Assessing the Validity of the Age at Onset Criterion for Diagnosing ADHD in DSM-5

Abstract

Objective: Research about Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV) age of onset criterion for ADHD led to increasing that criterion to 12 in Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5). Although an improvement from the previous age at onset of 7, whether this new criterion is adequate to capture all cases with ADHD remains uncertain. Method: We compared three groups of adults: (a) ADHD participants who met all DSM-5 criteria for ADHD (n = 182), (b) late-onset ADHD participants who met all criteria except for later age at onset (n = 17), and (c) non-ADHD participants who did not meet criteria for ADHD (n = 117). We assessed patterns of symptoms, psychiatric comorbidity, functional impairment, familial transmission, quality of life, social adjustment, and intelligence. Results: Compared with non-ADHD participants, all ADHD groups had poorer quality of life and had more impaired social adjustment. Compared with each other, the ADHD groups had similar patterns of psychiatric comorbidity, functional impairment, familial transmission, and intelligence. Conclusion: When assessing adult ADHD, self-reported onsets of ADHD after 12 years of age may be warranted.

Keywords

ADHD onset adult

Introduction

Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5; American Psychiatric Association [APA], 2013) increased the age of onset criterion for ADHD from 7 to 12 due to research showing that cases of ADHD with onset later in childhood were as valid as earlier onsets (Faraone, Biederman, Doyle, et al., 2006; Faraone, Biederman, Spencer, et al., 2006; Faraone, Kunwar, Adamson, & Biederman, 2009; Faraone et al., 2007; Matte et al., 2014). For example, in a series of articles, we documented similar profiles of psychiatric comorbidities, neuropsychological impairment, substance use disorders, family transmission, and personality traits in between adults with ADHD onset prior to age 7 and those with later onsets, primarily in adolescence (Faraone, Biederman, Doyle, et al., 2006; Faraone, Biederman, Spencer, et al., 2006; Faraone et al., 2009; Faraone et al., 2007). Karam et al. (2009) described comparable profile of comorbidities with milder symptoms in late-onset ADHD. Vande Voort, He, Jameson, and Merikangas (2014) compared the prevalence and clinical correlates of Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; APA, 1994) versus DSM-5-defined ADHD in a population sample of U.S. youth. Patients in the study with later ages of onset did not differ from those with earlier ages of onset in terms of severity and patterns of comorbidity, providing further support to extension of age of onset criteria for ADHD in DSM-5. These, and other studies, led DSM-5 to increase the age at onset criterion from 7 to 12.

Although a clear improvement from the DSM-IV age at onset criterion of 7 years of age, some adult patients come to assessment with later ages at onset than 12. However, few studies have assessed the validity of adult ADHD when onset occurs subsequent to age 12. Peyre et al. (2014) compared psychometric properties of 18 ADHD symptoms in the three age of onset groups (<7 years, 7-12 years, and 12-18 years). They concluded that the age of onset criterion could be further raised to 18 years without changing the psychometric properties of ADHD symptoms. Response to treatment with oral release osmotic system (OROS) methylphenidate (MPH) was studied by Biederman, Mick, et al. (2006) who, using an open-label design, reported significant improvement in symptoms of ADHD in patients having a mean age of onset of 14.2 years. Similarly, in a study of atomoxetine, Surman et al. (2010) described statistically and clinically significant improvements in ADHD symptoms in patients with late-onset ADHD or sub-threshold ADHD with a mean age at onset of 13.6 years. In an open-label study by Reinhardt et al. (2007), the response to MPH was statistically better in late-onset adult ADHD patients with a mean age of 37.4 years when compared with typical ADHD patients.

These prior studies provide no clear guidance about the validity of onsets of ADHD later than age 12. For the present study, we hypothesized that comparing late-onset ADHD with DSM-5 onset ADHD would provide support for the validity of diagnosing onset after age 12 based on the retrospective reports of adults. Such data are important, not only for future revisions of the DSM but also for the appropriate diagnosis and management of ADHD among adult patients coded as Other Specified ADHD due to a late onset of symptoms.

Method

Participants

Males and females between the ages of 18 and 55 were eligible for the study. Exclusion criteria were deafness, blindness, psychosis, inadequate command of the English language, or a Full Scale IQ less than 80. We used two ascertainment sources: referrals to psychiatric clinics at the Massachusetts General Hospital (MGH) and advertisements in the Boston area. We recruited non-ADHD probands through advertisements in the Boston area. We ascertained and assessed the probands and all their available biological children, parents, and siblings. The study was approved by the institutional review board at MGH. Every participant, 18 and older, provided signed informed consent. Younger relatives provided signed assent, and parents provided signed informed consent.

Diagnostic Assessment Measures

Trained, lay interviewers, blind to ascertainment status, interviewed all adults (proband and relatives) with the Structured Clinical Interview for DSM-IV (First, Spitzer, Gibbon, & Williams, 1997) and modules from the Schedule for Affective Disorder and Schizophrenia for School-Age Children–Epidemiologic Version (Kiddie-SADS-E; Hodges, McKnew, Burbach, & Roebuck, 1987). For all child relatives, psychiatric data were collected from the mother when available. Child relatives 12 and older were directly evaluated. The interviewers were blind to the participant’s ascertainment group.

The interviewers had undergraduate degrees in psychology and were extensively trained. First, they underwent several weeks of classroom style training, learning interview mechanics, diagnostic criteria, and coding algorithms. Then, they observed interviews by experienced raters and clinicians. They subsequently conducted at least six practice (nonstudy) interviews and at least three study interviews while being observed by senior interviewers. Trainees were not permitted to conduct interviews independently until they executed at least three interviews that achieved diagnostic agreement with an observing senior interviewer. Age at onset was defined as the first emergence of impairing symptoms. Interviewers collected information for psychiatric diagnoses in child relatives (ages 6 to 18) with the Kiddie-SADS-E. Before interviewing for the study, interviewers completed a 4-month training program that included mastery of the instruments, learning about DSM-5 criteria, watching training tapes, observing interviews performed by experienced raters, rating several participants under the supervision of the project coordinator, and completing practice interviews. Throughout the study, they were supervised by board-certified child and adolescent psychiatrists or licensed psychologists.

We computed kappa coefficients of agreement by having experienced, board-certified child and adult psychiatrists and licensed clinical psychologists diagnose participants from audio-taped interviews. Based on 500 assessments from interviews of children and adults, the median kappa coefficient was .98. Kappa coefficients for individual diagnoses included the following: ADHD (0.88), conduct disorder (1.0), major depression (1.0), mania (0.95), separation anxiety (1.0), agoraphobia (1.0), panic (0.95), substance use disorder (1.0), and tics/Tourette’s (0.89). Initial diagnoses were prepared by the study interviewers and were then reviewed by a diagnostic committee of board-certified child and adolescent psychiatrists or licensed psychologists. The diagnostic committee was blind to the participant’s ascertainment group, all data collected from other family members, and all nondiagnostic data (e.g., cognitive functioning). Diagnoses were made for two points in time: lifetime and current (past month). The structured interview data were reviewed by the diagnostic committee so that it could make a best estimate diagnosis as described by Leckman et al. (1982). Definite diagnoses were assigned to participants who met all diagnostic criteria. Diagnoses were considered definite only if a consensus was achieved that criteria were met to a degree that would be considered clinically meaningful. By “clinically meaningful” we mean that the data collected from the structured interview indicated that the diagnosis should be a clinical concern due to the nature of the symptoms, the associated impairment, and the coherence of the clinical picture.

Other Assessment Measures

Using the methods of Sattler (1988), we estimated Full Scale IQ from the vocabulary and block design subtests of the Wechsler Adult Intelligence Scales–Third Edition (WAIS-III; Wechsler, 1997). Our interviewers assessed academic achievement with the Arithmetic and Reading Subtests of the Wide Range Achievement Test–Third Edition (Jastak & Jastak, 1993).

The Quality of Life Enjoyment and Satisfaction Questionnaire–Short Form (Q-LES-QSF; Endicott, Nee, Harrison, & Blumenthal, 1993) is an extensively used questionnaire that assesses the quality of life in several domains. Our research group previously described evidence for the validity of the Q-LES-QSF as a measure quality of life associated with ADHD (Mick, Spencer, Faraone, Biederman, & Zhang, 2007).

The Social Adjustment Scale–Self Report (SAS-SR; Weissman, 1999) is an extensively applied 54-question scale that assesses adaptation to and satisfaction with adult social roles. The scale evaluates six major areas of functioning for a participant, over the past 2-week period: work (capturing function related to work as an employee, housewife, or student), social and leisure activities, relationship with extended family, intimate relationship, parental role, and role within the family unit. Within each role, questions assess performance at role tasks, friction with others, other aspects of interpersonal relationships, as well as feelings about and satisfaction with the role. T-scores higher than 65 indicates significant impairment.

To measure deficient emotional self-regulation (DESR), we used eight items from the self-report Current Behavior Scale developed by Barkley (1997) to assess DESR as we have elsewhere (Surman et al., 2011). We chose those items designated by Barkley (R.A.B., personal communication) as measuring DESR. The scale asks participants to describe their behavior in the prior 6 months. Responses to each item on this scale range from 0 (never or rarely) to 3 (very often).

Statistical analyses

Our analyses used logistic regression for binary outcomes, ordinal logistic regression for ordinal outcomes, Poisson regression for count data, multinomial logistic regression for categorical outcomes, and Gaussian regression for continuous outcomes. For each psychiatric disorder in Table 2, we used Cox Proportional Hazard models to predict age of onset and proportion at risk. We then permuted group membership across participants and recalculated our omnibus and pairwise statistics for permutation tests. Permutation testing calculates the probability of achieving a result through random assignment of participants into groups by measuring the number of recalculated results greater than the observed statistic during a prescribed number of iterations. Using permutation tests allowed us to reinforce our results with low frequencies of psychiatric disorders in our not-ADHD and subthreshold ADHD groups. For the same reasons, we used Fisher’s exact test to analyze differences in frequency of participants receiving different treatment options, numbers of relatives with full ADHD, and subthreshold ADHD across ADHD diagnosis groups.

Because multiple members of a single family cannot be considered independent of one another because they share genetic, cultural, and social risk factors, we used Huber–White robust estimates of variance in analyses using relatives so that p values would be accurately estimated. We used the following strategy to balance our risk for Type I and Type II errors when adjusting for multiple comparisons. For each domain of analysis (as defined by the tables), we applied the Bonferroni criterion to the omnibus test for each variable in the domain. If that was significant, we used the .05 alpha level to assert significance for pairwise comparisons.

Results

The probands (n = 316) were divided into three groups based on retrospective reports of ADHD symptoms. The subgroups included full ADHD (n = 182 probands, n = 108 relatives), if they met the DSM-5 symptom and impairment criteria for ADHD with onset before 12 years of age; late-onset ADHD (n = 17 probands, n = 11 relatives), if they met all the criteria for ADHD other than age of onset. The remaining participants were classified as not-ADHD (n = 117 probands, n = 133 relatives). There was no statistical difference between the three groups in age of relatives, gender of probands, or gender of relatives. The not-ADHD participants were younger, were more likely to be never married, were more likely to be Caucasians, and had a lower socioeconomic status (Table 1). We adjusted for demographic differences in subsequent analyses.

Table 1.

Demographic Features.

	Not ADHD	Late-onset ADHD	Full ADHD	Test statistic	Omnibus p value
	M ± SD	M ± SD	M ± SD	Test statistic	Omnibus p value
Age of proband	29.8 ± 8.9	36.4 ± 10.0^a(**⁾	36.0 ± 10.8^b(***⁾	χ² = 24.95	<.001
Age of relatives	31.9 ± 18.5	23.3 ± 6.8	23.9 ± 18.8	χ² = 2.5	.29
Socioeconomic status	1.7 ± 0.6	2.5 ± 1.2^a(**⁾	2.0 ± 1.0^b(**⁾	χ² = 13.82	.001
	N (%)	N (%)	N (%)
Marital status				χ²(4) = 12.27	.02
Never married	90 (77)	9 (53)	106 (58)
Married	18 (15)	6 (35)	48 (26)
Divorced	9 (8)	2 (12)	28 (15)
Ethnicity				χ² = 6.95	.03
White	91 (77)	14 (82)	164 (89)
Other	27 (23)	3 (18)	21 (11)
Gender of proband
n (%) male	56 (47)	10 (59)	93 (50)	χ²(2) = 0.82	.7
Gender of relatives
n (%) male	32 (40)	2 (22)	97 (47)	χ²(2) = 2.93	.2

Note. For pairwise comparisons:

Controls versus late-onset ADHD.

Controls versus full ADHD.

p ≤ .05. **p ≤ .01. ***p ≤ .001.

Our first familial test of the validity of late-onset ADHD was to compare rates of full ADHD among relatives of each group. The rates of full ADHD among relatives were 50% for relatives of late-onset ADHD probands and 54% for relatives of full ADHD probands. Although the rates of full ADHD in relatives did not differ between these two ADHD proband groups, the risk to relatives of controls (9%) was significantly lower than the risk to relatives in each ADHD proband group (both ps < .01). Our second familial test of the validity of late-onset ADHD compared the risk for late-onset ADHD between relatives of probands having full ADHD and the relatives of controls. The difference was significant, 12% vs. 2%; χ²(1) = 4.1, p = .04.

Psychiatric Comorbidities

There were no significant differences in lifetime psychiatric comorbidities between the late-onset and full ADHD groups (Table 2). With a few exceptions, most of the comorbidities were significantly more prevalent in each ADHD group compared with the control group. The total number of lifetime disorders was lowest for controls (1.2 ± 1.3), and higher for the late-onset ADHD (4.4 ± 2.7) and full ADHD (3.9 ± 2.3) groups (χ² = 66.2, p < .001). The two ADHD groups did not differ significantly from one another but each ADHD group differed significantly from the controls (both ps < .001).

Table 2.

Lifetime Comorbid Disorders.

	Not ADHD	Late-onset ADHD	Full ADHD	Test statistic	Omnibus p value
Major depressive disorder	7 (6)	6 (35)^a***	58 (31)^b***	21.08	<.001
Bipolar disorder	0 (0)	3 (18)	8 (4)	4.53	.1
Oppositional defiant disorder	1 (1)	5 (29)^a***	66 (36)^b***	17.62	<.001
Conduct disorder	3 (3)	4 (24)^a**	51 (28)^b***	20.72	<.001
Alcohol abuse	38 (32)	9 (53)	101 (55)^b***	16.0	<.001
Alcohol dependence	7 (6)	4 (24)^a*	52 (28)^b***	18.54	<.001
Drug abuse	20 (17)	8 (47)^a**	83 (45)^b***	24.57	<.001
Drug dependence	7 (6)	2 (12)	39 (21)^b***	12.44	.002
Agoraphobia	3 (3)	2 (12)	27 (15)^b**	8.17	.02
Social phobia	10 (9)	8 (47)^a***	51 (28)^b***	22.43	<.001
Simple phobia	16 (14)	5 (29)	36 (19)	3.36	.186
Panic disorder	7 (6)	4 (24)^a*	34 (18)^b**	8.61	.014
Generalized anxiety disorder	4 (3)	5 (29)^a***	45 (24)^b***	16.84	<.001

Note. For pairwise comparisons:

Controls versus late-onset ADHD.

Controls versus full ADHD.

Late-onset ADHD versus full ADHD.

p ≤ .05. **p ≤ .01. ***p ≤ .001.

Regarding current psychiatric comorbidity, only social phobia was more prevalent in the late-onset ADHD group compared with controls (Table 3). The prevalence of substance abuse, agoraphobia, social phobia, and generalized anxiety disorder was greater in the full ADHD group compared with controls. We found no significant difference in the current prevalence of comorbidities between the full ADHD and late-onset ADHD groups (Table 3). The total number of current disorders was lowest for controls (0.2 ± 0.6), and higher for the late-onset ADHD (1.4 ± 1.1) and full ADHD (1.1 ± 1.5) groups, χ²(2) = 28.1, p < .001. The two ADHD groups did not differ significantly from one another but each differed significantly from the controls (both ps < .001).

Table 3.

Current Comorbid Disorders.

	Not ADHD	Late-onset ADHD	Full ADHD	χ²(2)	Omnibus p value
Major depressive disorder	0 (0)	0 (0)	13 (7)	0.00	1.0
Bipolar disorder	0 (0)	1 (6)	5 (3)	0.51	.77
Alcohol abuse	3 (3)	1 (6)	8 (4)	1.71	.42
Alcohol dependence	1 (1)	0 (0)	3 (2)	0.57	.75
Drug abuse	2 (2)	0 (0)	11 (6)^b*	4.8	.091
Drug dependence	7 (6)	2 (12)	39 (21)	0.39	.82
Agoraphobia	2 (2)	2 (12)	18 (10)^b*	4.91	.09
Social phobia	4 (4)	4 (24)^a*	29 (16)^b*	7.4	.025
Simple phobia	13 (12)	3 (18)	22 (13)	0.35	.839
Panic disorder	2 (2)	2 (12)	9 (5)	3.31	.19
Generalized anxiety disorder	1 (1)	2 (12)	27 (16)^b**	6.94	.031

Note. For pairwise comparisons:

Controls versus late-onset ADHD.

Controls versus full ADHD.

Late-onset ADHD versus full ADHD.

p ≤ .05. **p ≤ .01. ***p ≤ .001.

Functional Impairments and Quality of Life

Both ADHD groups required more special classes and academic tutoring compared with controls (Table 4). In addition, the full ADHD group reported repeated grades and arithmetic disability significantly more than controls. Both ADHD groups were more likely to have been arrested compared with controls (Table 4) whereas only the full ADHD group reported significantly more tickets and accidents compared with controls. We found no significant difference in these functional impairments between the two ADHD groups in any domain (Table 4).

Table 4.

Functional Impairments.

School	Not ADHD	Late-onset ADHD	Full ADHD	χ²(2)	p value
School	n (%)	n (%)	n (%)	χ²(2)	p value
Arithmetic disability	9 (8)	4 (24)	32 (18)^b*	5.54	.063
Reading disability	1 (1)	0 (0)	7 (4)	2.36	.31
Repeated grade	6 (5)	1 (6)	41 (22)^b***	13.30	.001
Special class	2 (2)	3 (18)^a**	26 (14)^b**	9.90	.007
Academic tutoring	24 (20)	0 (47)^a*	87 (48)^b***	23.48	<.001
	M ± SD	M ± SD	M ± SD
Legal
Ever arrested	0.1 ± 0.2	0.4 ± 0.5^a***	0.3 ± 0.5^b***	20.27	<.001
Ever convicted	0.0 ± 0.1	0.2 ± 0.4	0.1 ± 0.3	3.60	.1655
No. of convictions	1.0 ± 1.4	1.3 ± 1.5	0.9 ± 1.1	0.38	.8279
No. of imprisoned	0.0 ± 0.0	0.0 ± 0.0	0.1 ± 0.4	0.00	1.0
Prison term duration	0.0 ± 0.0	0.0 ± 0.0	0.2 ± 0.7	0.00	1.0
Ever imprisoned	0.0 ± 0.0	0.0 ± 0.0	0.1 ± 0.2	0.00	1.0
Driving
Tickets	2.4 ± 1.4	2.9 ± 1.8	3.7 ± 1.9^b***	20.98	<.001
Accidents	1.1 ± 1.2	1.9 ± 1.4	2.6 ± 1.6^b***	32.76	<.001

Note. For pairwise comparisons:

Controls versus late-onset ADHD.

Controls versus full ADHD.

Late-onset ADHD versus full ADHD.

p ≤ .05. **p ≤ .01. ***p ≤ .001.

The late-onset ADHD group scored significantly worse on sense of well-being, overall life satisfaction, and functioning in daily life compared with the full ADHD group (Table 5). Compared with controls, both ADHD groups scored worse on quality of life ratings for physical health, mood, work, household activities, social relationships, family relationships, leisure time activities, sex drive and interest, economic status, living/housing situation, vision, medications, sense of well-being, overall life satisfaction, and functioning in daily life when compared with control group (Table 5).

Table 5.

Quality of Life.

	Not ADHD	Late-onset ADHD	Full ADHD	χ²(2)	Omnibus p value
	M ± SD	M ± SD	M ± SD	χ²(2)	Omnibus p value
Physical health	4.2 ± 0.8	3.4 ± 0.8^a**	3.7 ± 1.0^b***	15.78	<.001
Mood	4.0 ± 0.7	2.6 ± 1.0^a***	3.2 ± 1.0^b***	37.3	<.001
Work	3.8 ± 0.9	2.6 ± 1.1^a***	3.2 ± 1.2^b***	24.45	<.001
Household activities	3.9 ± 0.7	2.7 ± 1.0^a***	3.1 ± 1.1^b***	32.49	<.001
Social relationships	4.0 ± 0.8	2.8 ± 1.2^a***	3.3 ± 1.2^b***	27.49	<.001
Family relationships	4.1 ± 0.8	2.9 ± 1.1^a***	3.4 ± 1.2^b***	24.02	.001
Leisure time activities	3.9 ± 0.8	2.6 ± 0.9^a***	3.2 ± 1.2^b***	20.12	<.001
Sex drive, interest, perform	4.0 ± 0.8	3.1 ± 1.2^a*	3.2 ± 1.3^b***	16.86	<.001
Economic status	3.4 ± 0.8	2.3 ± 1.2^a***	2.8 ± 1.2^b***	20.61	<.001
Living/housing situation	3.9 ± 0.9	3.3 ± 1.0^a*	3.5 ± 1.2^b**	10.42	.006
Vision	4.5 ± 0.6	3.6 ± 0.9^a***	4.0 ± 1.0^b***	17.86	<.001
Medication	4.3 ± 0.6	3.6 ± 0.5^a*	3.6 ± 0.9^b**	8.74	.013
Sense of well-being	4.3 ± 0.6	2.9 ± 0.7^a**^c	3.4 ± 1.0^b**^c	48.75	<.001
Overall life satisfaction	4.1 ± 0.6	2.6 ± 1.1^a*^c	3.3 ± 1.1^b*^c	41.51	<.001
Functioning in daily life	4.4 ± 0.6	2.9 ± 0.8^a**^c	3.5 ± 1.0^b**^c	51.24	<.001

Note. For pairwise comparisons:

Controls versus late-onset ADHD.

Controls versus full ADHD.

Late-onset ADHD versus full ADHD.

p ≤ .05. **p ≤ .01. ***p ≤ .001.

The late-onset ADHD group scored significantly worse on ratings of work functioning, social/leisure functioning, and total social adjustment scores when compared with full ADHD group (Table 6). Compared with controls, both ADHD groups scored worse on all measures of social adjustment: work functioning, social/leisure functioning, extended family functioning, primary relationship functioning, parenting functioning, and family unit functioning (Table 6).

Table 6.

Social Adjustment Scores.

	Not ADHD	Late-onset ADHD	Full ADHD	χ²(2)	Omnibus p value
	M ± SD	M ± SD	M ± SD	χ²(2)	Omnibus p value
Work functioning	52.3 ± 9.5	83.1 ± 20.4^a**^c	65.6 ± 17.6^b*^c	41.83	<.001
Social/leisure functioning	48.6 ± 6.3	66.7 ± 13.2^a**^c	59.1 ± 12.6^b**^c	41.74	<.001
Extended family functioning	58.8 ± 11.7	80.8 ± 19.5^a***	71.8 ± 18.3^b***	37.84	<.001
Primary relationship functioning	49.7 ± 10.1	64.4 ± 14.6^a**	62.1 ± 14.1^b***	13.74	.001
Parenting functioning	54.2 ± 12.3	61.2 ± 14.4	56.9 ± 12.1	1.78	.41
Family unit functioning	52.8 ± 11.7	69.5 ± 15.4^a***	65.2 ± 13.9^b***	17.7	<.001
Total	51.7 ± 8.1	76.7 ± 15.5^a**^c	67.1 ± 14.4^b**^c	58.35	<.001

Note. For pairwise comparisons:

Controls versus late-onset ADHD.

Controls versus full ADHD.

Late-onset ADHD versus full ADHD.

p ≤ .05. **p ≤ .01. ***p ≤ .001.

Other Clinical Features

The late-onset ADHD group scored lower on lifetime reports of hyperactive and inattentive symptoms compared with full ADHD group (Table 7). Compared with controls, both ADHD groups reported more symptoms of emotional dysregulation, hyperactivity-impulsivity, and inattentiveness. Compared with late-onset ADHD, the full ADHD group had significantly more hyperactive-impulsive and inattentive symptoms (Table 7). Compared with controls, both ADHD groups scored significantly worse on neuropsychological tests of reading, arithmetic, and intelligence (Table 7), but the two ADHD groups did not differ from one another.

Table 7.

DESR, Symptoms, Intelligence, and Achievement.

	Not ADHD	Late-onset ADHD	Full ADHD	χ²(2)	Omnibus p value
	M ± SD	M ± SD	M ± SD	χ²(2)	Omnibus p value
Emotion	3.0 ± 2.7	13.3 ± 5.5^a***	11.4 ± 5.5^b***	72.03	<.001
Hyperactive-impulsive symptoms^d	0.4 ± 0.7	4.6 ± 2.4^a***	6.0 ± 2.4^b**^c	51.67	<.001
Inattentive symptoms^d	0.3 ± 0.7	6.4 ± 1.4^c***	7.8 ± 1.4^c***	12.89	.002
WRAT reading scale score	108.6 ± 8.7	104.4 ± 8.4	105.5 ± 10.2	3.76	.15
WRAT Arithmetic Scale Score	106.8 ± 12.2	92.5 ± 12.6^a***	97.6 ± 13.9^b***	22.85	<.001
WISC block design estimated IQ	109.6 ± 14.2	101.7 ± 15.9	106.3 ± 15.6	2.72	.26
WISC vocabulary estimated IQ	119.2 ± 13.2	108.4 ± 14.2^a*	114.7 ± 14.6	6.23	.04
WISC Full Scale IQ	116.3 ± 12.5	105.9 ± 13.7^a*	112.0 ± 14.2	6.0	.05

Note. DESR = Deficient Emotional Self-Regulation; WRAT=Wide Range Achievement Test; WISC=Wechsler Intelligence Scale for Children.

For pairwise comparisons:

Controls versus late-onset ADHD.

Controls versus full ADHD.

Late-onset ADHD versus full ADHD.

Based on lifetime symptom report.

p ≤ .05. **p ≤ .01. ***p ≤ .001.

Discussion

Our data about the clinical features of patients and the pattern of transmission of ADHD among relatives provide evidence for the validity of diagnosing ADHD in adults who self-report onset at age 12 or later. This was consistent across several domains of validation, including family history, psychiatric comorbidity, functional impairments, quality of life, emotional dysregulation, ADHD symptoms, and neuropsychological functioning. These findings highlight the complexities of assessing age at onset retrospectively in adults.

Comparisons With Non-ADHD Controls

The fraction of participants meeting full criteria for ADHD at the time of the interview did not differ between the late-onset and full ADHD groups (which were defined based on a lifetime history of ADHD). When compared with the not-ADHD participants, both groups had an increased lifetime prevalence of mood, impulse control, substance use, and anxiety disorders, each of which are known to be elevated among ADHD youth (Faraone et al., 2015; Jensen et al., 2001), clinically referred ADHD adults (Biederman et al., 1993), and ADHD youth followed into adulthood (Biederman et al., 2010; Cheung et al., 2015). Compared with controls, we found evidence of significant and similar functional impairments in both the late-onset and full ADHD groups on the SAS. These impairments were in domains known to affect ADHD children, adolescents, and adults: school functioning, legal difficulties, and traffic accidents. Compared with the not-ADHD group, both the full and late-onset ADHD groups had achieved lower grades and occupational levels. These results confirm other studies showing ADHD adults to be at risk for driving difficulties (Jerome, Habinski, & Segal, 2006), legal problems (Chang, Lichtenstein, D’Onofrio, Sjolander, & Larsson, 2014), and school failure (Biederman, Faraone, et al., 2006; Fredriksen et al., 2014). In terms of quality of life assessment, both the full ADHD and late-onset ADHD patients scored lower than the not-ADHD subgroup in all quality of life ratings.

These comparisons of clinical features between the two ADHD groups and the non-ADHD controls show that the late-onset cases have many of the clinical features seen in patients meeting full criteria for the disorder, which provides evidence for the validity of the late-onset diagnoses. Consistent with this, our two tests of familial transmission supported the validity of retrospectively reported, late-onset ADHD: (a) Relatives of late-onset patients had a significantly elevated risk for full ADHD, and (b) relatives of full ADHD patients had an increased risk for late-onset ADHD. These familial findings argue against the idea that late-onset cases are partial mimics of ADHD.

Differences Between Late-Onset ADHD and Full ADHD

If late-onset ADHD were secondary to another disorder, we should have found an increased risk for psychiatric comorbidity among the late-onset cases. We did not. We found no significant differences, and in most cases, rates of disorders were similar between the two ADHD groups. For example, the total number of lifetime comorbid disorders was 4.4 for late-onset ADHD and 3.9 for full ADHD. For current comorbid disorders, these numbers were 1.4 and 1.1, respectively. These data do not suggest that other psychiatric disorders are mimicking a late-onset form of ADHD.

The late-onset cases had fewer current inattentive symptoms and fewer current hyperactive-impulsive symptoms. These findings suggest that they had a milder form of ADHD, which is expressed with fewer symptoms having a later onset. Yet despite these lower levels of symptoms and similar levels of psychiatric comorbidity, we did find evidence of worse outcomes for late-onset ADHD in social functioning and quality of life compared with the full ADHD group. On the SAS, the late-onset group reported worse functioning in work and in social/leisure activities. On the Quality of Life scale, the late-onset-ADHD group scored lower than the full ADHD group in their ratings of well-being, overall life satisfaction, and functioning in daily life. Thus, late-onset ADHD presents with fewer symptoms that cause greater levels of impairment.

Clinical and Research Implications

As McGough and Barkley (2004) concluded, clinicians should be flexible when applying DSM criteria for ADHD to adults. Our data suggest that such flexibility is important regarding DSM-5’s new age at onset criterion when diagnosing adults with ADHD. Like prior studies, ours was not designed to determine what age at onset criterion would be the most valid. Such a study would require sufficient numbers of participants for each age category to be considered as potentially valid. Ideally, the next DSM field trial for ADHD would incorporate such sampling considerations.

An alternative interpretation of our findings is that the DSM-5 age at onset criterion for ADHD is correct but that, due to recall biases, patients that actually onset prior to age 12 report their ages at onset to be greater than 12. If that idea is correct, then increasing the reliability of the age at onset assessment should decrease evidence for the validity of late-onset ADHD. The reliability of an age at onset assessment could be increased by using multiple informants or by selecting adult participants from prospective follow-up studies. Although our study cannot determine if such a bias exists, data from longitudinal follow-up studies would be able to make that determination. The complete elimination of recall biases will not be possible until the development of valid biomarkers. In their absence, future work should develop more sensitive self-report measures of age at onset. For example, it is possible that retrospective recall of childhood impairment might be more reliably assessed than self-reported age of onset of symptoms.

Since DSM-5 modified the diagnostic criteria for ADHD, there have been studies showing an increase in the prevalence of ADHD that have been explained as being due to a decrease in the diagnostic threshold for the disorder. Van de Glind et al. (2014) reported an increase in prevalence from 5.4% to 7.6% in Hungary and from 31.3% to 32.6% in Norway in treatment seeking substance use disorder patients. Matte et al. (2014) found a 27% increase in the expected prevalence of ADHD among young adults, comparing DSM-IV with DSM-5 criteria. In our study, the prevalence of late onset ADHD among relatives of controls was only 2%. This provides a rough estimate of how the population prevalence might increase if the age at onset criterion were increased. Along with an increase in prevalence, there would likely be an increase in the heterogeneity of an already heterogeneous disorder which could create a disconnect with the current, robust literature about childhood onset adult ADHD.

Various authors have commented upon the implications of increasing the age of onset to 12. Epstein and Loren (2013) stated that patient’s or parent’s inability to recall the age of onset prior to 7 gave false negative results, and the increase might reduce some of the diagnostic issues relating to recalling the age of onset. Coghill and Seth (2011) expressed their support for the DSM-5 criteria if they result in more ADHD patients receiving help. Both groups of investigators mentioned the arbitrariness of the age at onset criterion along with need for studies assessing the performance of the DSM-5 criteria in the population.

Is adult onset ADHD a distinct entity?

Our data indicate that adults who do not recall onset of ADHD in childhood but report ADHD symptoms in adulthood have a profile of psychiatric comorbidity, familial transmission, and functional impairment that is similar to that seen in those meeting the age at onset diagnostic criterion. Does this confirm the existence of an adult onset ADHD that occurs without preceding ADHD symptoms in childhood as has been asserted by others (Moffitt et al., 2015)? We think not. Our results suggest that such diagnoses are valid, but they cannot determine whether these people truly had later onset or could not recall their symptoms.

The existence of adult onset ADHD was hypothesized by a longitudinal population study that followed a birth cohort of 1,037 individuals to age 38 (Moffitt et al., 2015). At the adult follow-up, they diagnosed 31 participants with ADHD who had not been diagnosed with ADHD at earlier assessments during childhood or adolescence. Yet, compared with non-ADHD controls, this adult onset ADHD group had more teacher (but not parent) - rated symptoms of ADHD and more conduct disorder in childhood and were more likely than controls to have had a combined parent/teacher report of ADHD symptom onset prior to age 12. Thus, it is not clear what fraction of these cases had true adult onset of ADHD and in what fraction the onset had been missed at earlier assessments. Nevertheless, such data call for more research to better define age at onset and to better define the subset of adults who show no evidence for onset of ADHD in childhood.

Limitations

The total number of study participants was large, but our sample of late-onset ADHD cases was small. Although this limits our power to detect smaller effects involving this group, we did find many significant differences, which suggest that power was adequate for clinically significant effects. Because our sample was referred for ADHD, our results cannot be generalized to nonreferred samples. An important method for diagnostic validation is the use of longitudinal studies to establish predictive validity. Future work should include this method when assessing the validity of late-onset diagnoses.

Our diagnoses of adult ADHD relied entirely on the self-report of adult participants. Thus, these findings may not generalize to diagnoses defined using data from informants or longitudinal assessments. Another limitation to generalizability is that our sample was skewed toward Caucasian participants belonging to middle or higher income group families and was not representative of lower socioeconomic strata or other ethnicities.

Despite these limitations, by using converging evidence from multiple domains in a family study design, we can reasonably conclude that there is evidence for the validity of diagnosing ADHD among adults who report all DSM-5 criteria for the disorder except for criterion later age at onset. More work is needed to better understand the nature of this subgroup and how it should be defined. Although choosing an optimal age at onset criterion requires more research, clinicians should not dismiss the diagnosis of ADHD in adults when onset occurs later than allowed by DSM-5. Our findings and those of others discussed above should be taken into consideration in planning field trials for future revisions of the DSM.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported in part by grants to S.V. Faraone from the National Institute of Health (R01MH57934).Professor Faraone was supported by the K.G. Jebsen Centre for Research on Neuropsychiatric Disorders, University of Bergen, Norway, the European Community’s Seventh Framework Programme (FP7/2007-2013) under Grant Agreement 602805 and U.S. NIMH Grant R01MH094469.

Author Biographies

Shanel Chandra, MD, is a board-certified psychiatrist in the Department of Psychiatry at Upstate Medical University. Her primary clinical and research interests include the study of mental health issues in children.

Joseph Biederman is chief of the Clinical and Research Programs in Pediatric Psychopharmacology and Adult ADHD at the Massachusetts General Hospital, director of the Alan and Lorraine Bressler Clinical and Research Program for Autism Spectrum Disorders at the Massachusetts General Hospital, and professor of Psychiatry at Harvard Medical School. Dr. Biederman is Board Certified in General and Child Psychiatry.

Stephen V. Faraone, PhD, is a distinguished professor in the Departments of Psychiatry and Neuroscience & Physiology at SUNY Upstate Medical University. He is also senior scientific advisor to the Research Program Pediatric Psychopharmacology at the Massachusetts General Hospital and a lecturer at Harvard Medical School. Dr. Faraone studies the nature and causes of mental disorders in childhood and has made contributions to research in psychiatric genetics, psychopharmacology, diagnostic issues and methodology.

References

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.

Barkley

(1997). ADHD in adults: Comorbidity and adaptive impairments (Grant number1R01MH054509-054501A054502). National Institute for the Mentally Handicapped.

Biederman

Faraone

S. V.

Spencer

T. J.

Mick

Monuteaux

M. C.

Aleardi

(2006). Functional impairments in adults with self-reports of diagnosed ADHD: A controlled study of 1001 adults in the community. Journal of Clinical Psychiatry, 67, 524-540. Retrieved from http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&;db=PubMed&dopt=Citation&list_uids=16669717

Biederman

Faraone

S. V.

Spencer

T. J.

Wilens

Norman

Lapey

K. A.

. . . Doyle

A. E.

(1993). Patterns of psychiatric comorbidity, cognition, and psychosocial functioning in adults with attention deficit hyperactivity disorder. American Journal of Psychiatry, 150, 1792-1798.

Biederman

Mick

Spencer

Surman

Hammerness

Doyle

. . . Schweitzer

(2006). An open-label trial of OROS methylphenidate in adults with late-onset ADHD. CNS Spectrums, 11, 390-396. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/16641844

Biederman

Petty

C. R.

Monuteaux

M. C.

Fried

Byrne

Mirto

. . . Faraone

S. V.

(2010). Adult psychiatric outcomes of girls with attention deficit hyperactivity disorder: 11-year follow-up in a longitudinal case-control study. American Journal of Psychiatry, 167, 409-417. doi:10.1176/appi.ajp.2009.09050736

Chang

Lichtenstein

D’Onofrio

B. M.

Sjolander

Larsson

(2014). Serious transport accidents in adults with attention-deficit/hyperactivity disorder and the effect of medication: A population-based study. JAMA Psychiatry, 71, 319-325. doi:10.1001/jamapsychiatry.2013.4174

Cheung

C. H.

Rijdijk

McLoughlin

Faraone

S. V.

Asherson

Kuntsi

(2015). Childhood predictors of adolescent and young adult outcome in ADHD. Journal of Psychiatric Research, 62, 92-100. doi:10.1016/j.jpsychires.2015.01.011

10.

Coghill

Seth

(2011). Do the diagnostic criteria for ADHD need to change? Comments on the preliminary proposals of the DSM-5 ADHD and Disruptive Behavior Disorders Committee. European Child & Adolescent Psychiatry, 20, 75-81. doi:10.1007/s00787-010-0142-4

11.

Endicott

Nee

Harrison

Blumenthal

(1993). Quality of Life Enjoyment and Satisfaction Questionnaire: A new measure. Psychopharmacology Bulletin, 29, 321-326.

12.

Epstein

J. N.

Loren

R. E.

(2013). Changes in the definition of ADHD in DSM-5: Subtle but important. Neuropsychiatry (London), 3, 455-458. doi:10.2217/npy.13.59

13.

Faraone

S. V.

Asherson

Banaschewski

Biederman

Buitelaar

J. K.

Ramos-Quiroga

J. A.

. . . Franke

(2015). Attention-deficit/hyperactivity disorder. Nature Reviews: Disease Primers, 1, 15020. doi:10.1038/nrdp.2015.20.

14.

Faraone

S. V.

Biederman

Doyle

Murray

Petty

Adamson

J. J.

Seidman

(2006). Neuropsychological studies of late onset and subthreshold diagnoses of adult attention-deficit/hyperactivity disorder. Biological Psychiatry, 60, 1081-1087.

15.

Faraone

S. V.

Biederman

Spencer

Mick

Murray

Petty

. . . Monuteaux

M. C.

(2006). Diagnosing adult attention deficit hyperactivity disorder: Are late onset and subthreshold diagnoses valid? American Journal of Psychiatry, 163, 1720-1729; quiz 1859.

16.

Faraone

S. V.

Kunwar

Adamson

Biederman

(2009). Personality traits among ADHD adults: Implications of late-onset and subthreshold diagnoses. Psychological Medicine, 39, 685-693. doi:10.1017/S0033291708003917

17.

Faraone

S. V.

Wilens

T. E.

Petty

Antshel

Spencer

Biederman

(2007). Substance use among ADHD adults: Implications of late onset and subthreshold diagnoses. American Journal on Addictions, 16(Suppl. 1), 24-34. doi:10.1080/10550490601082767

18.

First

Spitzer

Gibbon

Williams

(1997). Structured Clinical Interview for DSM-IV Axis I Disorders. Washington, DC: American Psychiatric Press.

19.

Fredriksen

Dahl

A. A.

Martinsen

E. W.

Klungsoyr

Faraone

S. V.

Peleikis

D. E.

(2014). Childhood and persistent ADHD symptoms associated with educational failure and long-term occupational disability in adult ADHD. Attention Deficit and Hyperactivity Disorders, 6, 87-99. doi:10.1007/s12402-014-0126-1

20.

Hodges

McKnew

Burbach

Roebuck

(1987). Diagnostic concordance between the Child Assessment Schedule (CAS) and the Schedule for Affective Disorders and Schizophrenia for school-age children (K-SADS) in an outpatient sample using lay interviewers. Journal of the American Academy of Child & Adolescent Psychiatry, 26, 654-661.

21.

Jastak

(1993). Wide Range Achievement Test (3rd ed). Wilmington, DE: Jastak Associates.

22.

Jensen

P. S.

Hinshaw

S. P.

Kraemer

H. C.

Lenora

Newcorn

J. H.

Abikoff

H. B.

. . . Vitiello

(2001). ADHD comorbidity findings from the MTA study: Comparing comorbid subgroups. Journal of the American Academy of Child & Adolescent Psychiatry, 40, 147-158.

23.

Jerome

Habinski

Segal

(2006). Attention-deficit/hyperactivity disorder (ADHD) and driving risk: A review of the literature and a methodological critique. Current Psychiatry Reports, 8, 416-426. Retrieved from http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&;db=PubMed&dopt=Citation&list_uids=16968625

24.

Karam

R. G.

Bau

C. H.

Salgado

C. A.

Kalil

K. L.

Victor

M. M.

Sousa

N. O.

. . . Grevet

E. H.

(2009). Late-onset ADHD in adults: Milder, but still dysfunctional. Journal of Psychiatric Research, 43, 697-701. doi:10.1016/j.jpsychires.2008.10.001

25.

Leckman

J. F.

Sholomskas

Thompson

W. D.

Belanger

Weissman

M. M.

(1982). Best estimate of lifetime psychiatric diagnosis: A methodological study. Archives of General Psychiatry, 39, 879-883. PMID: 7103676.

26.

Matte

Anselmi

Salum

G. A.

Kieling

Goncalves

Menezes

. . . Rohde

L. A.

(2014). ADHD in DSM-5: A field trial in a large, representative sample of 18- to 19-year-old adults. Psychological Medicine, 45, 361-373. doi:10.1017/S0033291714001470

27.

McGough

J. J.

Barkley

R. A.

(2004). Diagnostic controversies in adult attention deficit hyperactivity disorder. American Journal of Psychiatry, 161, 1948-1956. Retrieved from http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&;db=PubMed&dopt=Citation&list_uids=15514392

28.

Mick

Spencer

Faraone

S. V.

Biederman

Zhang

H. F.

(2007). Assessing the validity of the Quality of Life Enjoyment and Satisfaction Questionnaire Short Form in adults with ADHD. Journal of Attention Disorders. Retrieved from http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&;db=PubMed&dopt=Citation&list_uids=17934183

29.

Moffitt

T. E.

Houts

Asherson

Belsky

D. W.

Corcoran

D. L.

Hammerle

. . . Caspi

(2015). Is adult ADHD a childhood-onset neurodevelopmental disorder? Evidence from a four-decade longitudinal cohort study. American Journal of Psychiatry, 172, 967-977. doi:10.1176/appi.ajp.2015.14101266

30.

Peyre

Hoertel

Cortese

Acquaviva

De Maricourt

Limosin

Delorme

. (2014). Attention-deficit/hyperactivity disorder symptom expression: A comparison of individual age at onset using item response theory. Journal of Clinical Psychiatry, 75, 386-392. doi:10.4088/JCP.13m08638

31.

Reinhardt

M. C.

Benetti

Victor

M. M.

Grevet

E. H.

Belmonte-de-Abreu

Faraone

S. V.

Rohde

L. A.

(2007). Is age-at-onset criterion relevant for the response to methylphenidate in attention-deficit/hyperactivity disorder? Journal of Clinical Psychiatry, 68, 1109-1116. PMID: 17685750.

32.

Sattler

J. M.

(1988). Assessment of children (3rd ed.). San Diego, CA: Author.

33.

Surman

C. B.

Biederman

Spencer

Yorks

Miller

C. A.

Petty

C. R.

Faraone

S. V.

(2011). Deficient emotional self-regulation and adult attention deficit hyperactivity disorder: A family risk analysis. American Journal of Psychiatry, 168, 617-623. doi:10.1176/appi.ajp.2010.10081172

34.

Surman

C. B.

Hammerness

Petty

Doyle

Chu

Gebhard

. . . Biederman

(2010). Atomoxetine in the treatment of adults with subthreshold and/or late onset attention-deficit hyperactivity disorder-not otherwise specified (ADHD-NOS): A prospective open-label 6-week study. CNS Neuroscience & Therapeutics, 16, 6-12. doi:10.1111/j.1755-5949.2009.00124.x

35.

van

Glind

Konstenius

Koeter

M. W.

van Emmerik-van Oortmerssen

Carpentier

P. J.

Kaye

. . . van den Brink

(2014). Variability in the prevalence of adult ADHD in treatment seeking substance use disorder patients: Results from an international multi-center study exploring DSM-IV and DSM-5 criteria. Drug and Alcohol Dependence, 134, 158-166. doi:10.1016/j.drugalcdep.2013.09.026

36.

Vande Voort

J. L.

J. P.

Jameson

N. D.

Merikangas

K. R

. (2014). Impact of the DSM-5 attention-deficit/hyperactivity disorder age-of-onset criterion in the US adolescent population. Journal of the American Academy of Child & Adolescent Psychiatry, 53, 736-744. doi:10.1016/j.jaac.2014.03.005

37.

Wechsler

(1997). Wechsler Adult Intelligence Scale III [manual] (3rd ed.). San Antonio, TX: The Psychological Corporation.

38.

Weissman

M.M.

(1999). Social Adjustment Scale – Self Report (SAS-SR): User’s Manual. Toronto: Multi-Health Systems: