Addressing the Challenge of Under-Diagnosed Adult ADHD

There is a well-documented gap between the known prevalence of adult ADHD and rates of diagnosis and treatment. In Germany, epidemiological studies of nationally representative community samples have found prevalence rates ranging from 3.1% to 4.7%. Yet studies of publicly insured individuals age 18 to 69 years old report rates of diagnosed ADHD between 0.04% and 0.4%. So, even in a country with universal health insurance more than nine out of ten adults with ADHD go undiagnosed.

Many factors contribute to underdiagnosis: stigma, culturally influenced perceptions, and lack of motivation by those affected. Another crucial factor is the lack of recognition of ADHD symptoms by clinicians.

A research team surveyed 144 psychologists, 32 physicians, and two occupational therapists. Almost three in five participants were psychotherapists, a quarter were neuropsychologists, and one in seven were psychiatrists.

Four out of five clinicians stated they had received only “a few hours” of ADHD-specific training. One in four stated they had not examined guidelines for diagnosing ADHD. A lack of formal training among the vast majority and unfamiliarity with current diagnostic guidelines in a significant minority were surprising findings among clinicians who regularly work with adults with ADHD.

Many clinicians had difficulty identifying core features of adult ADHD as defined by the DSM-5 and International Classification of Diseases, Tenth Revision (ICD-10). Roughly one in five stated that hyperactivity had little relevance to adult ADHD. The only core feature correctly identified by more than half the respondents was having “difficulties concentrating.” Impairments in social behavior or aggression and memory impairment were not identified as being clearly “relevant” or “irrelevant” to adult ADHD.

The authors concluded, “these findings appear to indicate some uncertainty or at least a lack of consensus among clinicians about what symptoms are relevant to ADHD in adulthood and it is likely that this uncertainty contributes to diagnostic inaccuracy.”

Most respondents reported using self-report scales of ADHD symptoms and using unstructured interviews. While slightly more than half agreed that collateral reports are important to diagnosis, only about a third reported regularly using them. This is a problem given the limited accuracy of self-reported childhood symptoms for documenting the childhood-onset of the disorder. Semi-structured interviews are also known to improve the accuracy of diagnosis but are rarely used in clinical practice.

Over half of psychologists and a quarter of physicians reported using cognitive or neuropsychological testing, even though this is at variance with German (and other) guidelines, which specify that such testing is suitable for clarifying strengths and weaknesses, but not for ruling out or confirming a diagnosis of ADHD. The European Consensus Statement also states that cognitive/neuropsychological testing should only be used as a secondary or supplementary assessment tool.

While three out of four clinicians recommended stimulant drug treatment, psychologists tended to be more hesitant to do so. This is likely because German psychologists receive little training in pharmacotherapy, and do not have prescription privileges. Given the demonstrated efficacy of stimulant treatment, this points to a need to better educate psychologists in this regard.

Almost three in four respondents cited “lack of clinician knowledge and experience” as a barrier to ADHD diagnosis. Most clinicians also stated they were either “uncertain” or only “somewhat certain” of their ability to diagnose ADHD. That suggests that more extensive ADHD-specific training is needed.

A limitation of the survey was the relatively low participation by physicians. It is also likely that the findings are not reflective of practices in ADHD specialty clinics.

The authors concluded, “Further training is needed to improve clinicians’ understanding of ADHD in adulthood and to align diagnostic practices with guideline recommendations. Whereas discrepancies between respondents regarding the relative importance of peripheral symptoms (e.g., memory problems) were most common, a lack of consensus was found even for core symptoms listed by diagnostic criteria. Particularly among psychologists, improved awareness regarding the benefits of stimulant medications is needed to bring their treatment recommendations in line with evidence-based guidelines.”

Brooke C. Schneider, Daniel Schöttle, Birgit Hottenrott, Jürgen Gallinat, and Steffen Moritz, “Assessment of Adult ADHD in Clinical Practice: Four Letters—40 Opinions,” Journal of Attention Disorders (2019) DOI: 10.1177/1087054719879498.

How do undiagnosed but symptomatic adults compare with those diagnosed with ADHD?

The study team began with a representative sample of 69,972 U.S. adults aged 18 years or older who completed the 2012 and 2013 U.S. National Health and Wellness Survey. These adults were invited to complete the Validate Attitudes and Lifestyle Issues in Depression, ADHD and Troubles with Eating (VALIDATE) study, which included 1) a customized questionnaire designed to collect data on sociodemographic and clinical characteristics and lifestyle, and 2) several validated work productivity, daily functioning, self-esteem, and health-related quality of life (HRQoL) questionnaires. Of the 22,937 respondents, 444 had been previously diagnosed with ADHD, and 1,055 reported ADHD-like symptoms but had no previous clinical diagnosis.

There were no significant differences between the two groups in terms of age, education, income, health insurance, and most comorbid disorders. But those who had not been previously diagnosed were significantly more likely to be first-generation Americans (p<.001), nonwhite (p<.001), unemployed (p=.024), or suffer from depression, insomnia, or hypertension.

After matching the two groups for sociodemographic characteristics and comorbid conditions, covariate comparisons were made between 436 respondents diagnosed with ADHD and 867 previously undiagnosed respondents. Among respondents who were employed, diagnosed individuals registered a mean work productivity loss of 29% as opposed to 49% for the previously undiagnosed (p<.001). They also registered a 37% level of activity impairment versus a 53% level among the undiagnosed (p<.001). On the Sheehan Disability Scale, which ranges from 0 (no impairment) to 30 (highly impaired), the diagnosed group had a mean of 10, as opposed to a mean of 15 for the undiagnosed (p<.001). Diagnosed respondents also significantly outperformed undiagnosed ones on the Rosenberg Self-Esteem Scale (19 versus 15, on a scale of 0 to 30, p<.001), and on two quality-of-life scales (p<.001).

Applying a linear regression mixed model to the matched sets, the diagnosed still scored 16 points better than the undiagnosed on the WPAI:GH Productivity Loss scale (p<.001), 14 points better on the WPAI:GH Activity Impairment scale (p<.001), 4.5 points better on the Sheehan Disability Scale (p<.001), almost 4 points on the Rosenberg Self-Esteem Scale (p<.0001), with comparable gains on the two quality-of-life scales (p<.001 and p<.0001).

The authors concluded, “This comparison revealed that individuals who had been diagnosed with ADHD were more likely to experience better functioning, HRQoL , and self-esteem than those with symptomatic ADHD. This result appears to be robust, withstanding several levels of increasingly rigorous statistical adjustment.” That points to substantial benefits from the treatment that follows diagnosis of adult ADHD.


Manjiri Pawaskar, Moshe Fridman, Regina Grebla, and Manisha Madhoo, “Comparison of Quality of Life, Productivity, Functioning and Self-Esteem in Adults Diagnosed With ADHD and With Symptomatic ADH,” Journal of Attention Disorders, Published online May 2, 2019

The Relationship Between Executive Function Deficits and DSM-5-Defined ADHD Symptoms

Michael J. Silverstein , Stephen V. Faraone, Terry L. Leon, Joseph Biederman, Thomas J. Spencer, and Lenard A. Adler

Journal of Attention Disorders. 1–11: 2018. DOI: 10.1177/1087054718804347

The Diagnostic and Statistical Manual of Mental Disorders (5th ed.) (DSM-5) still defines ADHD symptoms in terms nine inattentive (IA) and nine hyperactive-impulsive (H-I) symptoms, to form the core eighteen symptoms of the disorder; this is in spite of a large literature that indicates that higher level symptoms of organization, planning and prioritization known as Executive Function Deficits (EFDs) common co-travel with symptoms of ADHD and are highly impairing to adults with ADHD. The investigators examined the relationship of core ADHD IA and HI symptoms and EFDs and the predictive utility of the Adult ADHD Investigator Symptom Rating Scale (AISRS) in identifying those with adult ADHD and Executive Dysfunction (ED). The AISRS is a clinician-administered, severity based (0-3), semistructured interview, containing adult ADHD specific prompts, developed to evaluate ADHD symptoms at baseline and during treatment. The Adult ADHD Self-Report Scale (ASRS) Symptom Checklist was also administered. Both the AISRS and ASRS Symptom Checklist were expanded to not only include the 18 core DSM-5 symptoms of ADHD, but also nine additional symptoms of EFD and four symptoms of Emotional Dyscontrol (EC). Executive Function was also assessed via the BRIEF-A, a well-normed scale to assess EF, with patients with global executive complex (GEC) T scores T >= 65 (1.5 standard deviations above the mean, 93 percentile) being indicative of ED. Subjects were recruited from referrals to a university adult ADHD program or a primary care clinical practice; 297 subjects participated (171 with adult ADHD). (IA) and (H-I) symptoms on the AISRS and ASRS Symptom Checklist were moderately to strongly correlated with and highly predictive of EFDs (with correlations being stronger for IA symptoms). Receiver operating characteristic curve analysis showed that an AISRS DSM 18-item score of ⩾ = 28 was most predictive of clinical ED. This study is to clinicians because it highlights the importance of assessing EFDs in addition to core symptoms of IA and HI when evaluating patients with adult ADHD.

Self vs Clinician Rating Scales

Silverstein et al. (2017) recently published a study which reported a validation of two expanded adult ADHD scales, self-report (Adult ADHD Self Report Rating Scale (ASRS) v1.1 Symptom Checklist) and clinician/investigator (Adult ADHD Investigator Symptom Rating Scale – AISRS); both scale were expanded beyond the classic 18 DSM symptoms of inattention (IA) and hyperactivity-impulsivity (HI) to include 13 additional commonly co-traveling items of executive function deficits (EFDs) (eg. difficulty with organization, planning, task execution and procrastination) and emotional control deficits (EC) (eg. moodiness, over-reactivity of mood), creating 31 item scales.

Data was examined on 297 individuals (either referred adults with ADHD or individuals in a primary care practice, who were mostly controls) at the NYU School of Medicine. The psychometric properties of both scales were examined in terms of internal consistency (Cronbach’s alpha) and cut-off’s from normative data to predict a diagnosis. The internal consistency was high on both scales for measuring not only DSM IA and HI symptoms but also symptoms of EFD and EC (subscales ranging from 0.84 to 0.96); however, the internal consistency for EC was less for EFD, indicating that EC symptoms are less homogeneous than those of EFD, as had been suggested in a factor analysis by Adler et al (2017). Cut-off scores predictive of adult ADHD on the AISRS for dsm IA and HI 18 symptoms were between 23 and 26; this validates the empirical choice utilized in many clinical trials in adult ADHD of a cut-off score of 24.

This study highlights the importance to clinicians of assessing not only DSM symptoms of IA and HI, but also the co-traveling symptoms of EFD and EC and that clinician or self-report scales can be utilized in clinical practice. Also, the assessment of EFD and EC is especially important for clinicians as these symptom sets are less robustly responsive to pharmacotherapy than classic symptoms of IA and HI and may be more amenable to the addition of psycho-social interventions, such as cognitive behavioral therapy.

Adler LA, Faraone SV, Spencer TJ, Berglund P, Alperin S, Kessler RC. The structure of adult ADHD. Int J Methods Psychiatr Res. 2017 Mar;26(1). doi: 10.1002/mpr.1555. Epub 2017 Feb 17

Silverstein MJ, Faraone SV, Alperin S, Leon TL, Biederman J, Spencer TJ, Adler LA. Validation of the Expanded Versions of the Adult ADHD Self-Report Scale v1.1 Symptom Checklist and the Adult ADHD Investigator Symptom Rating Scale. J Atten Disord. 2018 Feb 1:1087054718756198. doi: 10.1177/1087054718756198.

ADHD Screener

Ustun et al. (2017) recently published an updated version of the adult ADHD screener which is validated for DSM-5: the ASRS v1.1 Screener: DSM-5. The prior DSM-IV version of the screener was established using two populations: a community-based sample from the National Co-Morbidity Survey (NCS-R) and a sample of individuals from a health care plan.

The first step was to recalibrate the new screener using these same two samples, but applying updated DSM-5 criteria; symptoms included not only core symptoms of inattention (IA) and hyperactivity-impulsivity (HI) as defined in DSM, but additional co-traveling symptoms of executive dysfunction (eg: deficits in organization, planning, working memory) or emotional dysregulation (eg: over emotionality, changeable mood).

The symptoms of executive dysfunction have been shown to carry a high symptom burden and in many ways drive the symptom presentation when present in a recent factor analysis (Adler et al. 2017). The selection and weighting of the symptoms was selected by SLIM artificial intelligence – six items were selected: four were from DSM classic symptoms of IA and HI, but two were symptoms of executive dysfunction beyond those defined in the DSM. The process was again repeated and validated in a new sample of referred individuals for ADHD evaluations and controls from primary care practices from the NYU School of Medicine as second validation. The screener is again self-report and rated on a frequency basis of 0-4 (never to very often), with a cut-off score of > = 14 indicating a positive screen. The weighting of items in the screener is not evenly distributed and the scoring algorithm will shortly be available through an educational program on this website.

The ASRS v1.1 Sceener: DSM-5 has a high degree of sensitivity and specificity (first sample: 91.4%; 96.0%, respectively; second NYU sample: 91.9%, 74.0%, respectively). Given the high sensitivity and specificity, the new screener can be a highly effective tool for clinicians to identify individuals at risk for adult ADHD who merit further evaluation and a full diagnostic evaluation.

Adler LA, Faraone SV, Spencer TJ, Berglund P, Alperin S, Kessler RC. The structure of adult ADHD. Int J Methods Psychiatr Res. 2017 Mar;26(1). doi: 10.1002/mpr.1555. Epub 2017 Feb 17.

Ustun B, Adler LA, Rudin C, Faraone SV, Spencer TJ, Berglund P, Gruber MJ, Kessler RC. The World Health Organization Adult Attention-Deficit/Hyperactivity Disorder Self-Report Screening Scale for DSM-5. JAMA Psychiatry. 2017 May 1;74(5):520-526. doi: 10.1001/jamapsychiatry.2017.0298.