ADHD in Adults - Persistence of Childhood ADHD

Completely revised 10/2026

 

Author: Ulrich Brennecke
Review: Waldemar Zdero, M.A. in Psychology

ADHD is not just a childhood disorder. For about two-thirds of those diagnosed as children, ADHD persists into adulthood. The symptoms often change over time. In a survey conducted in 20 countries, ADHD was actually slightly more common among 18- to 44-year-olds than the prevalence reported retrospectively for childhood.

When ADHD goes into remission, this usually does not mean that the symptoms have disappeared. It simply means that the symptoms are no longer sufficient to meet the diagnostic criteria. In a large long-term study, up to 90% of people with ADHD still had residual symptoms or significant limitations in young adulthood. Disadvantages often persist in school, vocational training, and the workplace. For many people, ADHD follows a fluctuating course. Symptoms subside and then return later. Stress and the changing demands of daily life are likely to contribute to this. In studies, the persistence of ADHD into adulthood was associated, among other factors, with severe ADHD in childhood and with additional disorders such as social behavior disorders or depression.

Some people are not diagnosed until adolescence or adulthood (late-onset ADHD). In most cases, symptoms were already present in childhood but were overlooked or were still mild. Whether ADHD can develop for the first time in childhood without any prior signs is a matter of debate. After careful evaluation, over 90% of the seemingly late-onset cases were ruled out. If ADHD is not recognized until after childhood, it usually becomes apparent between the ages of 12 and 16. The lack of evidence of symptoms in childhood should not be a reason to deny treatment to adults.

In women, ADHD is more often diagnosed in adulthood. In one study, women were diagnosed on average about 4 years later than men. Women with a late diagnosis often also have anxiety disorders or depression. One possible reason for the delayed onset is the hormone estradiol, which can mask symptoms and declines with age, especially during menopause.

In adults, visible physical hyperactivity usually decreases and often transforms into inner restlessness. Inattention tends to persist more often than hyperactivity and impulsivity. This is often accompanied by problems with organization, emotional overreactions, and mood swings. Adults are generally better able to structure their lives in ways that minimize the impact of ADHD symptoms.

The recommended daily doses of stimulants for adults differ barely from those for children. Regardless of age, treatment should begin with a low dose and be gradually increased. In a large meta-analysis of numerous studies, ADHD medications had a slightly weaker effect on core symptoms in adults than in children and adolescents. Many people with ADHD discontinued their medication as young adults, most commonly between the ages of 18 and 19.

ADHD can be diagnosed well into old age. In older adults, ADHD can be mistaken for early-stage dementia. In one study, older adults with ADHD had a quality of life that was similarly impaired to that of younger people with the condition. Medication may also be appropriate for seniors. Treatment should begin with a low dose and be gradually increased. Care must be taken to account for cardiovascular diseases, interactions with other medications, and changes in how the body metabolizes the active ingredients that occur with age. Lisdexamfetamine is approved for adults with no upper age limit. Methylphenidate can only be used in older adults off-label. According to ADxS, a blanket rejection of stimulants solely on the basis of age is not justified.

ADHD does not, as was believed until the end of the last millennium, (E 4)¹ automatically end when a person reaches adulthood.

ADHD persists into adulthood in about two-thirds of the people with ADHD who were diagnosed as children. Many experience changes in their symptoms, while others may experience complete remission or remission with fluctuating phases. Up to 90% of people with ADHD still have residual symptoms or significant limitations in young adulthood compared to people without ADHD. (Cohort study, n = 558, 16 years, E 2b.)² Various reasons for the persistence of ADHD into adulthood are discussed, such as a low socioeconomic background, high stress levels during childhood, or late-onset trauma. One theory posits that ADHD is already present in childhood but is masked by coping mechanisms and only becomes apparent in adulthood. Late-onset ADHD—that is, a first occurrence (or at least first recognition) after childhood—is a topic of discussion and appears to occur particularly among women. However, after applying differential diagnostic procedures, over 90% of apparent late-onset cases are ruled out. Most (if not all) late-onset cases occur between the ages of 12 and 16. (Narrative review, E 4.)³
The symptoms of ADHD differ in adults compared to children, with hyperactivity decreasing and inner restlessness, inattention, and organizational problems becoming more prominent.

(E 4): ADHD can be diagnosed even in old age. (Narrative review, E 4)³ (Narrative review, E 4)⁴ For older adults (age 50 and older), it is recommended to start ADHD medication at a low initial dose and increase it gradually. (Systematic review, E 4.)⁵

1. Prevalence of ADHD in Adults¶

In Germany, 4.8% of children aged 3 to 17 had a medical or psychological diagnosis of ADHD (according to parents), and ADHD was suspected in an additional 4.9%. (Cross-sectional study, n = 14,836, E 3.)⁶

(E 3): The prevalence of ADHD among adults was:

• 6.76% (95% CI 4.31 to 10.61) for symptomatic ADHD (regardless of whether it began in childhood), standardized to the 2020 global population (meta-analysis, k = 40, E 3.)⁷

• 1% to 4.7% in four population-based studies, depending on the diagnostic criteria (systematic review; full text not available to us; E 3.)⁸

• 4.5% at age 22 (Brazilian birth cohort born in 1993, n = 3,780, DSM-5, M.I.N.I.) (birth cohorts, n = 7,354, E 2b.)⁹

• 4.4% at age 30 (Brazilian birth cohort born in 1982, n = 3,574, DSM-5, M.I.N.I.) (birth cohorts, n = 7,354, E 2b.)⁹

• 4.4% among 18- to 44-year-olds in the U.S. (cross-sectional study, n = 3,199, partially funded by manufacturers, E 3.)¹⁰

• 4.4% as the best available estimate (Narrative Review, secondary citation from Polanczyk & Rohde 2007, E 4.)¹¹ For children and adolescents, the review cited 5.29%

• 2.8% among 18- to 44-year-olds in 20 countries (WHO survey), higher than the prevalence reported retrospectively for childhood (2.2%) (cross-sectional study, k = 20 surveys, n = 26,744, E 3.)¹² This could be consistent with the possibility that the heritability of adult ADHD is lower than that of childhood ADHD, meaning that the role of environmental factors in the development of ADHD may be greater in adults than in children. (Narrative review of twin studies, E 4)¹³

• 2.58% (95% CI 1.51 to 4.45) for persistent ADHD (onset in childhood), standardized to the 2020 world population (meta-analysis, k = 40, E 3.)⁷

• 1% to 2.5%, calculated based on a persistence rate of 35% to 50% and a childhood prevalence of 3% to 5% (Narrative Review, E 4.)¹⁴

Further studies can be found in Krause. (E 4)¹⁵

As early as 1905, Gustav Specht and, in 1910, Hermann Paul Nitsche described adults with core ADHD symptoms using the terms “chronic mania” and “chronic manic states,” respectively. Specht viewed this as a “completely independent mental illness.” Adult ADHD is therefore not a construct of modern psychiatry. (Historical Analysis, E 4.)¹⁶ Neurologists of the early 20th century used terms such as “chronic mania” to describe patients who, in all likelihood, would be diagnosed with adult ADHD today. Their concepts show clear parallels to today’s diagnostic criteria. (Historical Analysis, E 4.)¹⁷

Medical descriptions of ADHD as a whole date back much further. For more on this, see Historical Descriptions and Names of ADHD

Of 5,593 15-year-olds insured by AOK who had been diagnosed with ADHD, only 31.2% still had an ADHD diagnosis by age 21. The proportion taking ADHD medication fell from 51.8% to 6.6% during the same period. (Cohort study, n = 5,593, 6 years, E 2b.)¹⁸ This describes diagnostic (administrative) persistence (coding in billing data), not true persistence. The authors interpret the sharp decline as an indication of care-related problems during the transition.

Longitudinal studies largely support the vulnerability hypothesis, according to which mental disorders that begin in childhood and adolescence (including ADHD, anxiety disorders, depression, and social behavior disorders) are often precursors to mental disorders in adulthood. (Narrative review; full text not available to us; E 4.)¹⁹

2. Patterns of ADHD Persistence into Adulthood¶

It is by far more the rule than the exception that ADHD present in childhood persists into adulthood.

(E 2b): The reported rates of ADHD persistence from childhood into adulthood varied considerably. Systematic reviews cited:

• 60% to 86% for persistent symptoms in 7 North American controlled prospective follow-up studies (Montreal, New York, Milwaukee, Pittsburgh, MGH, Berkeley, MTA) (Narrative review, k = 7, E 4.)²⁰
• 4% to 80% (Narrative review, secondary citation from Polanczyk & Rohde 2007, E 4.)¹¹
• 4% to 77% (41 estimates from 12 longitudinal samples) (Systematic review, k = 12, E 2b.)²¹
• 5.7% to 77% for the full-body scan in the same 7 follow-up studies (narrative review, k = 7, E 4.)²⁰
• 5% to 66% in prospective follow-up studies (Narrative Review, E 4.)²²
• 4% to 66% for symptom persistence in follow-up studies (systematic review; full text not available to us; E 3.)⁸
• up to 50% (Narrative Review, E 4.)¹⁴
• 40% to 50% in the few estimates that combined self- and peer-ratings, required the presence of impairment, and used an age-appropriate symptom threshold (Systematic Review, k = 12, E 2b.)²¹
• 30% to 50%, primarily persistence of attention deficit (Narrative Review, E 4.)²³ Many symptoms classified as inattention are actually impairments in executive functions.

The impairment associated with ADHD persists into adolescence in 85% of people with ADHD and into adulthood in 50% to 70%. The DSM-IV criteria are significantly better suited to children than to adolescents and adults. (Narrative review; full text not available to us; E 4.)²⁴

In a population-based study, the cumulative incidence among individuals aged 5 to 19 years ranged from 7.5% to 16%. (Systematic review; full text not available to us; E 3.)⁸

The persistence and remission rates found depended on:

• Informants (self-assessment / parental assessment / combined assessment) (Systematic review, k = 12, E 2b.)²¹ (Cohort study, N = 321, n = 200, E 2b.)²⁵ (Cohort study, N = 218, full text not available to us, E 2b.)²⁶ (Narrative review, k = 7, E 4.)²⁰ (Narrative review, E 4.)²² (Narrative review, E 4.)²⁷ (Narrative review, secondary citation of Polanczyk & Rohde 2007, E 4.)¹¹ (Commentary, E 4.)²⁸
• Diagnostic criteria: While many older longitudinal studies recruited children based on DSM-II, they used the current DSM version (DSM-III, DSM-III-R, or DSM-IV) during follow-up, meaning that remission rates may have been influenced by changes in diagnostic criteria rather than solely by changes in symptoms (literature review, k = 7 follow-up studies, industry-funded, E 4.)²⁹ (Cohort study, N = 218, full text not available to us, E 2b.)²⁶ (Narrative review, k = 7, E 4.)²⁰ (Narrative review, E 4.)²² (Narrative review, E 4.)²⁷ (Narrative review, secondary citation of Polanczyk & Rohde 2007, E 4.)¹¹
• Symptom threshold (DSM or normative values, age-specific thresholds) (Systematic review, k = 12, E 2b.)²¹ (Cohort study, N = 321, n = 200, E 2b.)²⁵ (Narrative review, E 4.)²² (Commentary, E 4.)²⁸
• Criterion of impairment (systematic review, k = 12, E 2b.)²¹ (cohort study, N = 321, n = 200, E 2b.)²⁵
• Data collection method (questionnaire or interview, instrument) (Systematic review, k = 12, E 2b.)²¹ (Comment, E 4.)²⁸
• Recruitment and sampling characteristics (e.g., exclusion of aggressive children, comorbidity) (Narrative review, E 4.)²² (Narrative review, E 4.)²⁷
• Definition of remission (narrative review, secondary citation from Polanczyk & Rohde 2007, E 4.)¹¹
• Selective attrition (rate of dropouts) (Narrative review, E 4.)²²

(E 2b): Self-reports yielded significantly lower rates than parental or combined assessments. (Cohort study, N = 321, n = 200, E 2b.)²⁵ (Cohort study, N = 218, full text not available to us, E 2b.)²⁶ Exclusively self-reports and a strict threshold of 6 DSM symptoms resulted in very low estimates. (Systematic review, k = 12, E 2b.)²¹ Earlier follow-up studies based on self-reports likely significantly underestimated persistence. (Cohort study, N = 218, full text not available to us, E 2b.)²⁶ People with ADHD underreported their current symptoms, while people without ADHD overreported them. Parental assessment was more diagnostically sensitive than self-assessment; an additional self-assessment did not provide any significant added benefit. (Cohort study, N = 321, n = 200, E 2b.)²⁵ Parental reports were more closely associated with important areas of life than self-reports. (Cohort study, N = 218, full text not available to us, E 2b.)²⁶

The recommendations included third-party assessments, a lower symptom threshold, and the consideration of impairment (cohort study, N = 321, n = 200, E 2b.)²⁵; and for studies on persistence, remission, and late-onset ADHD, thorough assessment in adulthood that combines self-reports and parental reports. (Commentary, E 4.)²⁸

(E 2b): Depending on the informant and criterion, ADHD persisted in individual cohorts among:

• 65% to 86% did not remit (Milwaukee, age 27), secondary citation from Cherkasova et al. (narrative review, k = 7, E 4.)²⁰
• 77% with full-spectrum or subsyndromal ADHD (MGH, age 22), secondary citation from Cherkasova et al. (narrative review, k = 7, E 4.)²⁰
• 76.6% according to DSM-5 (Berkeley), secondary citation via Cherkasova et al. (narrative review, k = 7, E 4.)²⁰
• 75.6% with increased symptom severity based on the combined assessment (Pittsburgh/PALS, mean age 20.2 years) (cohort study, N = 321, n = 200, E 2b.)²⁵
• 71.8% with increased symptom severity as assessed by parents (Pittsburgh/PALS, mean age 20.2 years) (cohort study, N = 321, n = 200, E 2b.)²⁵
• 66% based on parental reports and developmental criteria (above 2 SD), controls 8% (Milwaukee, ages 19 to 25) (cohort study, N = 218, full text not available to us, E 2b.)²⁶
• 60.4% with clinically significant impairment based on the combined assessment (Pittsburgh/PALS, mean age 20.2 years) (cohort study, N = 321, n = 200, E 2b.)²⁵
• 57% according to DSM-IV (Berkeley), secondary citation via Cherkasova et al. (narrative review, k = 7, E 4.)²⁰
• 55.6% with clinically significant impairment as assessed by parents (Pittsburgh/PALS, mean age 20.2 years) (cohort study, N = 321, n = 200, E 2b.)²⁵
• 49.9% according to DSM-5 (MTA, age 25), secondary citation from Cherkasova et al. (narrative review, k = 7, E 4.)²⁰
• 46% based on parental reports and DSM criteria; controls: 1.4% (Milwaukee, ages 19 to 25) (cohort study, N = 218; full text not available to us; E 2b.)²⁶
• 41.5% according to DSM-III in late adolescence (New York), secondary citation from Cherkasova et al. (narrative review, k = 7, E 4.)²⁰
• 24.4% with self-reported increased symptom severity (Pittsburgh/PALS, mean age 20.2 years) (cohort study, N = 321, n = 200, E 2b.)²⁵
• 22.2% in middle adulthood (New York), secondary citation from Cherkasova et al. (narrative review, k = 7, E 4.)²⁰
• 19.7% met the DSM-IV-TR symptom threshold in the combined assessment (Pittsburgh/PALS, mean age 20.2 years) (cohort study, N = 321, n = 200, E 2b.)²⁵
• 14.6% with clinically significant impairment in self-assessment (Pittsburgh/PALS, mean age 20.2 years) (cohort study, N = 321, n = 200, E 2b.)²⁵
• 12.0% met the DSM-IV-TR symptom threshold as assessed by parental reports (Pittsburgh/PALS, mean age 20.2 years) (cohort study, N = 321, n = 200, E 2b.)²⁵
• 12% based on self-report and development-related criteria (above 2 SD), controls 10% (Milwaukee, ages 19 to 25) (cohort study, N = 218, full text not available to us, E 2b.)²⁶
• 9.6% met the DSM-IV-TR symptom threshold based on self-report (Pittsburgh/PALS, mean age 20.2 years) (cohort study, N = 321, n = 200, E 2b.)²⁵
• 5.7% according to DSM-III-R in early adulthood, self-report only (New York), secondary citation from Cherkasova et al. (narrative review, k = 7, E 4.)²⁰
• 5% based on self-reports and DSM criteria; controls: 0% (Milwaukee, ages 19 to 25) (cohort study, N = 218; full text not available to us; E 2b.)²⁶

Remission was equally dependent on its definition. Among boys with ADHD aged 18 to 20, it was:

• Syndromic remission (no longer meets the full diagnostic criteria) in over 60% of patients (cohort study, n = 128, 4 years, E 2b.)³⁰
• functional remission (fewer than 5 symptoms and no impairment) 10% (cohort study, n = 128, 4 years, E 2b.)³⁰

(E 4): The most consistent predictors of poor functional outcomes in adulthood were:

• Comorbidity, particularly social behavior disorders (narrative review, k = 7, E 4.)²⁰ (narrative review, E 4.)²⁷
• Persistence of ADHD (Narrative Review, k = 7, E 4.)²⁰
• Severity of ADHD in Childhood (Narrative Review, E 4.)²⁷

A review identified ADHD severity, a comorbid social behavior disorder, major depression, and ADHD treatment as the main predictors of persistence. (Narrative review, E 4.)²⁷

An early meta-analysis of 9 prospective studies (follow-up 4 to 16 years after diagnosis) described an exponential decline in ADHD rates, which roughly halved every 5 years. Assuming a childhood prevalence of 4%, this would extrapolate to approximately 0.8% at age 20 and 0.05% at age 40. (Meta-analysis, k = 9, E 2b.)³¹ This projection was based solely on the loss of the full-blown syndrome according to criteria at the time and relied predominantly on self-reports. It was refuted by later studies, which found significantly higher persistence rates using third-party assessments, age-appropriate thresholds, and the inclusion of partial remission.

According to a review of follow-up studies, the adult outcomes of children with ADHD fell into three groups: (Narrative review, E 4.)³²

• Persistent significant problems with concentration, impulsivity, and social interaction (about 50 to 60 percent)
• largely unremarkable functional level, no different from that of controls (about 30%)
• significant psychiatric and/or antisocial problems (about 10 to 15 percent)

2.1. Complete continuity into adulthood¶

(E 2b): According to systematic reviews, the persistence of ADHD (definition and follow-up period vary by study), possibly with altered symptoms, was found in:

• 50% to 75% of children diagnosed with ADHD (narrative review, E 4.)³³

• two-thirds to three-quarters with problematic ADHD symptoms in early and middle adolescence (New York and Montreal cohorts) (Narrative review, E 4.)³⁴

• Up to two-thirds continue to experience ADHD symptoms even as adults (narrative review, E 4.)³⁵

• (E 3): approximately 66% (consensus, E 4)³⁶ (genome-wide association study, n = 77,914, figure from the introduction, E 3.)³⁷

• 65% at age 25 (complete or partial remission) (meta-analysis, k = 21, E 2b.)³⁸

• about 60% (Narrative Review, E 4.)³⁹

• 55% in prospective studies published after 2011 (from childhood or adolescence through adulthood) (Systematic review, k = 39, of which 27 were quantitatively analyzed; full text not available to us, E 2b.)⁴⁰

• 43% on average across all prospective studies (from childhood or adolescence through adulthood) (systematic review, k = 39, of which 27 were quantitatively analyzed; full text not available to us, E 2b.)⁴⁰

• approximately two-fifths with clinically significant ADHD symptoms persisting into late adolescence (New York and Montreal cohorts) (Narrative review, E 4.)³⁴

• 34% in prospective studies from 1985 to 2011 (systematic review, k = 39, of which 27 were quantitatively analyzed; full text not available to us; E 2b.)⁴⁰

• about one-third meet all diagnostic criteria by age 18, with a further age-related decline through their mid-20s (Narrative Review, E 4.)³⁵

• at least one-third with symptoms in adulthood (estimate based on longitudinal studies) (Narrative review, E 4.)⁴¹

• 32% full-body (weighted average from follow-up studies) (literature review, n = 7, manufacturer-funded, E 4.)⁴²

• 15% complete persistence (full-field) at age 25 (meta-analysis, k = 21, E 2b.)³⁸

(E 2b): In individual cohorts, persistence was observed in:

• 86.5% continued to meet DSM-5 criteria 6 years later (ADHD combined type, mean age 11.4 years at the start of the study, 17.4 years at follow-up) (Cohort study, n = 347, 6 years, E 2b.)⁴³ An additional 8.4% met subthreshold criteria, 5.1% were in remission, and 48.5% continued to experience functional impairment.

• 83.3% during adolescence based on a norm-referenced cutoff score (cohort study, N = 189, n = 123, 8 years, E 2b.)⁴⁴

• 79% full-spectrum (DSM-IV, parent interview) during adolescence and young adulthood (median age 18.8 years) among individuals diagnosed with the combined subtype in childhood (cohort study, n = 116, 6.6 years, E 2b.)⁴⁵

• 78% when including subsyndromic, functional, and medication-treated persistence (boys) (cohort study, N = 215, n = 110, 10 years, E 2b.)⁴⁶

• 77.1% when including subsyndromic, functional, and medication-treated persistence (girls, mean age 22 years) (Cohort study, N = 187, n = 96, 11 years, industry-funded, E 2b.)⁴⁷

• 75.6% continued to experience disruptive ADHD symptoms 6 years after diagnosis at ages 7 to 10 (Turkey) (cohort study, n = 73, 6 years, E 2b.)⁴⁸

• 71.5% during adolescence according to DSM-III-R (cohort study, N = 189, n = 123, 8 years, E 2b.)⁴⁴

• 71% (95% CI 61–79%) with full-spectrum ADHD or residual criteria with impairment (girls, up to age 16) (Cohort study, n = 123 with ADHD, n = 106 controls, 5 years, E 2b.)⁴⁹ Even girls whose symptoms had remitted remained functionally impaired compared to controls. More severe behavioral impairment at the start of the study predicted persistence of symptoms.

• 66.6% still had an ADHD diagnosis 6 years after being diagnosed at ages 7 to 10 (Turkey) (cohort study, n = 73, 6 years, E 2b.)⁴⁸ 24.4% in complete remission

• 65.8% with full-blown or subclinical ADHD 4 years after diagnosis between the ages of 6 and 12 (French population sample) (cohort study, N = 1,012, n = 875, 4 years, full text not available to us, E 2b.)⁵⁰

• 65.8% full-spectrum or subsyndromal (mean age 19 years, China) (cohort study, n = 193, 6.8 years, E 2b.)⁵¹

• 65.6% (combined subtype, 12 years after diagnosis, DSM-5 symptom threshold) (follow-up study, N = 115, n = 61, 12 years, E 2b.)⁵²

• 61.4% based on developmental criteria (4 hyperactivity-impulsivity and 5 inattention symptoms) (girls, mean age 19.6 years) (Cohort study, n = 140 with ADHD, n = 88 controls, approximately 10 years, E 2b.)⁵³

• 58% still had the diagnosis 25 years later (11 out of 19 male adolescents with ADHD) (cohort study, n = 19 with ADHD, 26 controls, 25 years; full text not available to us, E 2b.)⁵⁴

• 57% in the maternal assessment using a norm-referenced cutoff score (women with childhood ADHD, mean age 20 years) (cohort study, N = 58, n = 34, E 2b.)⁵⁵

• 55.9% overall rate based on the DSM-IV threshold (71 out of 127) (girls, mean age 19.6 years) (cohort study, n = 140 with ADHD, n = 88 controls, approximately 10 years, E 2b.)⁵³

• 39.7% to 55.9% (DSM-5, depending on the time of assessment, up to a mean age of 25 years) (cohort study, n = 558, 16 years, E 2b.)²

• 53% in the last two examinations (children with predominantly elevated manic symptoms; mean age at the end of the study: 16.3 years) (cohort study, n = 431, 8 years, E 2b.)⁵⁶

• 49.5% full-spectrum (DSM-IV) between the ages of 11 and 16 (average age at diagnosis: 7.3 years) (cohort study, n = 186, E 2b.)⁵⁷ An additional 33.3% were subthreshold (more than half of the required symptoms), and 17.2% had only a few symptoms remaining

• 49% with a continuous course (children with predominantly elevated manic symptoms; mean age at the end of the study: 16.3 years) (cohort study, n = 431, 8 years, E 2b.)⁵⁶

• 45% 18 years after the childhood diagnosis (cohort study, n = 421, of whom 154 had childhood ADHD, 18 years, E 2b.)⁵⁸ Persistent ADHD was associated with major depression, internalizing problems, stress, and poorer sleep.

• 44% based on parental assessment 8 years after a school screening during elementary school (ethnically diverse U.S. community sample; 15% of children with subclinical symptoms, 2% in the comparison group) (Case-control study, n = 94 with ADHD, n = 75 subthreshold, n = 163 controls, 8 years, E 3.)⁵⁹ Gender, ethnicity, and poverty did not moderate the results.

• 44% with symptoms above the 93rd percentile plus impairment (Milwaukee cohort, mean age 27 years) (cohort study, N = 202, approximately 20 years, E 2b.)⁶⁰

• 44% met all DSM-IV criteria according to parental assessment; controls 1%; ϕ = 0.47 (girls, mean age 19.6 years) (Cohort study, n = 140 with ADHD, n = 88 controls, approximately 10 years, E 2b.)⁵³

• 43% persistent ADHD, controls 4% (replication cohort of hyperactive boys, mean age 18 years) (cohort study, N = 172, n = 94, ages 8 to 14, E 2b.)⁶¹

• 41% on the maternal assessment (DSM symptom count) (women with childhood ADHD, mean age 20 years) (cohort study, N = 58, n = 34, E 2b.)⁵⁵

• 40% with any ADD diagnosis or including individual symptom components or combinations of two symptoms; controls: 3% (hyperactive boys, aged 6 to 12 at diagnosis, 16 to 23 at follow-up) (Cohort study, N = 201, n = 101, approximately 9 years, E 2b.)⁶² (Cohort study, n = 100, 9 years, secondary citation from Remschmidt & Herpertz-Dahlmann 1989, E 2b.)⁶³

• 39% with a clinical symptom threshold based on an interview or self-assessment plus functional impairment (mean age 21.4 years, clinical sample) (cohort study, N = 85, n = 59, 10 years, E 2b.)⁶⁴

• approximately 36% with at least one moderately or severely impairing ADHD symptom; controls: 2% (Montreal, mid-20s) (narrative review, E 4.)³⁴

• 35% full-spectrum (DSM-IV) (boys) (cohort study, N = 215, n = 110, 10 years, E 2b.)⁴⁶

• 33.3% full-screen (DSM-IV) (girls, mean age 22 years) (cohort study, N = 187, n = 96, 11 years, manufacturer-funded, E 2b.)⁴⁷

• 31% full-spectrum (DSM-III-ADHD), controls 3% (hyperactive boys, ages 6 to 12 at diagnosis, ages 16 to 23 at follow-up) (Cohort study, N = 201, n = 101, approximately 9 years, E 2b.)⁶² (Cohort study, n = 100, 9 years, secondary citation from Remschmidt & Herpertz-Dahlmann 1989, E 2b.)⁶³ An additional 5% with residual form

• 27.8% with full-blown or subclinical ADHD at an average age of 18, 9 years after diagnosis between the ages of 6 and 12 (French population sample) (cohort study, N = 1,012, n = 492, 9 years, full text not available to us, E 2b.)⁶⁵

• 23% based on self-assessment using a norm-referenced cutoff score (women with childhood ADHD, mean age 20 years) (cohort study, N = 58, n = 34, E 2b.)⁵⁵

• 22.2% at an average age of 41, 33 years after diagnosis; controls: 5.1% (p < 0.001) (Men without childhood disorder) (Cohort study, N = 271, 33 years, E 2b.)⁶⁶

• 22% met all DSM-IV criteria based on self-report; controls: 2%; ϕ = 0.28 (girls, mean age 19.6 years) (Cohort study, n = 140 with ADHD, n = 88 controls, approximately 10 years, E 2b.)⁵³

• 21.1% by age 18 (cohort study, n = 2,232, age 18, E 2b.)⁶⁷

• 17.2% (based on the 1993 Pelotas Birth Cohort Study) (cohort study, n = 5,249, ages 18 to 19, E 2b)⁶⁸

• 16.7% full-spectrum (DSM-5) at an average age of 18, 9 years after diagnosis at ages 6 to 12 (French population sample) (cohort study, N = 1,012, n = 492, 9 years, full text not available to us, E 2b.)⁶⁵ An additional 11.1% had subclinical ADHD

• 12% in self-report (DSM symptom count) (women with childhood ADHD, mean age 20 years) (cohort study, N = 58, n = 34, E 2b.)⁵⁵

• 11% including clinically significant individual symptoms (full-blown disorder according to DSM-III-R, recruitment according to DSM-II, data collected solely through self-report; children with primarily aggressive behavior were not included) (Cohort study, N = 186, n = 91, 16 years, E 2b.)⁶⁹

• 10.8% consistently at all measurement time points up to a mean age of 25 years (cohort study, n = 558, 16 years, E 2b.)²

• 8% full-spectrum according to DSM-III-R (recruitment based on DSM-II, data collected solely through self-report; children with primarily aggressive behavior were not included) (Cohort study, N = 186, n = 91, 16 years, E 2b.)⁶⁹(Narrativer Review, E 4.)³⁴

• 4% full-spectrum ADHD according to DSM-III-R, no other cases with impairing individual symptoms (mean age 24.1 years; children with pronounced aggression were not included) (Cohort study, N = 158, n = 85, E 2b.)⁷⁰ (Note: significantly stricter ADHD criteria: DSM-III-R compared to DSM-II)

(E 2b): The observed persistence depended heavily on the informant. (Cohort study, n = 140 with ADHD, n = 88 controls, approximately 10 years, E 2b.)⁵³ (Cohort study, N = 58, n = 34, E 2b.)⁵⁵ Among girls with ADHD, the agreement between parental and self-reports in young adulthood (mean age 19.6 years) was low (κ = 0.22). (Cohort study, n = 140 with ADHD, n = 88 controls, approximately 10 years, E 2b.)⁵³

(E 3): In cases of childhood ADHD diagnosed retrospectively, ADHD persisted into adulthood in:

• 72.8% among adults in addiction treatment (cross-sectional study, N = 1,276, n = 290, manufacturer-funded, E 3.)⁷¹
• 70.0% (95% CI 52.8 to 82.9) in a representative population sample (ages 18 to 44) (cross-sectional study, n = 3,309, E 3.)⁷² Childhood prevalence 2.9%, adult prevalence 2.1%
• 63.0% of men meeting the full DSM-IV criteria (Australian adults, mean age approximately 32 years) (cross-sectional study, retrospective, twin registry, n = 3,795, E 3.)⁷³
• 55.3% meeting full DSM-IV criteria (Australian adults, mean age approximately 32 years) (Cross-sectional study, retrospective, twin registry, n = 3,795, E 3.)⁷³ Adult prevalence 1.1%
• 50.3% according to DSM-IV criteria without the onset criterion (Australian adults, mean age approximately 32 years) (Cross-sectional study, retrospective, twin registry, n = 3,795, E 3.)⁷³ Adult prevalence 2.3%
• 47.6% of women meeting the full DSM-IV criteria (Australian adults, mean age approximately 32 years) (cross-sectional study, retrospective, twin registry, n = 3,795, E 3.)⁷³
• 40.2% with disruptive ADHD symptoms (Australian adults, mean age approximately 32 years) (Cross-sectional study, retrospective, twin registry, n = 3,795, E 3.)⁷³ Adult prevalence 2.7%

According to an overview of the New York and Montreal cohorts, two-thirds of children who had previously been hyperactive showed no signs of a mental Disorder by their mid-20s. (Narrative review, E 4.)³⁴

At age 25, participants in the MTA study with childhood ADHD (combined type) continued to exhibit significantly more severe ADHD symptoms than a local comparison group without ADHD. ADHD scores compared to the control group (CAARS) for: (Cohort study (follow-up to an RCT), n = 515 with ADHD, n = 258 without ADHD, age 16, E 2b.)⁷⁴

• Symptom severity (average of parental and self-reports), d = 1.11
• Discrepancy between parental and self-reports (parents reported more symptoms), d = 0.60
• Severity of symptoms consistent with long-term stimulant use (constant, fluctuating, barely noticeable); no significant differences

Among adolescents with combined-type ADHD (mean age 17.4 years), current symptom severity was higher (explaining 20.9% of the variance) in: (cohort study, n = 347, 6 years, E 2b.)⁴³

• ADHD in one parent
• higher severity of ADHD symptoms at the start of the study
• higher levels of impairment reported by parents at the start of the study
• drug treatment, no effect

Compared with unaffected children, children with ADHD from a U.S. community sample had an increased risk during adolescence for: (case-control study, n = 94 with ADHD, n = 75 subclinical, n = 163 controls, 8 years, E 3.)⁵⁹

• oppositional defiant behavior, OR = 12.9 (5.6 to 30.0)
• Anxiety/depression, OR = 10.3 (2.7 to 39.3)
• significant functional impairment, OR = 3.4 (1.7 to 6.9)
• Contact with the juvenile justice system, OR = 3.1 (1.0 to 9.1)
• reduced quality of life, OR = 2.5 (1.3 to 4.7)

2.2. Partial persistence into adulthood¶

A reduction in ADHD, in the sense of a decrease in the number of symptoms, is referred to as partial remission. For most people with ADHD, however, significant limitations remain—for example, in terms of educational and career achievements—compared to people without the condition.
The percentage of people with ADHD in childhood who have partially persistent ADHD as adults has been estimated at:

• 69% full-blown or subsyndromal after 11 years (girls, aged 15 to 30 at follow-up; 42% still undergoing treatment) (Cohort study, 11 years, secondary citation from Geissler & Lesch 2011, E 2b.)⁷⁵
• 66% of those who were reported to be hyperactive as children reported at least one ADHD symptom that caused impairment during a follow-up examination in adulthood. (Narrative review, secondary citation from Weiss et al. (2002), E 4.)⁷⁶
• 60% (partial diagnosis, including full-blown disease; weighted average from follow-up studies) (literature review, n = 7, manufacturer-funded, E 4.)²⁹
• approximately 60% (complete or as a partial syndrome) (Narrative review, E 4.)⁷⁷
• 29.2% subsyndromal; an additional 10.4% functionally impaired without meeting the symptom criteria (girls, 11 years after recruitment) (Cohort study, N = 187, n = 96, 11 years, industry-funded, E 2b.)⁴⁷
• 22% subsyndromal; an additional 15% functionally impaired without meeting the symptom criteria (boys, 10 years after recruitment) (Cohort study, N = 215, n = 110, 10 years, E 2b.)⁴⁶
• 15.6% sustained partial remission through the age of 25 (cohort study, n = 558, 16 years, E 2b.)²
• 0.23% for stable partial remission (1 out of 431 children, through adolescence) (cohort study, n = 431, 8 years, E 2b.)⁵⁶

A study found that young adults whose childhood ADHD (combined subtype) had remitted below the DSM-5 symptom threshold no longer showed significantly elevated scores compared to unaffected individuals in the following areas: (Follow-up study with cross-sectional comparison, N = 115, n = 61, 12 years, E 3.)⁵²

• Inhibition and cognitive flexibility
• Working memory
• Impulsivity/emotional instability
• Self-concept issues

while

• Hyperactivity/restlessness
• Shortcomings in planning and organization

continued.

Children with ADHD who had moderate to severe persistent symptoms 15 years later (n = 22) were more likely than those with no or only mild residual symptoms (n = 39) to experience moderate to severe problems with: (Cohort study, N = 102, n = 61, 15 years, E 2b.)⁷⁸

• Alcohol consumption (15 out of 22 vs. 13 out of 39, p = 0.01)
• antisocial behavior (14 out of 22 versus 7 out of 39, p = 0.001)
• emotional problems (17 out of 22 versus 16 out of 37, p = 0.02)

There were no differences among the subgroups in terms of drug use. Hyperactive individuals without persistent symptoms did not differ from the control group in terms of antisocial behavior.

15 years after diagnosis (mean age 25 years), individuals who had been hyperactive as children were more likely than controls to: (cohort study, N = 102, n = 61, 15 years, E 2b.)⁷⁹

• at least one ADHD symptom ranging from mild to severe (restlessness, difficulty concentrating, impulsivity, explosiveness) (66% versus 7%)
• motor restlessness observed during the interview (44% versus 9.7%)
• a subjective feeling of restlessness

Despite a persistent ADHD diagnosis, boys with ADHD functioned very differently up to mid-adolescence: 20% functioned well in all three areas (academic, social, emotional), 20% functioned poorly in all three, and 60% fell somewhere in between. Maternal psychopathology, larger family size, comorbidity, and impulsivity hindered normalization. Thus, functional level and syndrome persistence are partially independent of one another. (Cohort study, N = 153, n = 85, 4 years, E 2b.)⁸⁰

 

Adults diagnosed with ADHD (combined type) in childhood could be classified into three profiles approximately 12 years later based on their self-regulation (executive functions, effortful control, emotional lability). A “normalized” profile (49.18%) differed from that of non-affected individuals only in terms of the hyperactivity index. A profile with moderate impairments accounted for 27.86%. A profile with severe generalized impairments (22.96%) had the lowest self-regulation abilities and was impaired in all functional areas (self-concept, quality of life, comorbidity). (Cross-sectional study with childhood diagnosis, N = 115, n = 61, E 3.)⁸²

2.3. Complete remission of childhood ADHD¶

(E 2b): Remission (the definition of which varies depending on the study, but usually refers to levels falling below the diagnostic threshold without partial remission) was observed

• following a diagnosis in childhood of:
  • 60% syndromic remission (disappearance of the full clinical picture) among 18- to 20-year-olds with ADHD in childhood; remission rate depends heavily on the definition (narrative review, secondary citation from Mick et al. 2004, E 4.)⁸³
  • approximately 40% (calculated: no partial diagnosis, weighted average from follow-up studies) (literature review, k = 7, manufacturer-funded, E 4.)⁴²
  • 35% at age 25 (calculated: neither complete remission nor partial remission) (meta-analysis, k = 21, E 2b.)³⁸
  • 34.4% (below the DSM-5 symptom threshold, residual deficits persisted) (follow-up study, N = 115, n = 61, 12 years, E 2b.)⁵²
  • 31.4% achieved at least temporary complete remission, of whom 59.4% experienced a subsequent relapse (cohort study, n = 558, 16 years, E 2b.)⁸⁴
  • 31% remission rate at the last two examinations (mean age at the end of the study: 16.3 years) (cohort study, n = 431, 8 years, E 2b.)⁵⁶
  • 30% symptomatic remission (fewer than half of the required symptoms) among 18- to 20-year-olds with childhood ADHD (narrative review, secondary citation from Mick et al. 2004, E 4.)⁸³
  • 22.9% complete remission (girls, 11 years after enrollment) (cohort study, N = 187, n = 96, 11 years, manufacturer-funded, Phase 2b)⁴⁷
  • 22% complete remission (boys, 10 years after enrollment) (cohort study, N = 215, n = 110, 10 years, Phase 2b)⁴⁶
  • 10% functional remission (symptomatic remission with full functional capacity) among 18- to 20-year-olds with childhood ADHD (narrative review, secondary citation from Mick et al. 2004, E 4.)⁸³
  • 9.1% sustained complete remission by the end of the study (mean age 25 years) (cohort study, n = 558, 16 years, E 2b.)⁸⁴
  • 0% (E 4)⁸⁵
• following diagnosis in adulthood in cases of:
  • 30.2% no longer met the DSM-IV criteria (7 years later), regardless of changes in comorbidity or cognitive demands and regardless of age (cohort study, n = 344, of whom n = 227 were followed up, mean age 34.1 years, 7 years, E 2b.)⁸⁶
  • 12.4% achieved complete remission (fewer than 4 symptoms, 7 years later) (cohort study, n = 344, of whom n = 227 were followed up, mean age 34.1 years, 7 years, E 2b.)⁸⁶
  • 5.7% over 13 years
    • A study examined adults with ADHD three times at intervals of 7 and 6 years, from an average age of 34 to an average age of 47. Of the 157 people with ADHD who were examined three times and whose data were evaluable, 9 (5.7%) showed a course of remission, 68.8% showed a stable course, and 25.5% showed an unstable course. Cross-sectionally, 10.2% were in remission at the third assessment. (Cohort study, N = 323, n = 157, 13 years, E 2b.)⁸⁷

(E 2b): The ADHD diagnosis was more likely to persist, while the loss of the diagnosis was less common

• for ADHD diagnosed in adulthood (persistence after 7 years) in: (cohort study, n = 344, of whom n = 227 were followed up, mean age 34.1 years, 7 years, E 2b.)⁸⁶
  • a higher number of inattention symptoms, OR = 8.05 (2.54 to 25.45)
  • social phobia, OR = 3.59 (1.12 to 11.47)
  • oppositional defiant behavior, OR = 3.12 (1.20 to 8.11)
  • a higher number of hyperactivity/impulsivity symptoms, OR = 1.18 (1.04 to 1.34)
• in children (follow-up period of 3 to 10 years) for: (Retrospective cohort study, n = 376, ages 3 to 10 years, E 2b.)⁸⁸
  • moderate versus severe baseline severity (CGI-S), loss to follow-up 47.36% versus 17.54%, p = 0.023
  • Lack of response to treatment: Loss of diagnosis in responders (CGI-I ≤ 2) versus non-responders: 42.85% versus 24.15%, p < 0.001

Of 376 children with ADHD, 117 no longer met the diagnostic criteria within 3 to 10 years. Gender, subtype, discontinuation of treatment, comorbidity, and chronic conditions were not associated with the loss of the diagnosis. (Retrospective cohort study, n = 376, 3 to 10 years, E 2b.)⁸⁸

Remission usually does not mean that the symptoms have completely disappeared, but merely that they no longer reach the level required for a diagnosis. Even after symptoms have gone into remission, educational delays and lower academic performance often persist. (Narrative review, E 4)⁸⁹

(E 2b): In cases of remitted childhood ADHD, even after the ADHD diagnosis had gone into remission, problems persisted compared to unaffected individuals or control groups in the following areas:

• executive functions
  • Women who had ADHD as girls showed improvements in their executive functions from childhood (9.5 years) through early adulthood (25.6 years) similar to those of the control group, but consistently lagged behind the control group, even when their ADHD had gone into remission by early adulthood (Cohort study, N = 228, n = 140, 16 years, E 2b.)⁹⁰
  • Male adolescents and young adults with both persistent and remitted ADHD, with stable results over 10 years and independent of the course of ADHD (cohort study, n = 217, 10 years, E 2b.)⁹¹
  • Weaker interference control (Stroop test, η² = 0.147) in both persistent and remitted ADHD, with no difference between the two groups (cross-sectional study, n = 104, E 3.)⁹² The authors view interference control as a possible core feature of ADHD across the lifespan.
• IQ, reading, and math performance were equally stable over 10 years among male adolescents and young adults with persistent and remitted ADHD (cohort study, n = 217, 10 years, E 2b.)⁹¹
• Time reproduction in people with ADHD who were in young adulthood (mean age 27 years), with shorter and more variable time reproductions at longer intervals, regardless of whether ADHD persisted or was in remission (Cohort study, N = 202, approximately 20 years, E 2b.)⁶⁰ The time errors were associated with nonverbal working memory and task fluency, not with anxiety or depression.
• Slower reaction times among individuals aged 19 to 25 who had previously been hyperactive, with or without persistent ADHD (cohort study, N = 218, E 2b.)⁹³ Attention and inhibition had improved since adolescence in all groups, but the gap compared to the control group persisted.
• Functional level 18 years after the childhood diagnosis (mean age 27.4 years): Childhood ADHD was associated with a lower level of functioning (small to large effects) across more than 60% of the assessed characteristics from all areas of life; among those who remitted, the effects were smaller than among those with persistent ADHD; ADHD persisted in 45% of participants (cohort study, n = 421, including 154 with childhood ADHD and 138 siblings without childhood ADHD; 18 years; E 2b.)⁵⁸ Siblings without childhood ADHD were largely similar to those without ADHD. The distributions of the results overlapped significantly between the groups, and many differences were small.
• Self-esteem and well-being at age 17 among individuals with a subclinical, remitting course of ADHD symptoms (14.4%) (cohort study, n = 10,262, 14 years, E 2b.)⁹⁴
• School adjustment among boys with a history of hyperactivity who no longer had a DSM-III diagnosis between the ages of 16 and 23 (52 out of 101), but not in terms of occupational adjustment, temperament, alcohol abuse, or antisocial behavior (cohort study, n = 132, approximately 9 years, E 2b.)⁹⁵ These differed barely from controls without a diagnosis (n = 80). Although they reported concentration problems (21% vs. 8%), distractibility, and hyperactivity (“constantly on the go,” 21% vs. 9%) more frequently between the ages of 16 and 18, the interviewers did not perceive this as causing any functional impairment. The authors concluded that there was a bimodal distribution of dysfunction: children who were hyperactive earlier in life either develop a diagnosable Disorder or remain largely unremarkable.

Compared with those in remission, individuals with persistent ADHD exhibited increased problems in terms of: (Cohort study, n = 421, including 154 with childhood ADHD, 138 siblings without childhood ADHD, age 18, E 2b.)⁵⁸

• Major depression (even toward people who aren’t affected)
• internalizing problems (even toward those not affected)
• perceived stress (even toward those not affected)
• Sleep quality (including compared to people who are not affected)
• Baseline attention performance: remitted patients and controls (cross-sectional study, n = 104, E 3.)⁹²
• Increased inhibition errors on the CPT and ADHD-related behavior during testing at ages 19 to 25, observed only in formerly hyperactive children with persistent ADHD compared to controls (cohort study, N = 218, E 2b.)⁹³

2.4. Fluctuating ADHD¶

(E 2b): Studies report a fluctuating course of ADHD, in which ADHD disappeared (remitted) and reappeared (recurred) multiple times during development. (Narrative review, E 4)⁸⁹ (Cohort study (follow-up to an RCT), N = 558, 16 years, E 2b)² (Cohort study, N = 323, n = 157, 13 years, E 2b)⁸⁷ (Cohort study, n = 344, 7 years, E 2b)¹⁰² (Narrative review, E 4)³(Narrativer Review, E 4.)¹⁰³

Two studies suggest that symptoms fluctuate in adulthood rather than repeatedly disappearing and reappearing:

• Among 157 adults with ADHD (mean age 34 to 47 years, 3 measurement points over 13 years), different courses were observed: (Cohort study, N = 323, n = 157, 13 years, E 2b)⁸⁷
  • 68.8% stable
  • 25.5% unstable (fluctuating)
  • 5.7% relapsing-remitting
• After 7 years, an increase in symptoms was observed with regard to: (cohort study, n = 344, 7 years, E 2b)¹⁰²
  • Hyperactivity: 25%
  • Impulsivity: 17%
  • Inattention: approx. 13%

Older review articles still assumed a chronic course and interpreted a phasic course as an indication of a mood disorder. (Narrative review, E 4.)¹⁰⁴

Symptoms and comorbidities of ADHD can also fluctuate and change over the course of a person’s life. Due to the instability of the symptom profile across the developmental course, the DSM-IV subtypes were replaced by presentation types in the DSM-5. (Narrative review, E 4.)¹⁰³

(E 2b): ADHD had a fluctuating course, with periods of remission and relapse,

• in the MTA study, for:
  • 63.8% during childhood and early adulthood (cohort study [follow-up to an RCT], N = 558, 16 years, E 2b)²
    • 31.4% achieved complete remission at least once; of these, 59.4% experienced a full or partial relapse. Only 9.1% remained in sustained remission until the end of the study (mean age 25 years), 10.8% showed stable persistence, and 15.6% remained in sustained partial remission. (Cohort study (follow-up to an RCT), N = 558, 16 years, E 2b.)²
    • A further analysis of the MTA study examined children with ADHD-C a total of 9 times, from baseline (mean age 8.46 years) through the 16-year follow-up (mean age 25.12 years). The fluctuating group (63.8%) showed an average of 3.58 fluctuations. Within each individual, there were 6 to 7 symptoms between peak and trough values. Remission phases typically first occurred during adolescence and were associated with higher environmental demands, particularly at a younger age. The severity of ADHD in this group was moderate. The group with stable, persistent ADHD (10.8%) showed an early and persistent risk for affective disorders, substance use problems during adolescence and young adulthood, lower medication adherence, and a poorer response to treatment during childhood. (Cohort study (follow-up to an RCT), N = 483, 16 years, E 2b.)¹⁰⁵
    • Protective factors included
      • in the recovery group (9.1%): very low levels of parental psychopathology
      • in the partial remission group (15.6%): higher rates of comorbid anxiety (cohort study [follow-up to an RCT], N = 483, 16 years, E 2b.)¹⁰⁵
• in a cohort of children with ADHD onset before age 12 (annual follow-up over 8 years; 86% with elevated manic symptoms) in: (cohort study, n = 431, 8 years, E 2b.)⁵⁶
  • 15% with unstable symptoms at the end of the study
  • 13% fluctuating trend
  • 53% persistent ADHD
  • 31% in remission
• in a cohort, the following were met after 10 years: (Controlled cohort study, N = 215, n = 110, 10 years, E 2b.)⁴⁶
  • 65% of boys with ADHD no longer meet all the DSM-IV criteria
  • However, 78% met at least one definition of persistence (full or subsyndromal criteria, a GAF score of 60 or lower, or ongoing ADHD treatment).
• in three cohorts of children and adolescents (at least two transitions between meeting and not meeting the ADHD criteria, proportion of people with ADHD) for: (Cohort study, 3 cohorts, n = 9,735 + 258 + 149, E 2b.)¹⁰⁶
  • the clinical NCR cohort (ages 4 to 18, 3 to 9 examinations) for:
    • 41.9% for 5 or more examinations
    • 30% among participants who remained unmedicated throughout the study
    • 26.6% in total
    • 20% for 4 exams
    • 14% for 3 examinations
    • 9.8% when using a stricter definition of remission, as in the MTA study
  • the population-based ABCD cohort (ages 9 to 13, 3 examinations) for:
    • 31.3% among participants who were not on medication at any time
    • 29.3% in total
    • 19.7% when using a stricter definition of remission, as in the MTA study
  • the NKI-Rockland cohort (ages 6 to 18, 3 examinations) for:
    • 18.1% among participants who were not on medication at any time
    • 17% in total

In the latter three cohorts of children and adolescents, the fluctuating group was never the largest subgroup. Its proportion increased as the number of assessments rose. Fluctuations were also observed among participants who remained unmedicated throughout the study. When a stricter definition of remission was applied, as in the MTA study, the proportion decreased. The authors see similarities to recurrent-remitting affective disorders or a pronounced sensitivity to developmental environmental changes. (Cohort study, 3 cohorts, n = 9,735 + 258 + 149, E 2b.)¹⁰⁶

A study found fluctuating behavioral problems (including inattention and hyperactivity) among children with ADHD in Shanghai during the COVID-19 restrictions, peaking around day 260 of isolation, followed by a decline and then a renewed increase after the lockdown. (Cohort study, n = 1,102, E 2b.)¹⁰⁷

ADHD symptoms assessed in toddlers aged 24 months showed moderate stability through preschool age (36 to 64 months), with correlations between total scores ranging from r = 0.56 to 0.60. (Cohort study, N = 256, E 2b.)¹⁰⁸

We believe that a certain proportion of fluctuating ADHD (but not all of it) is likely due to daily variations in symptom severity and the margin of diagnostic uncertainty when symptom severity is near the diagnostic threshold. In addition, stress is likely to increase the severity of ADHD symptoms. (ADxS assessment)

(E 2b): The severity of ADHD symptoms in childhood and adolescence was influenced by:

• the transition to middle school, which temporarily interrupted the developmentally expected significant decline in inattention, hyperactivity, and impulsivity among children with ADHD, with effects on: (Cohort study, N = 370, n = 258, 6 years, E 2b.)¹⁰⁹
  • Impulsivity in parental judgment, d = 0.25
  • Hyperactivity as rated by parents, d = 0.24
  • Inattention in parental judgment, d = 0.22
  • Inattention in the teacher’s assessment, d = 0.20
  • Hyperactivity as assessed by the teacher, d = 0.14
• Environmental demands, stressors, and allostatic load; comorbid symptoms; compensatory abilities; lifestyle; social support; and family chaos (assumption for adolescents) (Narrative review, E 4.)¹¹⁰ Such contextual fluctuations could cause people with ADHD to temporarily or permanently fall above or below the diagnostic threshold, even though impairing problems persist.

The proportions of 7 severity categories of inattention and hyperactivity/impulsivity among participants aged 5 to 28 were as follows: (Cohort study, n = 485 ADHD, 665 siblings, 399 controls; partially manufacturer-funded; E 2b.)¹¹¹

• consistently low, 44.3%
• difficult to lose weight through a combination of factors, 13%
• stable, mild, 12.9%
• a moderate combined decline, 12.7%
• moderate combined, increasing, 7.5%
• heavy, combined, stable, 4.8%
• severely inattentive but stable, 4.8%

Severe symptoms in childhood were usually associated with moderate to severe symptoms in adolescence and young adulthood, despite a marked decrease. (Cohort study, n = 485 with ADHD, 665 siblings, 399 controls; partially industry-funded; E 2b.)¹¹¹

Only a relatively small number of people with ADHD show a stable, remitting course of the disorder. Studies on the neurobiology and genetics of ADHD that rely on a single measurement may overlook biologically based differences among people with different life histories. (Commentary, E 4.)¹¹³

Clinically elevated teacher ratings of inattention among elementary school children remained elevated in the following year in only 33 to 46 percent of cases. Even among children with a confirmed ADHD diagnosis (MTA), less than half remained elevated, while 25 percent fell within the normal range. This decline could not be explained by hyperactivity, oppositional behavior, anxiety at baseline, or the initiation of medication. (Cohort study, k = 3 samples, N = 79, 12 to 14 months, E 2b.)¹¹⁴

(E 2b): ADHD medications were frequently discontinued during adolescence and young adulthood. (General population cohort study, n = 41,700, 4 years, E 2b.)¹¹⁵ (Registry study, n = 19,233, manufacturer-funded, author(s) employed by the manufacturer, E 2b.)¹¹⁶ The proportion of patients with ADHD receiving medication was:

• 36.1% at age 21 (80.1% at age 18) in Sweden (registry study, n = 19,233, manufacturer-funded, author(s) employed by the manufacturer, E 2b.)¹¹⁶
• 27% after just under 4 years among adolescents and young adults (ages 15 to 21) (cohort study of the general population, n = 41,700, 4 years, E 2b.)¹¹⁵

Adolescents and young adults discontinued ADHD medications more frequently than other age groups. The discontinuation rate was higher than would be expected given the persistence of ADHD. (General population cohort study, n = 41,700, 4 years, E 2b.)¹¹⁵

Of 181 patients at a forensic psychiatric outpatient clinic in Switzerland, 12 (7%) had a confirmed ADHD diagnosis from childhood or adolescence. In adulthood, the diagnosis was consistently maintained in 1 person (8%); in 4 (33%), it was initially assigned in the first evaluation and later dropped; in 4 (33%), it was dropped and later reassigned; and in 3 (25%) it was never assigned. In 50% of the cases involving a change in diagnosis, the discontinuation of the ADHD diagnosis coincided with a newly diagnosed personality disorder (or vice versa). The authors discussed fluctuating courses similar to those seen in depression, as well as “lost” diagnoses and subliminal residual symptoms. (Retrospective chart review, N = 181, n = 12, E 2b.)¹¹⁷

3. Prevalence of Childhood ADHD in Adulthood¶

Studies on predictors found that ADHD persists

• based on diagnostic criteria or symptoms in cases of:
  • 85% met full or subthreshold criteria and had impairment after 4 years (boys, ages 6 to 17) (Cohort study, n = 128, 4 years, E 2b.)¹¹⁸ 15% had remitted, with roughly half doing so by age 12 and half thereafter.
  • 65% into adolescence in Velo-cardio-facial syndrome (22q11.2 deletion) (cohort study, n = 72, 4 years, E 2b.)¹¹⁹
  • 105 of 163 adolescents (ages 16 to 18) with childhood ADHD as determined by retrospective interviews (cohort study, n = 457, E 2b.)¹²⁰ went into remission.
  • 50% on average (ranging from 32.8% to 84.1% depending on the country) according to DSM-IV among adults with retrospectively reported childhood ADHD in 10 countries (cross-sectional study, retrospective, n = 629 with childhood ADHD, manufacturer-funded, E 3.)¹²¹
  • 50% (10 out of 20) according to DSM-5 among children diagnosed and treated with medication between 2012 and 2013; follow-up between 2020 and 2022 at ages 13 to 25 (Cohort study, N = 121, n = 20, E 2b.)¹²²
  • 49% (64 out of 130) with full-blown disease during adolescence (mean age 14.0 years, onset at 8.9 years) (cohort study, n = 130, 5 years, E 2b.)¹²³ 18% (24) had achieved complete remission.
  • 36.3% according to DSM-IV among 18- to 44-year-olds with retrospectively diagnosed childhood ADHD, blinded clinical interviews (cross-sectional study, retrospective, n = 3,197, manufacturer-funded, author employed by manufacturer, E 3.)¹²⁴
• Among Chinese children in early adulthood (mean age 18.8 years), depending on the definition of persistence: (cohort study, n = 399, median 6.58 years, E 2b.)¹²⁵ 27.32% were in complete remission.
  • 46.37% according to DSM-IV
  • 19.55% with persistent symptoms
  • 6.77% with functional persistence
• as a persistently severe course of symptoms in:
  • 14.3% remained consistently high in measurements taken at ages 8, 14, and 23 (decreasing from 24.7%, remaining consistently low at 61.0%) (Cohort study, n = 306, 24 years; full text not available to us, E 2b.)¹²⁶
  • 13% remained high from 3rd through 12th grade in a high-risk sample (children with increased behavioral problems in kindergarten), Parental assessment (persistently low at 71%; initially clinically significant, then decreasing below the clinical threshold to 16%) (cohort study, n = 413, E 2b.)¹²⁷
  • 10.6% remained high through age 16 (cohort study, n = 1,730, 5 years, E 2b.)¹²⁸

(E 2b): Groups with a persistently high course and those with a declining course had similar symptom severity in (early) childhood. (Cohort study, n = 413, E 2b.)¹²⁷ (Cohort study, n = 485 with ADHD, 665 siblings, 399 controls, partially funded by manufacturers, E 2b.)¹¹¹ A persistently high course was associated with more adverse consequences. At age 32, the group with a consistently high level of symptoms had more psychosocial impairments and a lower socioeconomic status. (Cohort study, n = 306, 24 years, full text not available to us, E 2b.)¹²⁶ In late adolescence, the group with persistently high levels was more antisocial than the group with low levels (more arrests, school dropouts, unemployment); in contrast, the group with decreasing levels did not differ from the group with low levels. (Cohort study, n = 413, E 2b.)¹²⁷ During adolescence, the course of ADHD is thought to diverge. A subgroup achieves a largely favorable outcome, while a larger group remains on a problematic trajectory. (Narrative review, E 4.)¹¹⁰

Predictive models based on childhood characteristics achieved a predictive accuracy (AUC) of: (Predictive model from cohort studies, n = 5,113 (development), validation on n = 2,040, n = 3,911, and n = 717, E 2b.)¹²⁹

• 0.82 (95% CI 0.79 to 0.83) for ADHD in young adulthood (without the onset criterion), internal validation
• 0.76 (0.73 to 0.80) for ADHD in young adults; external validation in the MTA
• 0.76 for a multivariate risk profile of persistence (cross-sectional study, retrospective, n = 629 with childhood ADHD, manufacturer-funded, E 3.)¹²¹
• 0.75 (0.71 to 0.78) for ADHD in young adults; external validation in E-Risk
• 0.57 (0.54 to 0.60) for ADHD in young adults in the Brazilian Pelotas cohort

The risk calculator accurately predicted ADHD in European and North American samples, but barely in the Pelotas cohort. The model did not predict anxiety disorders or depression in adulthood. (Predictive model based on cohort studies, n = 5,113 (development), validated on n = 2,040, n = 3,911, and n = 717, E 2b.)¹²⁹

Persistent ADHD was associated with an increased risk of: (Meta-analysis, k = 36, E 2b.)¹³⁰

• Major depression, OR 3.19 (95% CI 1.71 to 5.95)
• Substance use disorders, OR 2.12 (95% CI 1.53 to 3.17)
• Social behavior disorders and substance use disorders among hyperactive boys aged 16 to 23 who continued to have ADHD (cohort study, N = 201, n = 101, approximately 9 years, E 2b.)⁶² Excluding cases with persistent ADHD, former patients and controls no longer differed significantly in terms of antisocial disorders (13% vs. 8%) and substance use disorders (10% vs. 7%). Substance use disorders almost always followed a conduct disorder.

The differences in the presentation of symptom severity during childhood were no longer significant during adolescence (ages 16 to 18). Children with the combined type most often met only the criteria for the inattentive type during adolescence. (Cohort study, n = 457, E 2b.)¹²⁰
Polygenic ADHD risk and maltreatment had independent effects; no gene-environment interaction was found. The ADHD PGS was associated with maltreatment (β = 0.09, 0.01 to 0.17). Approximately 47% of this effect was mediated through ADHD symptoms at age 6, suggesting an evocative gene-environment correlation (children with ADHD symptoms are more likely to experience maltreatment). (Birth cohort, n = 4,231, age 18, E 2b.)¹³¹

Current ADHD symptoms (ASRS) were directly and, additionally, indirectly associated with childhood trauma (CTQ) and dissociative experiences (DES) and with retrospectively reported childhood ADHD symptoms (WURS), independent of demographic and clinical confounders. This was found among both college students (n = 1,148) and adults with ADHD (n = 202). (2 cross-sectional studies, N = 1,350; full text not available to us; E 3.)¹³²
Among women (ages 35 to 88), subclinical (n = 82) and full-blown ADHD (n = 43) were associated with more distressing childhood experiences and weaker parental attachment. (Cross-sectional study, population-based, N = 2,741, E 3.)¹³³
Some adults with ADHD do not exhibit significant impairments in daily life, relationships, or work and education, which points to prognostic heterogeneity. The authors hypothesize that preserved processing of delayed rewards (as described by the dual-pathway model) is the neurobiological correlate of perseverance, which is associated with greater academic success. (Narrative review, E 4.)¹³⁴

(E 4): The long-term outcome of ADHD is thought to depend on characteristics of the child and the family (Narrative Review, E 4.)¹³⁵ (Narrative review, E 4.)³² —specifically the child’s health, temperament, and IQ; socioeconomic status; emotional climate; parents’ mental health; family structure; and the broader social environment (Narrative Review, E 4.)¹³⁵ or on the severity of symptoms, comorbidity, and the child’s IQ, parental psychopathology, socioeconomic status, family stress, and treatment, all of which are thought to have cumulative and interactive effects. (Narrative review, E 4.)³² In particular, a comorbid social behavior disorder, a low IQ, and parental psychopathology were unfavorable factors. The long-term effects of treatment remained unclear. (Narrative review, E 4.)³²

The functional level of individuals who were hyperactive as children in young adulthood (ages 17 to 24) did not depend on individual childhood characteristics (ages 6 to 12), but rather on their cumulative interaction. The following were particularly common predictors: (Cohort study, n = 64, ages 10 to 12, E 2b.)¹³⁶

• socioeconomic status (education, occupation, contact with law enforcement)
• mental health of family members (emotional adjustment, contact with law enforcement, drug use)
• IQ (educational attainment, drug use)
• emotional instability, low frustration tolerance, and aggressiveness (encounters with the police, emotional adjustment, accident-related damage)

Attempts to predict, based on childhood characteristics, which hyperactive children would have an unfavorable outcome were largely unsuccessful. In one study, only IQ (negatively) and number of siblings (positively) were associated with antisocial personality disorder, and only IQ was associated with alcoholism; all predictors combined explained less than 10% of the variance. In another study, out of approximately 300 correlations between childhood characteristics and outcome, 29 were significant, none of them strong. (Narrative review, E 4.)¹³⁷

The CBCL scale for attention problems at age 11 was multidimensional (SCT items and ADHD items). A bifactor model predicted psychopathology, educational attainment, financial situation, and relationship skills in young adulthood somewhat better than simple sum scores. (Cohort study, N = 1,352, E 2b.)¹³⁸

Self-reported SCT during childhood and adolescence predicted symptoms of depression (β = 0.26) and inattention (β = 0.12) in adulthood 12 years later. However, SCT and depression were strongly correlated during adolescence (r = 0.84 in self-reports, r = 0.78 in parental reports). When considered together, neither predictor predicted adult inattention or depression. (Cohort study, twins, 449 twin pairs, 12 years, E 2b.)¹³⁹

There are various theories as to why ADHD that begins in childhood disappears in some adults but persists in others.

• Among children with ADHD from low socioeconomic backgrounds, the likelihood of remission (as reported by parents) was higher in U.S. states with more inclusive special education. For children from higher socioeconomic backgrounds, it did not depend on this factor. (Cross-sectional study, E 3.)¹⁴⁰ In our view, this suggests that families with low socioeconomic status are less able to compensate for their children’s deficits. This is likely to apply to all mental health issues. (ADxS assessment)
• A study of stress levels in children with ADHD found that very high stress levels, peaking in mid-adolescence, were associated with two persistent courses of severe combined ADHD (characterized by severe emotional dysregulation or by increasing internalizing problems). Low and decreasing stress levels were associated with partially remitting courses (predominantly inattentive type and moderate combined type). (Cohort study, N = 1,587, n = 609, 8 years, E 2b.)¹⁴¹ A large Swedish registry cohort found a dose-response relationship between cumulative stressors in childhood and treated ADHD in late adolescence and young adulthood (4 or more stressors: OR 5.5). (Cohort study, n = 543,650, E 2b.)¹⁴²
• Another study found that adolescents with late-onset ADHD were 11.8 times more likely to have been exposed to multiple traumas than people with ADHD (OR 11.82). Clinically and neurocognitively, late-onset ADHD did not differ from ADHD with childhood onset. However, people with late-onset ADHD experienced higher parental demands (d = 1.09) and were more likely to transfer to alternative schools (26.7% versus 0%). Compared to non-affected individuals, they exhibited deficits in metacognition (d = 1.25) and academic motivation (d = 0.80). In terms of exposure to multiple traumas, they did not differ significantly from people with childhood ADHD. After one year, 50% of people with late-onset ADHD met the DSM-5 symptom criteria, and 67.7% exhibited increased symptoms with clinically significant impairment. (Case-control study with 1-year follow-up, N = 50, n = 15, E 3.)¹⁴³ The lack of persistence in some cases could also indicate a high rate of misdiagnosis. (ADxS assessment)
  • Puberty is accompanied by significant changes in dopaminergic innervation in the PFC. During this period, the PFC is particularly vulnerable to dopaminergic stimulation. This explains the behavior of adolescents, which oscillates between novelty seeking and strong withdrawal. The high dopaminergic vulnerability during this developmental phase further explains the risks associated with alcohol and drug abuse or excessive media consumption. These can cause lasting disorders of the dopamine system, which increases the risk of impulsivity, addiction, or psychosis. (Full text not available to us, E 4)¹⁴⁴
    Delayed cortical maturation in the frontoparietal network has a beneficial effect on the brain’s ultimate performance, whereas acceleration—e.g., due to excessive dopaminergic stimulation—has an adverse effect. (Full text not available to us, E 4)¹⁴⁵ In ADHD, as in cases of high intelligence, cortical thickness reached its maximum later. (Longitudinal study, N = 446, E 2b.)¹⁴⁶ (Longitudinal study, n = 307, E 2b.)¹⁴⁷
    See ⇒ Giftedness and ADHD In the section ⇒ Differential Diagnosis of ADHD in the chapter ⇒ Diagnosis
• Anxiety symptoms at age 15 predicted mental disorders in late adolescence and early adulthood (adjusted for ADHD, ASD, and developmental coordination disorder) (Cohort study of twins, n = 14,106 + 9,211, E 2b.):¹⁴⁸
  • Anxiety disorder: 4.9 times
  • Depression: 4.8 times
  • any psychiatric diagnosis: 3.4 times
    This could be further evidence that middle adolescence, along with early childhood, represents a second particularly vulnerable developmental window. (ADxS assessment)
• One theory posits that brain functions continue to mature (convergence toward typical brain structure and function). (Narrative review, E 4.)¹⁴⁹ However, not every delay in brain maturation is necessarily a sign of ADHD. In ADHD, the cortex reached its maximum thickness at a median age of 10.5 years instead of 7.5 years. (Longitudinal study, N = 446, E 2b.)¹⁴⁶ Children with very high intelligence (IQ 121 to 149) also exhibited an initially accelerated and prolonged phase of cortical thickening (peaking in the superior frontal gyrus at around 11 years instead of 7 to 8 years), followed by an equally pronounced thinning beginning in early adolescence. (Longitudinal study, n = 307, E 2b.)¹⁴⁷
  ⇒ ADHD and Giftedness.

• Genetic factors associated with relative immaturity in childhood accounted for 10% to 14% of the variance in ADHD symptoms up to the age of 16 to 17, and only 4% by the age of 19 to 20. (Twin cohort study, n = 1,302 twin pairs, approximately 11 years, E 2b.)¹⁵⁰ This is consistent with the theory of post-maturation, but explains only a small portion of the symptom variance.
• Girls with childhood ADHD showed greater improvement from childhood (ages 6 to 12) through young adulthood (ages 17 to 24) than controls in sustained attention and global executive functions, and similar improvement in working memory and response inhibition. Those who showed the greatest improvement in global executive functions also exhibited the greatest reduction in inattention and hyperactivity/impulsivity. An improvement in response inhibition was associated with a reduction in hyperactivity/impulsivity. Changes in working memory were not associated with changes in symptoms, which the authors interpret as a possible indication of a core deficit. (Cohort study, n = 228, approximately 10 years, E 2b.)¹⁵¹
• Improvements in executive functions—particularly the development of working memory—are thought to be associated with a reduction in symptoms. (Narrative review, E 4.)¹¹⁰
• Another theory suggests that certain regions of the brain take on compensatory functions, thereby correcting the ADHD deficits in children. (Narrative review, E 4)¹⁴⁹
• Another model posits that certain ADHD deficits in children persist throughout life. (Narrative review, E 4)¹⁴⁹
• A deficit in response inhibition (stop-signal task) persisted in children with ADHD into adolescence, regardless of whether the ADHD persisted, partially remitted, or completely remitted. Working memory and reaction variability no longer differed significantly from controls in adolescence. Cognitive performance improved with age, regardless of the course of symptoms, and did not differ between persistent, partially remitted, and remitted ADHD, either in childhood or adolescence. (Cohort study, n = 130 (64 persistent, 42 partially remitted, 24 remitted), controls n = 40 and 28, E 2b.)¹⁵²
• Neither automatically controlled (lower-level) nor consciously controlled (higher-level) neurocognitive functions distinguished persistent ADHD from remitted ADHD. Both the persistent and remitted groups performed worse than those without ADHD, with the remitted group showing a smaller gap. Neurocognitive functions measured in childhood predicted ADHD several years later, regardless of the type of function. This does not support the maturation model proposed by Halperin and Schulz (2006), according to which consciously controlled functions continue to mature during remission. (Systematic review, k = 18, E 2b.)¹⁵³
• (E 3): A study of adults whose inattention persisted found an imbalance between connections within the brain’s default-mode network, on the one hand, and those between the default-mode network and the regions that support attention and cognitive control, on the other. In adults whose inattention had remitted, these differences were not observed compared to people with ADHD who had never been affected. (Cross-sectional study in a longitudinal cohort, n = 205, E 3.)¹⁵⁴ (Consensus, E 4.)¹⁵⁵
• In a resting-state fMRI study of adults with childhood ADHD, positive connectivity between the posterior cingulate and the medial prefrontal cortex (default mode network) was reduced only in those with persistent ADHD; those in remission did not differ from controls. In contrast, a reduced anticorrelation between the medial and dorsolateral prefrontal cortex was found in both ADHD groups, regardless of current status. (Narrative review, secondary citation via Uchida et al. 2015 on Mattfeld et al. 2014, E 4.)¹⁵⁶
• A comparison of adolescents with partial remission and those without remission revealed significantly lower activation of the vlPFC in the partial remission group. This improvement in vlPFC efficiency correlated with performance on a Go/No-Go task and fell between ADHD diagnoses and normal controls. (Cross-sectional study using fMRI, N = 15, E 3)¹⁵⁷
• A genetic analysis study identified four genome-wide significant loci for ADHD in childhood and one for ADHD diagnosed only in adulthood. Persistent ADHD showed higher polygenic risk scores for ADHD (ADHD-PRS) than the other two groups. ADHD in childhood showed greater genetic overlap with hyperactivity and autism, as well as the highest burden of rare protein-truncating variants in evolutionarily conserved genes. In contrast, late-onset ADHD showed greater genetic overlap with depression and no increased burden of rare protein-truncating variants. (GWAS, n = 23,312 cases, 38,303 controls; rare variant analysis: 7,650 cases, 8,649 controls; E 3.)³⁷

Children with ADHD (combined type) and their unaffected siblings caught up with unaffected peers on some neurocognitive measures between a mean age of 11.4 and 17.3 years, specifically in: (Cohort study, n = 838, including 339 people with ADHD and 271 unaffected siblings, E 2b.)¹⁵⁸

• People with ADHD affected 4 out of 9 measurements (44.4%)
• unaffected siblings: 5 out of 9 measurements (55.6%)

Verbal working memory and reaction time variability remained impaired. The change in neurocognitive performance was not associated with subsequent ADHD symptom severity or overall functioning (K-GAS). (Cohort study, n = 838, including 339 people with ADHD and 271 unaffected siblings, E 2b.)¹⁵⁸

 

4. Possible Reasons for ADHD Persisting into Adulthood¶

(E 4): The evidence regarding predictors of persistence versus remission is weak and largely inconsistent. (Narrative review, E 4.)¹⁰³ Known predictors do not adequately explain the course of the disease. (Narrative review, E 4.)¹¹⁰
The persistence of ADHD is not merely a continuation of early high symptom severity, but rather a product of genetic predisposition, socioeconomic context, and their interaction. These influences differed significantly between inattention and hyperactivity/impulsivity, which should be considered separately when examining the risk of persistence. (Editorial, E 4.)¹⁵⁹
The presentation and predictors of ADHD change over the course of development. In preschool-aged children, symptom severity, cognitive functions, and family factors predict the course of the disorder during school-age years. These factors remain predictive of long-term outcomes in school-aged children, with comorbidity emerging as an additional predictor. (Narrative review, industry-funded, E 4.)¹⁶⁰

4.1. Factors Influencing the Persistence of ADHD into Adulthood¶

 

The previous absolute statement—“It cannot be predicted in which people with ADHD will see their symptoms disappear in adulthood”—has been qualified. It was based on a single cohort reported by Barkley et al. (2008). A meta-analysis (16 studies, including 6 population-based retrospective studies and 10 clinical follow-up studies) identified severity, treatment, comorbid conduct disorder, and major depression in childhood as predictors of persistence.¹⁶¹
A stable course in middle adulthood was more common among individuals with more severe symptoms, those of female gender, those with comorbidities, and those with a genetic predisposition to depression.⁸⁷However, the specific predictors remain unclear on an individual basis.¹⁶²

The following factors influencing the persistence of ADHD into adulthood were examined:

4.1.1. ADHD Severity and Number of Symptoms in Childhood¶

• OR 8.05 for more symptoms of inattention in adults; persistence of the diagnosis after 7 years (cohort study, n = 227, 7 years, E 2b.)⁸⁶
• OR 2.33 for ADHD severity (meta-analysis, k = 16, E 2b.)¹⁶¹
• OR 2.0 for symptom severity (cross-sectional study, retrospective, n = 629 with childhood ADHD, industry-funded, E 3.)¹²¹
• OR 1.89 for initial ADHD severity and persistence of symptoms until an average age of 25 (MTA cohort) (cohort study, n = 453, approximately 17 years, E 2b.)¹⁶³
• OR 1.62 for inattention symptoms (lower among those who fully remitted as children than among those with fully persistent symptoms), adolescence (cohort study, n = 130, 5 years, E 2b.)¹²³
• OR 1.18 for more symptoms of hyperactivity/impulsivity in adults; persistence of the diagnosis after 7 years (cohort study, n = 227, 7 years, E 2b.)⁸⁶
• OR 1.15 (1.06 to 1.25) per hyperactivity/impulsivity symptom, adjusted, early adulthood (cohort study, n = 399, median 6.58 years, E 2b.)¹²⁵
• R² = 20.9% for ADHD severity 6 years later, in combination with ADHD in one parent and greater impairment at the start of the study (cohort study, n = 347, 6 years, E 2b.)⁴³
• 18% of adolescents had symptom scores below the threshold for all 9 inattention symptoms in childhood; 58% had 5 or fewer symptoms (cohort study, n = 130, 5 years, E 2b.)¹²³
• Attention problems reported in the parent questionnaire (CBCL) during adolescence, persistence of the diagnosis 25 years later (cohort study, n = 19 ADHD, 26 controls, 25 years; full text not available to us, E 2b.)⁵⁴ADHD symptoms reported by parents (not teachers) as predictors of ADHD symptoms and impairment during adolescence and young adulthood (cohort study, n = 116, 6.6 years, E 2b.)⁴⁵
• certain symptoms of inattention (following instructions, organizing tasks, avoiding tasks, forgetfulness, losing things), persistence from ages 16 to 18 (cohort study, n = 457, E 2b.)¹²⁰
• Number of ADHD symptoms (modeled nonlinearly) for ADHD in young adulthood (prediction model based on cohort studies, n = 5,113 (development), validation on n = 2,040, n = 3,911, and n = 717, E 2b.)¹²⁹
• Increased hyperactivity in Velo-cardio-facial syndrome (cohort study, n = 72, 4 years, E 2b.)¹¹⁹
• Severity of symptoms in childhood (narrative review, E 4.)¹⁶⁴
• Severity of symptoms starting in preschool age (narrative review, industry-funded, E 4.)¹⁶⁰
• Severity of childhood ADHD as the sole predictor aside from treatment during childhood; no modifiable risk factors (cross-sectional study, retrospective, n = 3,197, industry-funded, author employed by the manufacturer, E 3.)¹²⁴
• No correlation between the severity of the Disorder and full recovery (symptoms and impairment below the 84th percentile) by age 27 (E 4.)¹⁶⁵
• No increased persistence of ADHD symptoms or impairment in young adulthood despite initially more severe symptoms in children with mixed-type ADHD who also met the stricter ICD-10 criteria for a hyperkinetic disorder (HKS) (proportion with ICD-10 HKS in young adulthood: 5.5% (6 out of 109) compared to 6.8% (25 out of 367) without childhood HKS) (Cohort study (follow-up to an RCT), n = 476 (109 with, 367 without HKS in childhood), 16 years, E 2b.)¹⁶⁶
• No significant association with the number of inattention symptoms (cohort study, n = 399, median 6.58 years, E 2b.)¹²⁵
• No predictive value based on the number of symptoms in girls (cohort study, N = 187, n = 96, age 11, manufacturer-funded, E 2b.)⁴⁷

4.1.2. Comorbid social behavior disorder, oppositional defiant disorder, and behavioral problems¶

• OR 3.12 for comorbid oppositional defiant disorder in adults; persistence of the diagnosis after 7 years (cohort study, n = 227, 7 years, E 2b.)⁸⁶
• OR 1.85 for a comorbid social behavior disorder (meta-analysis, k = 16, E 2b.)¹⁶¹
• OR 1.81 (1.21 to 2.71) for oppositional defiant disorder or disorder of conduct, ADHD in young adulthood (prognostic model based on cohort studies, n = 5,113 (development), validation on n = 2,040, n = 3,911, and n = 717, E 2b.)¹²⁹
• OR 0.23 for comorbid oppositional defiant behavior (less common in children with complete remission than in those with complete persistence), adolescence (cohort study, n = 130, 5 years, E 2b.)¹²³
• more severe symptoms, greater functional impairment, and a higher likelihood of ADHD persisting into adulthood in cases of comorbid oppositional defiant disorder and social behavior disorders (narrative review, E 4.)¹⁰³
• Comorbidity with conduct disorder, boys (cohort study, n = 128, 4 years, E 2b.)¹¹⁸
• Disorder of behavior and bipolar disorder (increased only when persistent compared with controls), girls (cohort study, N = 187, n = 96, age 11, manufacturer-funded, E 2b.)⁴⁷
• Comorbid major depression and/or oppositional disorder, persisting from ages 16 to 18 (cohort study, n = 457, E 2b.)¹²⁰
• Social Behavior Disorders as Part of Multiple Psychiatric Comorbidities (Narrative Review, E 4.)¹⁶⁴
• comorbid symptoms, particularly oppositional behavior, in follow-up cohorts with similar severity of childhood symptoms (cohort study, n = 485 with ADHD, 665 siblings, 399 controls; partially funded by the manufacturer; E 2b.)¹¹¹
• Increased aggression and hyperactivity at school when levels remain high versus when they decline (cohort study, n = 413, E 2b.)¹²⁷
• Oppositional behavior in childhood, associated with a high and decreasing symptom score (cohort study, n = 306, 24 years; full text not available to us, E 2b.)¹²⁶
• Low aggression with a more favorable course (Narrative Review, E 4.)¹¹⁰
• OR 1.03 (0.87 to 1.24) for ODD/disorder of behavior, not significant (meta-analysis, k = 36, E 2b.)¹³⁰
• No significant association with oppositional defiant behavior (cohort study, n = 399, median age 6.58 years, E 2b.)¹²⁵
• No association with oppositional behavior after controlling for ADHD symptoms (cohort study, n = 116, 6.6 years, E 2b.)⁴⁵
• lower rates of persistence for conduct disorder in childhood (58.2% versus 79.6%), OR 0.2 (0.1 to 0.8), based on oppositional defiant disorder (OR 0.05, 0.01 to 0.3), but not on conduct disorder (OR 0.6, 0.1 to 3.1) (cross-sectional study, n = 3,309, E 3.)⁷²
• No association between behavioral problems in childhood (at the start of the study) and full recovery by age 27 (E 4.)¹⁶⁵
• No association between the severity of ADHD, ODD, or CD in adolescence and full recovery by age 27 (E 4.)¹⁶⁵

4.1.3. Traumatic childhood experiences, abuse, and an unfavorable family environment¶

• Early-life adversity (such as physical abuse or neglect) is thought, based on human and animal studies, to contribute significantly to the persistence of ADHD into adulthood and to vulnerability to adult ADHD. Proposed mechanisms include dysregulation of the HPA axis, epigenetic changes, sleep disturbances, oxidative stress, neuroinflammation, and impaired myelination, all of which may reinforce one another. (Narrative review, E 4.)¹⁶⁷
• OR 3.74 (1.89 to 7.42) for perceived parental rejection at age 11, with a persistently high symptom trajectory through age 16 compared to a low trajectory (cohort study, n = 1,730, 5 years, E 2b.)¹²⁸
• OR 2.38 (1.10 to 5.14) for perceived parental rejection, with a consistently high versus high, decreasing trend (cohort study, n = 1,730, 5 years, E 2b.)¹²⁸
• OR 2.14 (1.05 to 4.38) for perceived parental rejection, with a persistently high course compared to a moderately stable course (cohort study, n = 1,730, 5 years, E 2b.)¹²⁸ No interactions with DRD4, MAOA, and 5-HTTLPR were found
• OR 2.07 (1.26 to 3.38) for a high score on the Adverse Childhood Experiences (ACE) scale, with a partially remitting course between the ages of 7 and 16 in children not receiving medication, adjusted for gender and psychiatric comorbidity. The mean Adverse Childhood Experiences (ACE) scores among children not receiving medication with a history of ADHD between the ages of 7 and 16 were as follows: (cohort study, n = 457, 9 years, E 2b.)¹⁶⁸
  • 1.93 for partially remitting ADHD (n = 18)
  • 1.29 for persistent ADHD (n = 90)
  • 1.19 for remitting ADHD (n = 40)
  • 1.12 for subclinical ADHD (n = 25)
  • 1.03 in the control group (n = 269)
  • 1.0 for late-onset ADHD (n = 15)
• OR 1.28 (1.01 to 1.64) for probable childhood maltreatment and ADHD in young adulthood (prognostic model based on cohort studies, n = 5,113 (development), validation on n = 2,040, n = 3,911, and n = 717, E 2b.)¹²⁹
• β = 0.51 (0.26 to 0.77) for maltreatment at ages 6 and 11, and self-reported ADHD symptoms at age 18 (ASRS), independent of polygenic risk (birth cohort, n = 4,231, 18 years old, E 2b.)¹³¹
• β = 0.24 (0.18 to 0.29) for maltreatment at ages 6 and 11, and ADHD symptoms at age 18 as assessed by mothers (SDQ) (birth cohort, n = 4,231, age 18, E 2b.)¹³¹
• psychosocial stress (Rutter Index), boys (cohort study, n = 128, age 4, E 2b.)¹¹⁸
• dysfunctional family, more family conflicts, less cohesion, girls (cohort study, N = 187, n = 96, age 11, manufacturer-funded, E 2b.)⁴⁷
• Retrospectively reported child maltreatment was more strongly associated with a consistently high symptom class than with a low one, while sexual abuse was more strongly associated with a consistently high symptom class than with a decreasing one (cohort study, n = 306, 24 years; full text not available to us, E 2b.)¹²⁶

• Higher persistence rates for both subclinical and full-blown ADHD in a subgroup characterized by adverse childhood experiences compared to an “idiopathic” subgroup; women (cross-sectional study, population-based, N = 2,741, E 3.)¹³³
• Social Burden (Narrative Review, E 4.)¹⁶⁴
• Indirect association between current and retrospectively reported childhood ADHD symptoms via childhood trauma and dissociative experiences (2 cross-sectional studies, N = 1,350; full text not available to us, E 3.)¹³²
• No predictive value of childhood stress (cohort study, n = 130, 5 years, E 2b.)¹²³
• No predictive value for childhood stress and trauma after adjusting for severity (cross-sectional study, retrospective, n = 3,197, industry-funded, author employed by the manufacturer, E 3.)¹²⁴
• No significant association with overprotection and parental warmth (cohort study, n = 1,730, 5 years, E 2b.)¹²⁸
• No association with the parent-child relationship, MTA cohort (cohort study, n = 453, approximately 17 years, E 2b.)¹⁶³

4.1.4. IQ in Childhood¶

• OR 0.45 (0.24 to 0.84) for an above-average IQ (115 or higher, compared with an IQ below 90), adjusted (cohort study, n = 399, median 6.58 years, E 2b.)¹²⁵
• OR 0.89 (0.85 to 0.95) per 10 IQ points, ADHD in young adulthood (prognostic model based on cohort studies, n = 5,113 (development), validation on n = 2,040, n = 3,911, and n = 717, E 2b.)¹²⁹
• OR 0.99 (0.98 to 1.00) for a higher IQ (protective effect); no longer significant after excluding studies of low quality (meta-analysis, k = 36, E 2b.)¹³⁰
• Lower IQ (full-scale and performance IQ) associated with ADHD symptoms and impairment during adolescence and early adulthood (cohort study, n = 116, 6.6 years, E 2b.)⁴⁵
• Above-average intelligence with a more favorable course (Narrative Review, E 4.)¹¹⁰
• A high IQ facilitated better coping strategies (Narrative Review, E 4.)¹⁶⁴
• IQ in follow-up cohorts with similar severity of symptoms in children (cohort study, n = 485 with ADHD, 665 siblings, 399 controls; partially funded by the manufacturer; E 2b.)¹¹¹
• No association between childhood IQ and full recovery by age 27 (E 4.)¹⁶⁵
• No association with childhood IQ, MTA cohort (cohort study, n = 453, approximately 17 years, E 2b.)¹⁶³
• No predictive value for IQ (cohort study, n = 130, 5 years, E 2b.)¹²³
• No protective effect of a high IQ (120 or higher); the diagnosis persisted for 4.5 years in 78% of gifted children and adolescents with ADHD, compared to 82.4% of those with an IQ below 120 (no significant difference) (cohort study, N = 117, n = 38, 4.5 years, E 2b.)¹⁶⁹ All people with ADHD who no longer met the criteria exhibited subthreshold symptoms
• Intelligence does not predict the course of the disease (cohort study, n = 226, 6 years, E 2b.)¹⁷⁰

4.1.5. Mental health disorders in parents¶

• OR 2.4 for anxiety or mood disorders in the father (not the mother) (cross-sectional study, retrospective, n = 629 with childhood ADHD, manufacturer-funded, E 3.)¹²¹
• OR 2.2 for antisocial personality disorder in a parent (cross-sectional study, retrospective, n = 629 with childhood ADHD, manufacturer-funded, E 3.)¹²¹
• OR 1.41 (1.13 to 1.75) for maternal depression and ADHD in young adulthood (prognostic model based on cohort studies, n = 5,113 (development), validation on n = 2,040, n = 3,911, and n = 717, E 2b.)¹²⁹
• OR 1.35 for parental mental health problems, MTA study (over 450 participants, mean age 24.7 years): even after controlling for the severity of the child’s symptoms, a 28% increased risk of persistence (commentary, secondary citation via Agnew-Blais 2017 on Roy et al. 2016, E 4.)¹⁷¹
• OR 1.30 for parental mental health problems, MTA cohort (cohort study, n = 453, approximately 17 years, E 2b.)¹⁶³
• Familial multiple anxiety disorders (61% compared to 36% among those who discontinued treatment), girls (cohort study, N = 187, n = 96, age 11, manufacturer-funded, E 2b.)⁴⁷
• Exposure to maternal mental disorders, boys (cohort study, n = 110 with ADHD, n = 105 controls, age 11, industry-funded, E 2b.)¹⁷²
• Negative maternal mood during childhood, associated with a high and decreasing symptom score (cohort study, n = 306, 24 years; full text not available to us, E 2b.)¹²⁶
• Mental disorders in parents (narrative review, E 4.)¹⁶⁴
• Family factors starting in preschool age (narrative review, industry-funded, E 4.)¹⁶⁰

4.1.6. ADHD Treatment in Childhood¶

• OR 4.5 (1.7 to 11.8) for treatment during childhood, adjusted for severity (cross-sectional study, retrospective, n = 3,197, manufacturer-funded, author employed by the manufacturer, E 3.)¹²⁴
• OR 2.09 for ADHD treatment (meta-analysis, k = 16, E 2b.)¹⁶¹
• OR 2.18 (1.41 to 3.36) with drug therapy (cohort study, n = 399, median 6.58 years, E 2b.)¹²⁵
• OR 1.88 (1.28 to 2.75) for a history of stimulant treatment (meta-analysis, k = 36, E 2b.)¹³⁰
• ADHD diagnosis and ADHD medication in cohorts with similar severity of childhood symptoms (cohort study, n = 485 ADHD cases, 665 siblings, 399 controls; partially funded by the manufacturer; E 2b.)¹¹¹
• No association between the duration of stimulant treatment through adolescence and full recovery by age 27 (E 4.)¹⁶⁵
• No effect of drug treatment on symptom severity or overall functional status after 6 years (cohort study, n = 347, 6 years, E 2b.)⁴³
• No prediction based on treatment, girls (cohort study, N = 187, n = 96, 11 years old, manufacturer-funded, E 2b.)⁴⁷

4.1.7. Number of comorbid disorders and multimorbidity¶

• OR 1.7 for 3 or more comorbid disorders in childhood/adolescence (cross-sectional study, retrospective, n = 629 with childhood ADHD, industry-funded, E 3.)¹²¹
• OR 1.19 for the number of comorbid disorders, MTA cohort (cohort study, n = 453, approximately 17 years, E 2b.)¹⁶³
• OR 1.15 for childhood comorbidity, MTA study (commentary, secondary citation via Agnew-Blais 2017 on Roy et al. 2016, E 4.)¹⁷¹
• 42.5% (95% CI 33.9 to 51.1%) of children with ADHD symptoms persisting into age 17 who also had impairments in several other areas of development (IQ, social communication, pragmatic language, social behavior) between the ages of 7 and 9 (the most common age group), regardless of polygenic risk (cohort study, n = 9,757, ages 4 to 17, E 2b.)¹⁷³
• psychiatric comorbidity (higher CBCL scores for social behavior, aggression, delinquency, social problems, and cognitive problems), Boys (cohort study, n = 110 with ADHD, n = 105 controls, age 11, industry-funded, E 2b.)¹⁷²
• multiple psychiatric comorbidities (social behavior disorders, anxiety disorders, and mood disorders); lower comorbidity was associated with better coping strategies (narrative review, E 4.)¹⁶⁴
• Comorbidity as an additional predictor starting at school age (narrative review, industry-funded, E 4.)¹⁶⁰
• externalizing and social problems, analyzed using a narrative approach (meta-analysis, k = 36, E 2b.)¹³⁰
• No predictive value for comorbid disorders after adjusting for severity (cross-sectional study, retrospective, n = 3,197, industry-funded, author employed by the manufacturer, E 3.)¹²⁴
• No association with social and emotional problems after controlling for ADHD symptoms (cohort study, n = 116, 6.6 years, E 2b.)⁴⁵

4.1.8. ADHD in Parents and Familial Clustering of ADHD¶

• 69% of girls had a family history of ADHD, compared with 35% in the control group (cohort study, N = 187, n = 96, age 11, industry-funded, E 2b.)⁴⁷
• Familial clustering of ADHD in boys (cohort study, n = 128, 4 years, E 2b.)¹¹⁸
• ADHD in a parent and ADHD severity 6 years later (cohort study, n = 347, 6 years, E 2b.)⁴³
• Familial ADHD in Velo-cardio-facial syndrome (cohort study, n = 72, 4 years, E 2b.)¹¹⁹
• Attention problems in the father (cohort study, n = 457, E 2b.)¹²⁰
• Family history was a significant factor for full-spectrum ADHD (OR 7.56, 95% CI 3.54 to 16.17), but played a lesser role in persistence; women (cross-sectional study, population-based, N = 2,741, E 3.)¹³³
• No predictive value of parental ADHD (cohort study, n = 130, 5 years, E 2b.)¹²³

4.1.9. Comorbid major depression¶

• OR 2.2 for comorbid major depression (cross-sectional study, retrospective, n = 629 with childhood ADHD, industry-funded, E 3.)¹²¹

• OR 1.8 for comorbid major depression (meta-analysis, k = 16, E 2b.)¹⁶¹

• Affective disorders, boys (cohort study, n = 128, 4 years, E 2b.)¹¹⁸

• Comorbid major depression and/or oppositional disorder, persisting from ages 16 to 18 (cohort study, n = 457, E 2b.)¹²⁰

• Childhood depression in Velo-cardio-facial syndrome (cohort study, n = 72, 4 years, E 2b.)¹¹⁹

• Affective Disorders as Part of Multiple Psychiatric Comorbidities (Narrative Review, E 4.)¹⁶⁴

4.1.10. Comorbid anxiety disorders and internalizing disorders¶

• OR 6.9 (95% CI 1.4 to 35.1) for an anxiety disorder before age 16 (persistence rate 89.0% versus 59.3%) (Cross-sectional study, n = 3,309, E 3.)⁷²
• OR 3.59 for social phobia in adults; persistence of the diagnosis after 7 years (cohort study, n = 227, 7 years, E 2b.)⁸⁶
• Significantly higher persistence among children who failed to meet the HKS criteria solely due to an internalizing comorbidity (anxiety, depression) (cohort study [follow-up to an RCT], n = 476 (109 with, 367 without HKS in childhood), 16 years, E 2b.)¹⁶⁶
• Anxiety disorders, boys (cohort study, n = 128, 4 years, E 2b.)¹¹⁸
• Anxiety Disorders as Part of Multiple Psychiatric Comorbidities (Narrative Review, E 4.)¹⁶⁴
• Anxiety disorders, analyzed using a narrative approach (meta-analysis, k = 36, E 2b.)¹³⁰
• No association with anxiety after controlling for ADHD symptoms (cohort study, n = 116, 6.6 years, E 2b.)⁴⁵

4.1.11. Mixed type, or combined symptoms of inattention and hyperactivity¶

• OR 12.4 (4.5 to 34.5) for combined inattention and hyperactivity/impulsivity symptoms compared to hyperactivity-impulsivity symptoms alone (cross-sectional study, retrospective, n = 629 with childhood ADHD, manufacturer-funded, E 3.)¹²¹
• OR 2.4 for combined symptoms of inattention and hyperactivity in childhood compared to symptoms in only one domain (cross-sectional study, retrospective, n = 3,197, industry-funded, author employed by the manufacturer, E 3.)¹²⁴
• OR 1.75 (1.17 to 2.62) for the mixed type compared with the inattentive type (cohort study, n = 399, median 6.58 years, E 2b.)¹²⁵
• 28.2% persistence of hyperactivity symptoms alone (cross-sectional study, retrospective, n = 3,197, manufacturer-funded, author employed by the manufacturer, E 3.)¹²⁴
• 28.0% persistence in cases with inattention symptoms only (cross-sectional study, retrospective, n = 3,197, manufacturer-funded, author employed by the manufacturer, E 3.)¹²⁴
• Mixed type (Narrative review, E 4.)¹⁶⁴
• Combined subtype and hyperactive-impulsive symptoms, evaluated narratively (meta-analysis, k = 36, E 2b.)¹³⁰
• More frequent remission in the hyperactive-impulsive subtype (cohort study, n = 457, E 2b.)¹²⁰

4.1.12. ADHD-related impairment¶

• OR 3.4 for pervasive role impairment in childhood (cross-sectional study, retrospective, n = 3,197, industry-funded, author employed by the manufacturer, E 3.)¹²⁴
• OR 0.92 for impairment (lower among those with complete remission in childhood than among those with complete persistence), adolescence (cohort study, n = 130, 5 years, E 2b.)¹²³
• 63% with severe ADHD-related impairment persisting for over 11 years into young adulthood (mean age 22 years), compared with 33% among those who later remitted, boys (cohort study, n = 110 with ADHD, n = 105 controls, 11 years old, manufacturer-funded, E 2b.)¹⁷²
• Greater impairment at the start of the study was associated with greater ADHD severity 6 years later (cohort study, n = 347, 6 years, E 2b.)⁴³
• Functional impairment in follow-up cohorts with similar severity of symptoms in children (cohort study, n = 485 with ADHD, 665 siblings, 399 controls; partially funded by the manufacturer; E 2b.)¹¹¹
• OR 2.12 (0.99 to 4.54) for ADHD-related impairment, not significant (meta-analysis, k = 36, E 2b.)¹³⁰
• No prediction based on the impairment, girls (cohort study, N = 187, n = 96, age 11, manufacturer-funded, E 2b.)⁴⁷

4.1.13. Low socioeconomic status¶

• OR 1.58 (1.15 to 2.16) for the middle socioeconomic class compared to the upper socioeconomic class, ADHD in young adulthood (prognostic model based on cohort studies, n = 5,113 (development), validation on n = 2,040, n = 3,911, and n = 717, E 2b.)¹²⁹
• OR 1.55 (1.11 to 2.15) for lower versus upper socioeconomic status, ADHD in young adulthood (prognostic model based on cohort studies, n = 5,113 (development), validation on n = 2,040, n = 3,911, and n = 717, E 2b.)¹²⁹
• Lower socioeconomic status is associated with ADHD symptoms and impairment during adolescence and early adulthood (cohort study, n = 116, 6.6 years, E 2b.)⁴⁵
• Parents’ occupation is associated with fewer ADHD symptoms at age 19.8, regardless of baseline symptoms (cohort study, n = 53 with ADHD, n = 50 controls, 7 years, E 2b.)¹⁷⁴
• A higher socioeconomic status was associated with better coping strategies (narrative review, E 4.)¹⁶⁴
• low socioeconomic status, assessed using a narrative approach (meta-analysis, k = 36, E 2b.)¹³⁰
• No association with socioeconomic status or parental education, MTA cohort (cohort study, n = 453, approximately 17 years, E 2b.)¹⁶³
• No predictive value of sociodemographic factors after adjusting for severity (cross-sectional study, retrospective, n = 3,197, manufacturer-funded, author employed by the manufacturer, E 3.)¹²⁴

4.1.14. Higher polygenic risk¶

• β = 0.53 (0.03 to 1.03) for polygenic ADHD risk (PGS), self-reported ADHD symptoms at age 18 (ASRS), independent of maltreatment (birth cohort, n = 4,231, age 18, E 2b.)¹³¹
• β = 0.20 (0.08 to 0.32) for the ADHD-PGS, ADHD symptoms at age 18 as assessed by the mother (SDQ) (birth cohort, n = 4,231, age 18, E 2b.)¹³¹
• A higher polygenic risk of ADHD, independent of multimorbidity, is predictive of persistence through age 17 (cohort study, n = 9,757, ages 4 to 17, E 2b.)¹⁷³
• Polygenic risk for depression in course categories with similar severity of childhood symptoms (cohort study, n = 485 ADHD patients, 665 siblings, 399 controls; partially industry-funded; E 2b.)¹¹¹
• lower genetic risk with a more favorable clinical course (narrative review, E 4.)¹¹⁰

4.1.15. Neuropsychological and motor functions¶

• Explained variance of 5.6% for lower reaction time variability and better functional status over 6 years (mixed-type, mean age 11.5 years, weak predictive power) (Cohort study, n = 226, 6 years, E 2b.)¹⁷⁰
• 3.0% explained variance for improved working memory and reduced ADHD symptom severity over 6 years (cohort study, n = 226, 6 years, E 2b.)¹⁷⁰
• Motor coordination and visual perception in adolescence; persistence of the diagnosis 25 years later (cohort study, n = 19 ADHD, 26 controls, 25 years; full text not available to us, E 2b.)⁵⁴
• Better cognitive flexibility (intra- and extradimensional shifts) at an average age of 12.8 years was associated with fewer ADHD symptoms at age 19.8, regardless of baseline symptoms (cohort study, n = 53 with ADHD, n = 50 controls, 7 years, E 2b.)¹⁷⁴ The reduction in symptoms did not correlate linearly with neuropsychological improvement
• Higher rates of intrusion errors on a verbal list learning test (CVLT-C) in patients with velo-cardio-facial syndrome (cohort study, n = 72, 4 years, E 2b.)¹¹⁹
• Cognitive functions starting in preschool age (narrative review, industry-funded, E 4.)¹⁶⁰
• No prediction based on motor and cognitive inhibition, timing, processing speed, and motor skills (cohort study, n = 226, 6 years, E 2b.)¹⁷⁰
• No predictive value of executive functions (cohort study, n = 130, 5 years, E 2b.)¹²³
• No prediction by other cognitive measures (digit span, reaction time variability, omission and confusion errors, choice impulsivity) (cohort study, n = 116, 6.6 years, E 2b.)⁴⁵

4.1.16. Gender¶

• OR 0.72 (0.58 to 0.89) for female gender and ADHD in young adulthood (prognostic model based on cohort studies, n = 5,113 (development), validation on n = 2,040, n = 3,911, and n = 717, E 2b.)¹²⁹
• Male gender, associated with a high and decreasing symptom class (cohort study, n = 306, 24 years; full text not available to us, E 2b.)¹²⁶
• OR 0.93 (0.71 to 1.23) for male gender, not significant (meta-analysis, k = 36, E 2b.)¹³⁰
• No significant association with gender (cohort study, n = 399, median 6.58 years, E 2b.)¹²⁵

4.1.17. Age at the first onset of ADHD¶

• No connection to a full recovery by age 27 (E 4.)¹⁶⁵

• Early onset was associated with full-spectrum ADHD but played a lesser role in its persistence; women (cross-sectional study, population-based, N = 2,741, E 3.)¹³³

• Age at onset does not predict the outcome, girls (cohort study, N = 187, n = 96, 11 years, manufacturer-funded, E 2b.)⁴⁷

4.1.18. Emotional dysregulation¶

• A profile of deficient emotional self-regulation (CBCL-DESR) at follow-up in 50 of 101 adolescents with persistent ADHD, but in none of the 18 who had remitted (p < 0.001), DESR profile persisted over 4 years in 57% of participants (cohort study, N = 381, n = 177, 4 years, manufacturer-funded, E 2b.)¹⁷⁵
• Greater emotional dysregulation at home when levels remain high versus when they decline (cohort study, n = 413, E 2b.)¹²⁷
• No difference in emotional dysregulation (cohort study, N = 121, n = 20, E 2b.)¹²²

4.1.19. Other factors¶

• Lower psychological resilience and more autistic traits among those with persistent symptoms (cohort study, N = 121, n = 20, E 2b.)¹²² Autistic traits and resilience were negatively correlated

• Physical activity intensity measured by actigraphy (cohort study, n = 116, 6.6 years, E 2b.)⁴⁵

• Grade repetition and special classes, girls (cohort study, N = 187, n = 96, age 11, industry-funded, E 2b.)⁴⁷

• elevated immune and cardiometabolic markers (monocytes, neutrophils, hsCRP, insulin, leptin), with which persistence was primarily associated; women (cross-sectional study, population-based, N = 2,741, E 3.)¹³³

• low birth weight, associated with a high and decreasing symptom class (cohort study, n = 306, 24 years; full text not available to us, E 2b.)¹²⁶

• relatively young age in the school class (month of birth shortly before the school enrollment cutoff date), OR 1.02 (95% CI 0.99 to 1.06), no association (meta-analysis of individual participant data from prospective cohorts, k = 41, n = 4,708, ages 4 to 33, E 2b.)¹⁷⁶ Over time, an ADHD diagnosis among the youngest children in a class was therefore no more frequently found to be incorrect than among the oldest

 

4.2. Brain-Organ Correlations with Persistent ADHD¶

5. The Genetics of Adult ADHD¶

5.1. Are There Genetic Differences Between ADHD in Children and ADHD in Adults?¶

(E 2b): The initially lower heritability estimates in adults were attributed to assessor effects resulting from self-reporting alone. When multiple sources of information were included or in cases of clinically diagnosed ADHD, heritability in adults was comparable to that in children. (Twin study, registry, n = 59,514, E 2b.)²⁰⁷ (Narrative review, E 4.)²⁰⁸ (Narrative review, E 4.)²⁰⁹ Heritability was:

• 0.88 (95% CI 0.83 to 0.92) for clinically diagnosed ADHD across the entire lifespan (twin study, registry, n = 59,514, E 2b.)²⁰⁷
• about 75% in children (narrative review, E 4.)²⁰⁹
• 0.72 (0.56 to 0.84) for clinically diagnosed ADHD in adults (diagnosed at age 18 or older), unshared environment 0.28 (0.16 to 0.44) (twin study, registry, n = 59,514, E 2b.)²⁰⁷
• approximately 30 to 50% in initial estimates of heritability among adults (Narrative Review, E 4.)²⁰⁹
• 30 to 40% (moderate) in twin studies using self-reports from adults (narrative review, E 4.)²⁰⁸
• 30 to 40% in earlier, low heritability estimates for adults, which the authors attribute to rater effects resulting from self-reporting alone (twin study, registry, n = 59,514, E 2b.)²⁰⁷

One individual case involving a clear diagnosis of adult ADHD-C reported complete remission following a COVID-19 infection. After three-quarters of a year, the ADHD symptoms slowly returned. (ADxS experience) It is entirely conceivable that illnesses can alter the dopamine system. Just as certain viral illnesses can be a risk factor for ADHD because they can trigger an extracellular dopamine and norepinephrine deficiency in some brain regions, the reverse is also possible. (ADxS assessment)

Although studies on fluctuating ADHD are rare, fluctuating ADHD could explain the well-known phenomenon in which many people with ADHD stop taking their medication—which they initially found very helpful and necessary—after a few years, only to resume taking it several years later. (Assessment by ADxS)

5.2. A Genetic Perspective on ADHD in Childhood and Adulthood¶

5.3. The Persistence of ADHD into Adulthood from a Genetic Perspective¶