviations from the standardized processes can invalidate results and lead to misinformed and incorrect conclusions. Second, psychometric skills are required to fully understand and consider the selection of tests to give to an individual client. For example, if my client is an adult with suspected executive functioning struggles, it would facilitate my selection process to understand that the adult-version of the Behavior Rating Inventory of Executive Function (BRIEF-A) has demonstrated strong internal consistency, test-retest reliability, and convergent validity with other measures of executive functioning for the population into which my client fits. Understanding normative data to ensure that tests are appropriate for my client’s age, gender, ethnicity, racial group, language needs, etc. are also required in appropriate test selection. Finally, providers need to have specialized instrument-based knowledge. For example, during Module 4 of the Autism Diagnostic Observation Schedule (ADOS-2), it takes specialized knowledge for an examiner to accurately code the quality of social overtures in order to distinguish between those that are socially awkward and those that are instrumental (Hus & Lord, 2015).
Assessment, too, requires specialized skills. The import of the assessment process cannot be overstated and “requires a high degree of skill and sophistication to be implemented properly” as “no clinical question can be answered solely by a test score” (Meyer et al., 2001, p. 144). From my vantage, specialized assessment skills include honed clinical judgment, expertise in diagnostics, critical integration skills, and good therapeutic communication. In all assessment, but particularly adult autism assessment, sharp clinical judgment and diagnostic expertise are necessary to cut through the inevitable copse of likely co-occurring diagnoses, including ADHD, anxiety disorders, and disorders of aggression (Lord et al., 2018). Do a client’s social difficulties extend beyond anxiety-provoking situations and include fundamental challenges with social perspective-taking across contexts? What is the etiology of a deep-seated fear of negative evaluation in social situations? Understanding the answers to these questions assists with a complex differential, as many of the symptoms are cross-cutting. Additionally, a professional engaging in testing and assessment services must have critical integration skills in order to integrate often contradictory findings. Particularly in adult autism assessment, we have to be able to make sense of the mismatch between performance and objective test results and qualitative assessment findings (e.g., the presence of PAI SCZ-S elevation, an indication of social detachment, but a verbal history of team-based sports interest), and we must be able to explain these discrepancies to our clients and patients in jargon-free, understandable ways.
The above occupational activities–psychological testing and assessment–have long been situated in the subfields of health service psychology: clinical psychology, counseling psychology, and school psychology. Each of these fields have historically and generally gravitated towards a testing and assessment niche: forensic and neuropsychological testing in clinical psychology, personality and aptitude testing in counseling psychology, and educational and intellectual testing in school psychology. The paradigms in which we think and teach about testing and assessment in the Western world were developed by psychologists: Galton (English), Wundt (German), Cattell (American), Binet (French) (Gregory, 2013). For around 150 years, the field of psychology has held testing and assessment as one of its core occupational features. In fact, the accreditation standards from both the American Psychological Association (APA) and the Canadian Psychological Association (CPA) include assessment and testing, with the CPA standards noting that assessment is the use of “more than one type of assessment approach (e.g., intelligence testing, behavioural assessment, personality testing, psychoeducational assessment, diagnostic assessment, cognitive assessment, neuropsychological assessment)” (p. 16).
However, it is not only the field of psychology that makes use of psychological tests and assessment. I know this first-hand: I’m a licensed mental health counselor with discrete training at the master’s level (a terminal master's in clinical mental health counseling). The fields of professional counseling, social work, marriage and family therapy and others make use of testing and assessment materials to best serve their clients. Accreditation standards from the largest counseling and social work accrediting bodies in the United States–the Council for the Accreditation of Counseling and Related Educational Programs (CACREP) and the Council on Social Work Education (CSWE)–include testing and assessment in their standards. This is logical, given the need for counselors, social workers, and all mental health professionals to assess client welfare, symptoms, history, etc. However, harkening back to my previous point about definitional problems, how testing and assessment are defined in these standards is often vague, leaving a wide berth of interpretation. For example, CACREP standard G reads “use of assessments in academic/educational, career, personal, and social development” (p. 15). This descriptor lists, in effect, types of assessment, but not types of testing. For example, career assessment could include personality testing, aptitude testing, or intelligence testing. In fact, the only types of testing directly mentioned in the CACREP standards are “personality and psychological testing.” In reality, this is just a singular type of testing: personality. “Psychological testing” is vague enough to carry no pragmatic meaning for a standard.
The heterogeneous nature in which occupational standards are defined and applied has caused problems in other healthcare professions and has a term: scope creep. Scope creep happens when one field slowly begins to push into the occupational duties of another field. For example, nurse practitioners (mid-level healthcare providers with master’s degrees) weren’t always permitted to prescribe medications, as this practice fell within the scope of physicians. However, over time, nurse practitioners argued for, and now reserve the right to, prescribe medication. This same scope creep is currently happening with physician’s assistants (PA), who have been arguing for increasingly more practice independence. In mental health care, what I see is a type of scope creep from underprepared professional counselors and social workers in the realm of psychological assessment, a reality that is negatively impacting some of our most vulnerable populations.
Scope creep begins with accreditation standards and then moves into law. For decades, psychologists have been pushed out of masters training programs in counseling and social work--with CACREP formally banning psychologists from being core faculty in their programs in 2013. Who teaches the counselors to engage in psychological testing and assessment if psychologists aren't allowed to teach in their programs? Today, in my home state of Indiana, counselors are legally permitted to “administer and interpret appraisal instruments that the mental health counselor is qualified to employ by virtue of the counselor’s education, training, and experience” (IN Code 25-23.6-1-7.5). To break this down: “appraisal” is defined in the State Code as “the use or administration of career and occupational instruments, adaptive behavioral and symptoms screening checklists, and inventories of interests and preferences that are administered for the purpose of counseling persons to cope with or adapt to changing life situations that are due to problems in living…” (IC 25-23.6-1-1.5). This definition is arguably fairly specific, but some states are even more clear with what is and is not within-scope. For example, the Texas counselor standards specifies: “assessment…does not include the use of standardized projective techniques or permit the diagnosis of a physical condition or disorder” (OC 503.003.b.1; p. 102). Even more specific is the California counselor regulations which read:
"“Assessment” means selecting, administering, scoring, and interpreting tests, instruments, and other tools and methods designed to measure an individual’s attitudes, abilities, aptitudes, achievements, interests, personal characteristics, disabilities, and mental, emotional, and behavioral concerns and development and the use of methods and techniques for understanding human behavior in relation to coping with, adapting to, or ameliorating changing life situations, as part of the counseling process. “Assessment” shall not include the use of projective techniques in the assessment of personality, individually administered intelligence tests, neuropsychological testing, or utilization of a battery of three or more tests to determine the presence of psychosis, dementia, amnesia, cognitive impairment, or criminal behavior." (p.129).
Definitions of negation provide complementary data to the overwhelming positive definitions and legally defining what "psychological testing and assessment" could help curb some of this confusion.
Organizational preferences and legal policies don’t always align, but strong lobbying efforts can bring them closer together. There is some organizational alignment, particularly in the field of professional counseling, to see counselors as trained psychological testers and assessors. For example, the National Board of Forensic Evaluators (NFBE) are perhaps the most fervent proponents for counselor psychological testing and assessment privileges and make the argument that professional counselor standards include assessment and, therefore, counselors are trained in psychological testing and assessment. The NBFE and this 22-year-old position paper from an American Counseling Association working group stress something similar: with postgraduate training and experience, master’s-level counselors can give objective, projective, and intelligence tests. I do not disagree with this notion. With appropriate training, there is no ethical or legal reason that master’s-level clinicians could not provide psychological testing. In fact, states and provinces that license psychology at the master’s-level (e.g., Alberta, Kentucky, Michigan) are already doing this.
Almost universal across scope of practice codes for master-level clinicians are provisions for widening their scope of practice. Learning is a lifelong process–if a professional sees a need, and wants to help fill that need, they should be provided an avenue through which to specialize or re-specialize to do so. I am staunchly in favor of this practice: I did it myself. After earning my master’s degree in counseling, I went on to a Ph.D. in counseling psychology and sought pre-doc and post-doc training opportunities in the field that interested me: adult autism and ADHD assessment. My path is just one way, but there are many ways to become a competent assessor in the field of adult autism assessment. A current fear of mine, that is not unfounded, is that providers looking to re-specialize are unfamiliar with the complexities underlying adult autism and are not seeking the appropriate training to offer adequate and thorough autism assessments.
Perhaps the most current, and ultimate example, of the intersections of the above problems (scope creep, lack of standardization regarding definitions, lack of standardization regarding re-specialization) centers a very popular online autism and ADHD testing and evaluation service and repository of autism psychometrics, Embrace Autism. Embrace Autism was founded by two adult-diagnosed autistics, Natalie Engelbrecht and Eva Silvertant. In this section, I want to separate the educational side of Embrace Autism–it is a trove of information and support (although has also been criticized for inaccuracies)–and the evaluation and testing services. The latter have been the focus of much online criticism, particularly concerning the screening process, the assessment and testing process, and concerns of the service being a “diagnosis mill.” I would strongly encourage a thorough reading of the linked concerns but, for brevity, the writer alleges that Embrace Autism: a) does not engage in thorough screening, b) exclusively uses self-report measures for assessment of autistic traits, c) interacts very little with the client, and d) provides diagnoses that are not recognized by many geographic regions (including most U.S. states, the U.K., and some Canadian provinces).
In doing research for this post, what I learned was that the primary diagnostician for Embrace Autism, Natalie Engelbrecht, holds a doctorate in naturopathy and a license to practice in that field in Ontario (not a doctorate or license to practice psychology, as allegedly misrepresented elsewhere). There is one doctor of naturopathy program in Canada–the Canadian College of Naturopathic Medicine. Since 2015, naturopaths have been permitted to diagnose in the province of Ontario, from where Embrace Autism operates. Naturopathic medicine is a form of alternative medicine that has long been criticized for its weak empirical base, disconnect from predominant medical models, and purposeful misrepresentation as a field of medicine. A former naturopath referred to the field as “essentially witchcraft.” It is for these reasons that some countries do not license the practice of naturopathy at all (e.g., the U.K.), while others are divided: only around half of states in the U.S. recognize the field, and it’s only licenseable in five Canadian provinces. Because of these justifiable constraints and the limitation of naturopathic diagnoses, Embrace Autism allegedly has an option to pay to have your evaluation reviewed and “signed off” by a medical doctor who never interacts with the client–all for the low, low cost of around $2,000. I will add that this cost is prohibitive, but is not unique to Embrace Autism. In fact, my own out-of-network/out-of-pocket comprehensive evaluations cost the same amount (I write about my evaluation process near the end of this post) – that works out to around ~$150/hour for my comprehensive services.
The purview of this post is not to attack or discredit any one particular person (even if that professional has been formally reprimanded for practicing outside of the scope of their license and misrepresenting their credentials), but it is within the purview to discuss the unfortunate realities of standardization scarcity in adult autism diagnosis. I also don’t want any of these critiques to lead people who have sought testing and assessment services with Embrace Autism to doubt their own autistic identities–your identity is valid and yours. Embrace Autism is only one example of a seemingly-noble endeavor – increasing access to adult autism diagnostic services–wrapped in controversy surrounding standards of practice and money. In mental health fields, we have an adage: “see something, say something.” The notion is usually discussed in conversations of ethical dilemmas: if you see something unethical, you are implored by the nature of your license, and the ethical bodies to whom you answer, to say something. I have seen enough unscientific, evidence-lacking, money-grabbing adult autism assessment that I have to say something.
Staunchly missing from this conversation is the recognition of the complexity of adult autism diagnostics. Adult clients have layers upon layers of learned behavior, pathology, developed personality, and ingrained thinking patterns that we often do not contend with in the diagnosis of autism in young children. Giving an adult client a series of autism self-report measures is not enough to formulate a clear medical diagnosis: “Brief self-report measures do not have adequate specificity [to diagnose autism in adults]” (Lord et al., 2018, p. 5). Additionally, standardized structured and semi-structured interviews are not adequate diagnostic tools alone: “Our findings serve as a reminder that the ADOS-2 was not designed to be used as a standalone diagnostic measure” (Maddox et al., 2017, p. 7). Together, these tools also do not give the clinician the ability to rule-out diagnosis that commonly co-occur with autism: ADHD, learning differences, language needs, intellectual disability, personality disorders, or cognitive profile differences. About half of adults seeking diagnosis for autism for the first time had a lifetime prevalence of another co-occurring diagnosis (Pehlivanidis et al., 2020). When diagnosing autism in adults, we aren’t simply looking to “rule-in” autism, we’re looking to rule-out every other possible diagnosis while looking for positive evidence to confirm our diagnostic hypotheses. The systematic elimination of other diagnostic possibilities leads to the most accurate diagnosis, which is immensely important for client outcomes. The purpose of diagnosis is to guide intervention. Without a clear diagnosis, we don’t have a clear intervention. Here is an example taken from my own life:
When I was 14 years old, I spent some time in an intensive outpatient program after instances of self-harm. During this intensive experience, I was diagnosed with major depressive disorder. I lived with depression for much of my life, but medication never worked for me. I was in regular psychotherapy on-and-off, but I couldn’t fight the chronic fatigue, sadness, loss of direction, difficulty with activities of daily living (showering, brushing my teeth), frustration, irritability, amotivation, and apathy. It wasn’t until I began self-identifying as autistic that I came across the notion of autistic burnout, a type of cognitive and emotional overwhelm unique to autistic people living in unsupportive environments. Would you believe that when I began accommodating my autistic needs that my “depression” went away? Truthfully, I was likely never depressed – I was an undiagnosed autistic living in extreme burnout. I was treated for something I did not have which, obviously, did not work. Accurate diagnosis guides accurate intervention.
If you’ve made it this far, I can assume that you are intrinsically invested in the adult autism diagnosis process. Below, I share my comprehensive battery for assessing autism in verbally-fluent adults without co-occurring intellectual disability. This battery assists in my overarching goal of ruling-out as much as I am ruling-in (i.e., autism), by specifically targeting some of the most commonly co-occurring diagnoses: ADHD, OCD, Social Anxiety Disorder, and Social Pragmatic Communication Disorder. Many of these I can rule-out with my initial battery – some require additional testing, listed in point 5.
Screening Instruments
General Anxiety Disorder questionnaire (GAD-7)
Patient Health Questionnaire (PHQ-9)
Mood Disorder Questionnaire (MDQ)
PTSD Checklist (PCL-5)
Adult ADHD Self-Report Scale (ASRS v.1.1)
Yale-Brown Obsessive Compulsive Scale (Y-BOCS)
Ritvo Autism Asperger Diagnostic Scale-Revised (RAADS-R)
Autism Spectrum Quotient (AQ-10)
Adult Repetitive Behaviors Questionnaire (RBQ-2A)
Camouflaging Autistic Traits Questionnaire (CAT-Q)
Social Responsiveness Scale (SRS-2)
Interview:
Monteiro Interview Guidelines for Diagnosing the Autism Spectrum (MIGDAS-2)
Components of the Structured Clinical Interview for DSM-5 Disorders (SCID-5) depending on the results of symptom screeners
My own semi-structured historical interview
Cognitive tests
Behavior Rating Inventory of Executive Function (BRIEF-A), and/or
Comprehensive Executive Function Inventory (CEFI), and/or
Delis-Kaplan Executive Function System (D-KEFS)
I do not use the D-KEFS, but I know many evaluators who do–I have not been trained on it yet.
Personality tests
Personality Assessment Inventory (PAI)
Minnesota Multiphasic Personality Inventory (MMPI-3)
Contingent Assessments–based on the probability of co-occurring diagnoses and results of the above
Integrated Visual and Auditory Continuous Performance Task (IVA-2)
MOXO d-Continuous Performance Task (MOXO d-CPT)
Nelson-Denny Reading Test (NDRT)
Color Trails Test (CTT)
Conners Adult ADHD Rating Scale (CAARS)
Ruff 2-7 Selective Attention Test (Ruff 2-7)
Test of Memory Malingering (TOMM)
Dimensional Obsessive Compulsive Scale (DOCS)
Prodromal Questionnaire, Brief (PQ-B)
Integrated report
My integrated reports are usually 10 - 20 single-spaced pages.
The above battery has been refined to be specific to the population with whom I work. Your population may be different, so it may not be a 1:1 conversion for you. However, I have been adjacent to many professionals working with the same population I do, and their assessments are nowhere near as comprehensive as what I’ve listed here. Most recently, I had a concerned colleague with a client who sought external autism evaluation: a clinician charged their client $100, ostensibly gave them self-report autism measures, and diagnosed them with “mild Autism Spectrum Disorder.” This is not an evaluation; this is malpractice. As diagnosticians, we have to strive to know what something isn’t as much as we strive to know what it is. We are testing hypotheses, because we are scientists. Throwing inadequate screeners at a wall to see what sticks does a disservice to our clients and to our professions.
Autistic adults have spent much of their lives as the victims and survivors of systemic abuse, and the landscape of adult autism assessment is at a critical juncture. While increased awareness and accessibility to diagnosis represent significant progress – progress that I’ve benefited from – the field must confront its foundational issues before we undermine the very progress we’re celebrating. As clinicians, researchers, and members of the autistic community, we have an ethical and moral obligation to uphold rigorous standards that honor the complexity of autism in adults. What this means is: resisting the allure of simplified assessment protocols that sacrifice accuracy for speed, demanding appropriate training and specialization from providers, and centering the experiences of autistic people themselves. Our goal is not merely diagnosis, but a meaningful understanding that leads to appropriate support. If we can address these systemic challenges effectively, we can build an assessment infrastructure that truly serves autistic adults–one that acknowledges lived experiences while providing the diagnostic clarity needed to access appropriate accommodations and support. From this autistic psychologist: we have waited long enough for recognition; we should not have to settle for recognition that is incomplete or imprecise.
Comments