Comorbidity and sociodemographic characteristics of adult autism spectrum disorder and attention deficit hyperactivity disorder: epidemiological investigation in the World Mental Health Japan 2nd Survey

Abstract

Objectives

The present study aims to examine sociodemographic correlates and comorbid mental disorders among adult with autism spectrum disorders (ASD) and attention deficit hyperactivity disorders (ADHD) using the national representative data of Japan.

Methods

Analysis was conducted on the cross-sectional data derived from the World Mental Health Japan Survey 2, whose participants were community residents aged 20–75 years old. Multiple logistic regression was conducted on the data of those who were assessed for adult ASD (N = 2227) and ADHD (N = 2297). The association of adult ASD/ADHD with sociodemographics, mood disorders, anxiety disorders, substance use disorders, suicide ideation, hikikomori (social withdrawal), and internet addiction was examined, adjusting for sex and age. Gender difference in the association was also assessed based on the interaction terms of sex.

Results

None of the sociodemographics characteristics were significantly associated with adult ASD/ADHD. ADHD had increased risks for a variety of mental disorders, having the highest odds ratios for social phobia. However, a significant positive association was limited to social phobia and to drug abuse and dependence among those with adult ASD. Hikikomori and internet addiction were positively associated with ASD/ADHD, while suicide ideation was only associated with ADHD. Women with ADHD were more likely to have any one of mental disorders, as well as alcohol abuse and dependence, compared to men with ADHD.

Conclusions

Adults with ASD/ADHD in Japan had increased mental health needs. The specific needs of women with ADHD should be also considered in future clinical work and research.

Introduction

Autism spectrum disorders (ASD) and attention deficit hyperactivity disorders (ADHD) are variations of neurodevelopmental conditions, both of which are characterized by persistent cognitive and behavioral patterns that interfere with social interactions. ASD is defined by restricted and repetitive patterns of behavior, interests, or activities, and ADHD by inattention, hyperactivity, and impulsivity (American Psychiatric Association 2013). Emerging in early childhood, these traits could persist in adulthood and evoke interpersonal conflicts and impairments in one’s psychosocial functioning, such as increased risk for mental disorders (Barbaresi et al. 2013, Fayyad et al. 2007, Kessler et al. 2006, Mannuzza et al. 1993), lowered educational attainment (Brugha et al, Fayyad et al. 2007, Fredriksen et al. 2014, Shattuck et al. 2012), and work impairment (Brook et al. 2013, Fredriksen et al. 2014, Kessler et al. 2006, Kirino et al. 2015, Küpper et al. 2012, Shattuck et al. 2012). Despite the recent increase in clinical and societal attention on adult ASD/ADHD, mental health needs of adult ASD/ADHD are yet to be identified (Murphy et al. 2016, Ramsay et al. 2007) due to the lack of studies conducted in adult representative population, which are limited to some studies (Brugha et al. 2011, Fayyad et al. 2007, Kessler et al. 2006).

Previous studies consistently indicated deteriorated mental health among adults with ASD/ADHD. Mood disorders, anxiety disorders, and substance use disorders were more prevalent among adults with ASD (Croen et al. 2015, Maddox et al. 2015) and ADHD (Fayyad et al. 2007, Kessler et al. 2006) compared to those without these disorders. The investigation of the existing study was limited to common mental disorders, and other important mental health issues might have been overlooked. For example, the association of adult ASD/ADHD with social withdrawal has not been examined, even though the difficulties in social interaction (Fayyad et al. 2007, Taylor et al. 2015) might result in their avoiding social contacts. Recent studies found the increased risk of ADHD for internet addiction in youth (Ho et al. 2014, Ko et al. 2012). However, whether ADHD in adulthood is also associated with such addiction has not been examined yet, and no investigation was attempted for adult ASD. Suicide among adult ASD/ADHD is another area of research that was scarcely examined in adult general populations, although existing studies among the youth or clinical populations found an increased risk of suicide attempts among those with ASD/ADHD (Croen et al. 2015).

These studies indicate that adults with ASD/ADHD may have lower socioeconomic status because of the disturbance of social functioning arising from their persistent symptoms and deteriorated mental health. A few studies examined sociodemographic correlates of adult ASD (Taylor et al. 2015) and adult ADHD (Fayyad et al. 2007, Kessler et al. 2006) and found that they were less likely to be in marital relationship, to have higher educational attainment, or to stay in employment. However, sociodemographic correlates may vary across countries, because successful participation of adults with ASD/ADHD is largely determined by the sociocultural contexts they are in (Hodgkins et al. 2011, Norbury et al. 2013). For example, Japanese culture is highly contextual with complex social hierarchies and prescribed social relations, thus communication and social interaction in such a society could pose greater difficulties on adults with ASD/ADHD (Norbury et al. 2013). More research is needed across diverse sociocultural settings to identify the risk factors, embedded in the societies, for the promotion of their psychosocial well-being.

This study aims at exploratorily examining the association of adult ASD/ADHD with a wider range of mental disorders and related problems, including social withdrawal, internet addiction, and suicide ideation using the national representative data of Japan. We also examined their sociodemographic characteristics to investigate whether adults with these disorders tend to be in disadvantageous socioeconomic positions. Considering previous reports on gender variation in the impact of these disorders (Dalsgaard et al. 2002, Kreiser et al. 2015), we further examined gender difference in these associations.

Methods

Sample

Data for the present study were derived from the World Mental Health Japan Survey 2nd (WMHJ2), which was a cross-sectional survey conducted in collaboration with World Health Organization World Mental Health Survey (Kessler et al. 2008). The data were collected from 2013 to 2015 using a two-stage stratified random sampling method with the survey area as the primary sampling and survey participants as the secondary sampling unit. Survey sites were randomly selected from each of the three blocks that covers the whole area of Japan and participants from resident registries. As a result, the final sample was 2450 that represent national population of Japan (response rate 43.4%). We analyzed the data as of 8th of November 2017.

Survey participants were community residents aged 20–75 years old. Respondents were excluded from participation if they were (1) deceased or institutionalized, (2) had moved from the survey site, or (3) could not speak Japanese. For the present study, we used the data of those who were assessed for adult developmental disorders: N = 2227 for ASD, and N = 2297 for ADHD.

Survey instrument

Data were collected by face-to-face interviews using the computer-assisted personal interviews (CAPI) and self-administered questionnaire. The questionnaire built in CAPI was translated into Japanese from the World Mental Health (WMH) Survey Initiative version of the World Health Organization Composite International Diagnostic Interview (WMH-CIDI), a fully structured diagnostic interview to be conducted by trained interviewers (Kessler et al. 2004). CIDI demonstrated acceptable reliability and validity as a clinical assessment tool of common mental disorders (Haro et al. 2006). A self-administered questionnaire was developed for use in the WMHJ2. It additionally collected data on the factors possibly related to the mental health of the Japanese community residents, such as developmental disorders and internet addiction.

Measures

Autism spectrum disorders (ASD)

ASD was measured by the Japanese version of Autism-Spectrum Quotient Short Form (AQ-J-10) (Kurita, Koyama, & Osada, , & Osada, 2005). AQ-J-10 is a self-administered questionnaire of 10 items, which screens for adolescents and adults with current high-functioning pervasive developmental disorders. The AQ-J-10 asks respondents the degree of agreement on the respondents’ cognitive and behavioral patterns in the areas of social skill, attention switching, communication, and imagination. The response options were 0 for ‘disagree’ or ‘slightly disagree’ and 1 for ‘agree’ and ‘slightly agree’ (range 0–10). A total score of 7 or higher was defined as ‘having ASD’ (Kurita et al. 2005). The sensitivity and specificity of this cut-off point were 0.75 and 0.90, respectively (Kurita et al. 2005).

Attention deficit hyper activities disorders (ADHD)

ADHD was measured by the Japanese version of World Health Organization (WHO) adult ADHD self-report scale version 1.1 (ASRS-v1.1) (Adler et al. 2006, Kessler et al. 2005, Kessler et al. 2007). The ASRS-v1.1 was developed in conjunction with the WHO Composite International Diagnostic Interview, aiming at screening for current ADHD in adults who are 18 years or older. The ASRS-v1.1 consists of six items that ask about the frequency of the respondents’ inattention and hyperactivity-impulsivity using the five-point Likert scales: never, rarely, sometimes, often, and very frequently. ADHD was positive when a frequency of more than four items was higher than the standard frequency (Kessler et al. 2005). The sensitivity and specificity of this cut-off point were 0.64–0.69 and 0.99–1.00 (Kessler et al. 2007).

Common mental disorders

Respondents’ lifetime experience of common mental disorders was assessed based on DSM-IV adopted in the WMH- CIDI (Kessler et al. 2004). In the present study, the following nine disorders were considered: major depressive disorders, dysthymia, and bipolar I & II disorders as a group of mood disorders, panic disorders, social phobia, agoraphobia, and general anxiety disorders as a group of anxiety disorders, alcohol abuse or dependence, and drug abuse or dependence.

Other mental health problems

Suicide ideation. Respondents were asked whether they had ever seriously thought about committing suicide. A positive answer to this question was defined as having suicide ideation in their lifetime.

Hikikomori (Social withdrawal). Social withdrawal was assessed by using the data from the hikikomori section in the Japanese version of WMH-CIDI. Hikikomori, in Japanese, is a form of social withdrawal among those who retreat from social interaction for protracted periods of time. The definition of hikikomori in this study was based on the Japanese government guidelines; a state of social withdrawal for more than 6 months, rarely communicating with other than their family members and confining themselves in their house without going to school or work. The same types of withdrawal state caused by physical illness or injury were excluded from the hikikomori cases. Respondents who were 64 years old or younger were assessed for the lifetime experience of hikikomori.

Internet addiction. Internet addiction was assessed by the Japanese version of the compulsive Internet Use Scale (CIUS) (Meerkerk et al. 2009, Yong et al. 2017). CIUS measures excessive absorption, difficulties in setting priorities, and mood regulation with 14 items based on the five-point Likert scale (0 = never to 4 = very often; range = 0–56). We used a total score of 19 or higher, which indicated longer hours spent online and deteriorated mental health, as a cutoff point for having internet addiction (Yong et al. 2017).

Sociodemographics

Sociodemographics examined in this study were sex, age, marriage, education, and individual income. Age was grouped into five groups: 20–29, 30–39, 40–49, 50–59, and 60–75 years old. Marriage was categorized into ‘married,’ ‘divorced/separated,’ and ‘never-married.’ Education was the highest educational level attained and categorized into three groups: ‘Jr. high school/high school,’ ‘two-year college,’ and ‘university.’ Employment status was categorized into four groups: ‘employed,’ ‘student or homemakers,’ ‘retired,’ and ‘not-employed or others.’ Individual income was annual income before tax deduction earned by respondents, which was divided into tertile: low, middle, and high.

Analysis

In order to assess the association between the sociodemographic characteristics and ASD/ADHD, the prevalence of ASD/ADHD in each sociodemographic characteristic was calculated in cross-tabulations and evaluated for significance by chi-squared tests. All the sociodemographic variables were entered into one multiple logistic regression model to examine multivariate association between sociodemographics and ASD/ADHD.

The association of ASD/ADHD with common mental disorders and other mental health problems was examined by the prevalence of these disorders and problems for those with and without ASD/ADHD. Then, logistic regression was conducted to examine the bivariate association of these disorders and problems with ASD/ADHD, adjusted for age and sex. Gender difference in these associations was assessed using the interaction terms between gender and ASD/ADHD. Statistical significance was consistently tested at a 0.05 level, two-sided. All analyses were conducted using STATA (version 14, Stat Corp, College Station, TX, USA).

Ethical standards

The aim and procedure of this study were approved by the Research Ethics Committees of Graduate School of Medicine/Faculty of Medicine, The University of Tokyo (10131- (2), (3), and (4)), and therefore met the ethical standards laid down in the Declaration of Helsinki and ethical guidelines for epidemiological studies. The participation in this study was completely voluntary and confidentiality was assured. All participants gave their written informed consent prior to their participation.

Results

Prevalence of adult ASD/ADHD

The prevalence of ADS and ADHD was 5.1% and 8.2%, respectively, among WMHJ2 respondents (Table 1). These disorders were more prevalent among men than women significantly: 7.1% vs. 3.2% for ASD, and 10.2% vs. 6.4% for ADHD. The prevalence of ADHD was higher among the younger, those who never married, and those who were out of employment and university-graduates, but these tendencies were not observed at a significant level for the prevalence of ASD.

Table 1.

Prevalence of autism spectrum disorders (ASD) and attention deficit hyperactivity disorders (ADHD) by sociodemographic characteristics among WMHJ2 respondents.

		Respondents assessed for ASD (N = 2227)			Respondents assessed for ADHD (N = 2297)
		Prevalence of ASD (%)	χ2		Prevalence of ADHD (%)	χ2
Total		5.1			8.2
Sex
	Men	7.1	17.9***		10.2	10.9***
	Women	3.2	DF = 1		6.4	DF = 1
Age
	20–29 yr.	8.2	7.7		23.1	100.9***
	30–39 yr.	5.5	DF = 4		10.3	DF = 4
	40–49 yr.	5.4			8.5
	50–59 yr.	3.8			5.8
	60–75yr.	4.3			3.5
Marriage
	Married	4.7	8.0		6.7	38.2***
	Separated	2.9	DF = 2		5.5	DF = 2
	Never married	7.6			15.7
Education
	High school	4.8	1.5		6.8	10.6***
	College	4.7	DF = 2		7.5	DF = 2
	University	6.0			11.2
Employment
	Employed	4.9	7.3		8.8	10.2*
	Student/homemaker	4.2	DF = 3		5.7	DF = 3
	Retired	6.2			6.1
	Not employed/others	11.6			15.1
Individual income
	Low	5.0	2.7		6.3	5.4
	Middle	4.0	DF = 2		8.6	DF = 2
	High	5.9			9.5

ASD = Autism spectrum disorders, ADHD = Attention deficit hyperactivity disorders

^*p < 0.05, ***p < 0.001

Association of socioeconomic characteristics with adult ASD/ADHD

Table 2 shows the association between socioeconomic status and adult ASD/ADHD after the adjustment for sex and age. Those out of employment had an increased risk for ASD (OR = 2.61, 95% CI = 1.19–5.73) and ADHD (OR = 2.01, 95% CI = 1.00–4.02). However, a Wald test for the significance of variable as a whole showed that employment status did not have a significant association with ASD nor ADHD. None of the other socioeconomic characteristics were significantly associated with adult ASD/ADHD.

Table 2.

Logistic regression on the association of sociodemographics with adult developmental disorders.

		ADS						ADHD
		OR	95% CI			χ2†		OR	95% CI			χ2†
Marriage
	Married	1				1.1		1				0.2
	Separated	0.76	0.32	–	1.79			1.16	0.62	–	2.19
	Never married	1.24	0.72	–	2.11			1.02	0.65	–	1.58
Education
	High school	1				0.1		1				1.4
	College	1.08	0.65	–	1.78			0.96	0.64	–	1.44
	University	1.01	0.64	–	1.60			1.20	0.84	–	1.72
Employment
	Employed	1				7.1		1				6.3
	Student/homemaker	1.48	0.80	–	2.74			0.85	0.52		1.39
	Retired	1.46	0.69	–	3.07			1.66	0.80		3.44
	Not employed / others	2.61*	1.19	–	5.73			2.01*	1.00		4.02
Individual income
	Low	1				2.3		1				0.2
	Middle	0.68	0.40	–	1.16			1.01	0.66	–	1.55
	High	0.71	0.41	–	1.23			0.95	0.61	–	1.49

^† Wald test was conducted to test the significance of each parameter as a whole; marriage DF＝2, education DF＝2, employment DF = 3, individual income DF＝2.

^*p < 0.05

Association of major mental disorders with adult development disorders

Table 3 shows the results of the logistic regression on the association of major mental disorders with ASD. Those with ASD were more likely to have anxiety disorders (OR = 2.31, 95% CI = 1.07–5.00), especially social phobia (OR = 2.78, 95% CI = 1.04–7.39), and drug abuse and dependence (OR = 5.23, 95% CI = 1.04–25.53). Table 4 shows the same associations with ADHD. Those with ADHD had an increased risk for mood disorders (OR = 1.89, 95% CI = 1.11–3.19) and anxiety disorders (OR = 5.10, 95% CI = 3.03–8.60). Among these disorders, anxiety disorders, such as panic disorders (OR = 4.89, 95% CI = 1.40–17.11), social phobia (OR = 5.59, 95% CI = 2.81–11.12), agoraphobia (OR = 5.49, 95% CI = 1.53–19.75), and general anxiety disorders (OR = 5.17, 95% CI = 2.37–11.27) had a stronger association with ADHD, compared to major depressive disorders (OR = 1.89, 95% CI = 1.11–3.19).

Table 3.

Lifetime prevalence of common mental disorders among WMHJ2 respondents with and without ASD （N = 2227）.

	Prevalence (%)		OR†	95% CI
	With ASD (n = 113)	Without ASD (n = 2114)
Any mental disorders	26.6	22.1	1.10	0.71	–	1.70
Any mood disorders	7.1	6.1	1.26	0.60	–	2.66
Major depressive disorders	6.2	5.8	1.14	0.52	–	2.53
Dysthymia	0.9	0.9	1.04	0.14	–	7.92
Bipolar I & II disorders	0.0	0.2	NA
Any anxiety disorders	7.1	3.6	2.31*	1.07	–	5.00
Panic disorders	0.9	0.6	1.74	0.22	–	13.89
Social phobia	4.4	1.7	2.78**	1.04	–	7.39
Agoraphobia	0.0	0.5	NA
General anxiety disorders	2.7	1.5	2.17	0.64	–	7.36
Substance use disorders	16.8	15.6	0.86	0.51	–	1.45
Alcohol abuse or dependence	16.8	15.4	0.87	0.52	1	1.47
Drug abuse or dependence	1.8	0.3	5.25*	1.04	–	26.53

^† Odds ratio of those with ASD to those without ASD, adjusted for sex and age (ref = without ASD).

^*p < 0.05, **p < 0.01

Table 5.

Prevalence of suicide ideation, hikikomori, and internet addiction among those with adult developmental disorders

	Prevalence (%)
ASD†	With ASD	Without ASD	OR§	95% CI
Suicide ideation (n = 137)	12.4	7.5	1.65	0.89	–	3.04
Hikikomori (n = 28)	7.1	1.9	3.24*	1.28	–	8.21
Internet addiction (n = 310)	26.4	11.8	2.60***	1.57	–	4.31
ADHD ǂ	With ADHD	Without ADHD	OR §	95% CI
Suicide ideation (n = 142)	16.0	7.0	2.44***	1.55	–	3.82
Hikikomori (n = 30)	9.7	1.5	6.14***	3.06	–	12.35
Internet addiction (n = 315)	31.4	10.7	2.35***	1.61	–	3.43

^† Analysis was conducted on those who responded to both of the sections that assessed ASD and suicide ideation (N = 2087), hikikomori (N = 1643), and internet addiction (N = 2187).

^ǂ Analysis was conducted on those who responded to both of the sections that assessed ADHD and suicide ideation (N = 2155), hikikomori (N = 1679), and internet addiction (N = 2254).

^§ Odds ratio of those with developmental disorders to those without developmental disorders, adjusted for sex and age.

^**p < 0.01, ***p < 0.00

Association of suicidal ideation, hikikomori, and internet addiction with adult developmental disorders

The associations of hikikomori, internet addiction, and suicidal ideation with adult developmental disorders were presented in Table 5. Hikikomori had a significant association both with ASD (OR = 3.24, 95% CI = 1.28–8.21) and ADHD (OR = 6.14, 95% CI = 3.06–12.35), and the odds ratios were the highest among the three types of mental health problems. Internet addiction was also associated with ASD (OR = 2.60, 95% CI = 1.57–4.31) and ADHD (OR = 2.35, 95% CI = 1.61–3.43), but suicide ideation was associated only with ADHD (OR = 2.44, 95% CI = 1.55–3.82).

Gender difference

We examined gender difference in these associations, and found that women had increased risk for mental disorders as a whole and alcohol abuse and dependence at a significant level. The odds ratios of interaction terms (reference = men) show that women with ADHD were 1.98 times more likely to have mental disorders (95% CI = 1.03–3.81), and 2.79 times to have drug abuse and dependency (95% CI = 1.28–6.09), compared to their male counterparts. The other associations did not differ significantly by sex.

Discussion

In the present study, the increased risk for mental disorders was more evident in adult ADHD than in adult ASD, with 35% of those with ADHD having at least one of the common mental disorders in their lifetime compared to 27% of those with ASD. A variety of mood disorders and anxiety disorders, as well as suicide ideation, were significantly associated with adult ADHD, while significant association was limited to social phobia among those with adult ASD. The odds ratios of adult ADHD were the higher for anxiety disorders than for mood disorders, and the highest was 5.59 (95% CI = 2.81–11.12) for social phobia. The strongest association of anxiety disorders with adult ADHD, which was not evident in the previous studies (Fayyad et al. 2007, Kessler et al. 2006), was noteworthy in this Japanese study. Although the reason for this strong association is scarcely documented, existing studies suggested that persistent attentional impairments, stress intolerance and fear, and emotional dysregulation were related to the increased risk for anxiety disorders among adult ADHD (Sobanski 2006). It could be possible that Japanese culture that values permissiveness and self-regulation may be less tolerant of such adult ADHD’s attentional and behavioral patterns (Norbury et al. 2013).

On the other hand, drug abuse and dependence were significantly associated with adult ASD, but none of substance disorders were associated with adult ADHD. A qualitative study on life-experience of adult ASD by Kronenberg et al. (2014) indicated that those with ASD tend to experience confusion of thoughts and emotions, and have feelings of loneliness and boredom, and that they abused substance to ease these confusions and uncomfortable feelings. Further study is needed to appraise this speculation. Lack of association between substance disorders and ADHD needs further investigation, considering the existing evidence of their having higher risk for substance disorders (Fayyad et al. 2007, Kessler et al. 2006).

Hikikomori (social withdrawal) and internet addiction were also associated with adult ASD/ADHD. The comorbidity of hikikomori may be explained by the greater likelihood of anxiety disorders among them, because severe social anxiety disorders in early life was found to be a proxy of hikikomori in adulthood (Nagata et al. 2013). The increased risk for internet addiction among ADHD in children and adolescences was well examined, but no study examined the risk among ASD. Immediate and repeated stimulation and reward, as well as better performance in the controllable virtual reality, may have attracted them into internet addiction (Ko et al. 2012).

None of the sociodemographic characteristics was significantly associated with adult ASD/ADHD after adjusting for sex and age. Although the likelihood of ‘not-employed and others’ compared to ‘employed’ was significantly greater among those with ASD (OR = 2.61, 95% CI = 1.19–5.73) and ADHD (OR = 2.01, 95% CI = 1.00–4.02), a Wald test for ‘employment status’ was not significant. These results were inconsistent with previous studies that found greater likelihood of disadvantageous sociodemographic conditions, such as being a single and lowered educational attainment, among adults with ASD/ADHD (Brugha et al. 2011, Fayyad et al. 2007, Kessler et al. 2006, Taylor et al. 2015). In Japan, it could assume that social disadvantages of adults with ASD/ADHD might be limited to a less extent, although their deteriorated psychosocial functioning, including hikikomori and internet addiction, could most likely have brought difficulties in their social life. Considering the possibility that our study included non-clinical ASD/ADHD cases, further examination of sociodemographic correlates is required among those with severer adult ASD/ADHD.

A few studies examined gender difference in comorbid mental health disorders and found that girls had a greater risk for mood disorders (Kreiser et al. 2015) and psychiatric admission (Dalsgaard et al. 2002) than boys. Our results showed that women with ADHD had a greater risk for having mental disorders, especially alcohol abuse and dependence, compared to their male counterparts. As the proportion of ADHD was relatively low among women, their needs arising from gender-specific symptoms and social roles (Nussbaum 2012) could have been neglected, which may have resulted in an increased risk for mental disorders among women with ADHD.

Our results need to be understood in the light of the following limitations. First, our assessment of adult ASD/ADHD was not diagnostic, and the assessment threshold may be lowered due to the use of short screening scales. Indeed, the prevalence of adult ASD/ADHD in this study was higher than that of previous studies based on clinical interviews (Brugha et al. 2011, Delobel-Ayoub et al. 2015, Kessler et al. 2006). Moreover, the self-administered questionnaire, including the assessment of ASD/ADHD, might have limited the participation of those with low intelligence, which could have eventually excluded adults with severe ASD/ADHD from the present study. Thus, the lack of associations can be partly attributable to the moderate severity of ASD/ADHD in our study, while it could be also due to the sociocultural variations in which cognitive and behavioral patterns of these disorders can be less dysfunctional or problematic (Hodgkins et al. 2011). Second, our investigation was cross-sectional and does not guarantee a causal relationship. In addition, the positive association between adult ASD/ADHD could be found due to the cross-section design, as a review conducted by Marks et al. (2001) found that a significant association with comorbid mental disorders was found in retrospective studies, but not in prospective studies. However, this limitation does not bother the interpretation of our results severely in a sense that the aim of this study was to address the current mental health needs of adult ASD/ADHD.

Despite these limitations, our study highlighted the increased mental health needs of adult ASD/ADHD in Japan. A unique contribution of this study was that it found an increased risk for hikikomori and internet addiction among adult ASD/ADHD based on the national representative community sample. Assessment of these disorders in ASD/ADHD, as well as ASD/ADHD in these disorders, will be critical to prevent the interplay between ASD/ADHD and other mental health disorders from exacerbating their mental health and psychosocial functions (Sobanski 2006). To meet their diverse mental health needs, the combination of multimodal professional support, such as medications, psychotherapy, behavioral intervention, and coaching, need to be tailored for adults with ASD/ADHD (Murphy et al. 2016, Ramsay et al. 2007). Effective transition from child to adult health service to enhancement of life-long functioning and adaptive skills are also essential to increase their psychosocial well-being, although scientific investigation is lacking in this area of research (Murphy et al. 2016, Ramsay et al. 2007, Swift et al. 2014, Wilens et al. 2002). Gender difference in the comorbidity and women-specific needs should be also considered in future clinical work and research. We addressed the need for understanding the social disadvantages and resulting health problems among adult ASD/ADHD. Further investigation is required on the subgroup of people having more severe ASD/ADHD symptoms. The co-occurrence of ASD and ADHD also needs to be addressed in future studies, as 30.8% of those with ASD and 19.4% of those with ADHD in this study had concurrently had the other condition. Increased severity of symptoms and higher risks for behavioral problems among those who have both ASD and ADHD may result in greater psychosocial dysfunctions than those who have only one condition (Autism spectrum Australia, 2017). Interplay between ASD and ADHD requires further clinical and research attention.

Funding Statement

This study was funded by Ministry of Health, Labour, and Welfare and Japan Agency for Medical Research and Development.

Disclosure statement

None of the authors have conflicts of interest to be disclosed.