Sickness Absence Recommendation Among Outpatients With ADHD and Comorbidity: A Latent Class Analysis

Abstract

Objective: Determine sickness absence recommendation (SAR) prevalence for treatment-seeking patients with ADHD depending on comorbidity. Method: Population-based patient chart review of ADHD outpatients aged 19 to 29 years. The outcome, SAR, included both available financial benefit forms; activity compensation and sickness benefit. Latent class analyses (LCA) of demography and psychiatric comorbidities were performed both with and without SAR as an outcome variable. Results: Overall, 38% received SAR. Latent classes ranged from a from a small class of only females with personality disorders where all had SAR to larger groups characterized by lower comorbidity where 15% to 29% had SAR. In between these extremes were other classes of (a) neurodevelopmental disorders, for example, autism diagnosis or intellectual disabilities, and (b) high rate of anxiety disorder, where SAR rates ranged 46% to 65%. Conclusion: Treatment-seeking ADHD patients can be categorized into clinically relevant subgroups providing opportunities to structure rehabilitation efforts to the individuals’ needs.

Keywords

ADHD sickness absence comorbidity

Introduction

ADHD is a common neurodevelopmental disorder with pervasive symptoms of hyperactivity, inattentiveness, and impulsivity with prevalence rates reported between 2% and 5% in adults (Kessler et al., 2006; Polanczyk, de Lima, Horta, Biederman & Rohde, 2007; Simon, Czobor, Bálint, Mészáros, & Bitter, 2009). Adults with ADHD experience problems with working memory, planning, and anticipation. Furthermore, deficits in self-regulation of emotional arousal, verbal fluency, effort allocation, and application of organizational strategies are common (Gupta, Kar, & Srinivasan, 2011; Young, Morris, Toone, & Tyson, 2006; Young, Toone, & Tyson, 2003). Those deficits are closely linked with the ability to function at work and are associated with increased sickness absence (Adamou et al., 2013; Kessler, Lane, Stang, & Van Brunt, 2009; Kleinman, Durkin, Melkonian, & Markosyan, 2009; Søgaard & Bech, 2009). In Sweden, psychiatric outpatient departments are responsible for the diagnosis, treatment, and follow-up of patients with ADHD, including rehabilitation focusing on dysfunction in everyday life, for example, work.

Sickness absence is a common and widely used treatment, intended to give patients time to recover without losing their financial security when ill; however, health care systems and financial support for young people differ widely between countries (Helgesson, Johansson, Nordqvist, Lundberg, & Vingård, 2015). In Sweden, as well as most countries, a medical diagnosis of a disorder that causes reduced work ability is required to qualify for sickness absence benefits with psychiatric disorders being among the most common and has generally risen in recent years (Arvidsson et al., 2016). Most studies of young people receiving sickness absence benefits either use data from large registers or focus on small populations in specific work places, however, there are no studies of the rate of sickness absence among young outpatients with ADHD who receive specialist psychiatric outpatient care in Sweden.

In latent class analysis, a hypothesis-free algorithm predicts membership of underlying (latent) classes based on a set of predictor variables conditioning on an outcome. This method allows identification of patterns and commonalities of sets of predictor variables within a study population. The resulting classes may therefore contain patterns of comorbidities and diagnoses which would not be revealed in standard regression analysis.

The aim of the present study is to analyze the distribution of sickness absence and work ability among young psychiatric outpatients with ADHD diagnosis using latent class analysis, thereby identifying clinically relevant categories of comorbidities affecting work ability.

Method

Study Design and Population

This is a total population based cross-sectional patient chart review study of people aged 19 to 29 years in the eastern parts of Gothenburg, Sweden. The study group comprises all patients aged 19 to 29 years who received outpatient care at two outpatients’ departments in Gothenburg. Those two outpatient departments serve patients who live in the area of four large municipalities, in and nearby Gothenburg which is situated on the west coast of Sweden. Gothenburg city has a population of approximately 550,000 inhabitants in the urban area and 998,000 inhabitants in the metropolitan area and is the second-largest city in Sweden and the fifth-largest in the Nordic countries (Statistics Sweden, 2018). The areas’ population has increased successively during the last years.

Activity compensation as a benefit was introduced in Sweden in 2003. A person has to be between 19 and up to 30 years old to get this financial help. There are two different kinds of activity compensation, one issued as a sickness benefit for work ability and one as a support during longer time of completion of primary and high school studies. Activity compensation as a benefit is similar to disability pension except that it applies only to young people. It can only be issued for a maximum of three years at a time and it ends when an individual gets to the age of 30 years. Sickness benefit is another type of financial help for the young people with reduced working ability. Patients must have a reduced working ability of at least 25% to receive this kind of benefit. Sickness benefit can be issued from two different authorities, Social Insurance and Social services. These two forms of financial benefit, jointly known as sickness absence, were considered the main outcome of interest in the present study.

The study period was from March 1, 2014 to February 28, 2015. The outpatient departments in the study specialize in the treatment of patients with unipolar depression, anxiety-, personality-, and neurodevelopmental disorders. A search in the clinics database was performed where individuals with a diagnosis of ADHD who had an ongoing treatment at the clinic were considered eligible for the study. Patients with a diagnosis of schizophrenia, bipolar disorder type 1 and type 2 or substance use disorder were excluded from the study population for the following reasons: (a) the diagnostic process of ADHD with these comorbidities is more complicated and may give a selection bias and (b) the rate of sickness absence is strongly influenced by these diagnoses.

Measures

Data were gathered from electronic patient charts including age, sex, and psychiatric diagnoses (both primary and comorbid). This study focused only on the group with patients with ADHD either as primary or comorbid diagnosis. Psychiatric diagnoses were classified according to the ICD-10 (International Statistical Classification of Diseases and Related Health Problems–10th Revision) categories: mood disorders (F30-F39), anxiety disorders (F40-F48), personality disorders (F60-69), intellectual disability (F70-F79), and autism spectrum disorders (F80-F89), the last two collectively referred to as neurodevelopmental disorders.

Sickness absence is recommended by a medical doctor at the outpatient clinic and a medical certificate is issued. However, this is not a guarantee that the patient receives financial benefits, as the responsible authority, National Board of Health and Welfare, decides based on a doctors’ medical certificate as well as other information regarding work ability The current study focuses on doctors’ recommendation and the variables sickness benefit and activity compensation refer to the doctor’s assessment of recommending that the patients is not fit to work, study, look for a job, or have parental leave. Therefore, a patient’s current occupation was registered as either working/studying (including job-seeking or on parental leave) or being recommended for sickness benefit or activity compensation, grouped under the term sickness absence recommendation (SAR).

Analysis

The data were analyzed using polytomous latent class analysis. Using this method, the program uses predictor variables, in this case participant characteristics (age group, sex, and diagnoses other than ADHD), to distribute the participants into latent classes. A criterion variable (i.e., the outcome variable SAR) can be included in the model. The optimal number of classes in each model was decided by inspecting Akaike information criterion (AIC) and Bayesian information criterion (BIC). The poLCA simulation was set to run models with one to eight classes for maximum 5,000 iterations in 10 repetitions to maximize the global log-likelihood function. First, using AIC and BIC to compare the fit across numbers of classes in different trials to minimize the effect of the starting values on the results, for example, lower AIC and BIC scores indicate better fit for a model. The analyses were performed with and without SAR as a criterion variable in the model. In the results, frequencies of predictor and covariate characteristics in the resulting latent classes were described. Data were analyzed with R (R Core Development Team) using the “poLCA” package (Linzer & Lewis, 2014).

The study protocol was reviewed and approved by the central ethical review board of Sweden February 2, 2015 (Number: Dnr Ö 43-2014).

Results

A total of 1,031 individuals aged 19 to 29 years had contact with the two outpatient clinics, of those 516 had established treatment contact at the clinic with an ADHD diagnosis, and fulfilled the inclusion criteria, including complete information on the variables of interest. The 516 participants had a mean age of 23.9 years, and little over half were male (Table 1). The most common comorbid diagnosis was mood disorder, present in 41.1% of the cases, followed by anxiety and autism spectrum disorder (ASD) with 34.3% and 22.1%, respectively (Table 1). Fewer had personality syndromes and intellectual disabilities, 5.8% and 7.6%, respectively. More than half did not receive SAR whereas 4.5% were recommended sickness benefit and a third activity compensation. Of the 319 individuals without SAR, 147 were working, 74 were studying, and 49 were seeking work; however, there was no information on the occupational status of the remaining 28 individuals. A small subset of patients had no diagnosed comorbidity (15.9%) out of which 85.4% did not have SAR.

Table 1.

Study Participant Demographic and Diagnostic Categories.

Variable	Total, n = 516
Age in years	23.9 ± 0.85
Male sex	262 (50.8)
Comorbid diagnoses
Anxiety disorders	177 (34.3)
Personality disorder	30 (5.8)
Intellectual disability	39 (7.6)
Autism spectrum disorder	114 (22.1)
Mood disorders	212 (41.1)
No comorbidity	82 (15.9)
Sickness absence	197 (38.2)
Sickness benefit	23 (4.5)
Activity compensation	174 (33.7)

Note. Numbers presented as n (%) except age as mean ± standard deviation.

The first poLCA (without SAR as an outcome variable) had a best fit for three classes (Figure 1a) and the second model (with SAR), BIC had the best fit at three classes and AIC at four (Figure 1b). As BIC was only marginally higher for four classes than three, four classes were considered the best fit.

Figure 1.

BIC and AIC for models with different numbers of latent classes, left (model 1) without sickness absence recommendation (SAR) as an outcome and right with SAR (model 2).

The three classes predicted by Model 1 in which SAR was not a criterion variable comprised 12% to 54% of the participants (Table 2, left side). In the smallest class (12%), neurodevelopmental disorders were prominent (37% were diagnosed with intellectual disability) and 88% were diagnosed with ASD and 82% were male, mean age was 23 years and SAR was common (58%). In the medium-sized (34%) class, all had anxiety disorders, 47% were males and mean age was 24 years, less than half received SAR (46%). In the largest class (54%), 47% were males, comorbidities other than mood disorders (62%) were uncommon (29% had no comorbidities at all) and 29% received SAR.

Table 2.

Latent Classes by Age, Sex, and Diagnostic Categories by SAR Status Among Outpatients With ADHD in Gothenburg (n = 516).

Variable	Model 1: latent class analysis not including SAR			Model 2: latent class analysis including SAR
Variable	Neurodevelopmental disorder	Anxiety	Low comorbid	Personality disorder	Neurodevelopmental disorder	Anxiety	Low comorbidity
Proportion	60 (12)	177 (34)	279 (54)	18 (3)	99 (21)	177 (34)	222 (43)
Age in years	22.9 ± 3.1	24.0 ± 2.9	24.1 ± 2.8	26 ± 2.4	23 ± 2.6	24 ± 2.9	24 ± 2.8
Male sex	49 (82)	83 (47)	130 (47)	0	71 (72)	83 (47)	108 (49)
Comorbid diagnoses
Anxiety disorders	0	177 (100)	0	0	0	177 (100)	0
Personality disorder	0	2 (1)	28 (10)	18 (100)	2 (2)	2 (1)	8 (4)
Intellectual disability	22 (37)	6 (3)	11 (4)	3 (17)	23 (23)	6 (3)	7 (3)
Autism spectrum disorder	53 (88)	25 (14)	36 (13)	3 (17)	86 (87)	25 (14)	0
Mood disorders	0	40 (23)	172 (62)	5 (28)	25 (25)	40 (23)	142 (64)
No comorbidity	0	0	82 (29)	0 (0)	5 (5)	0	77 (35)
SAR	35 (58)	82 (46)	80 (29)	18 (100)	64 (65)	82 (46)	33 (15)
Sickness benefit	1 (3)	11 (13)	11 (14)	2 (11)	2 (3)	11 (13)	8 (24)
Activity compensation	34 (97)	71 (87)	69 (86)	16 (89)	62 (97)	71 (87)	25 (76)

Note. Variables are presented as n (%) except age as mean ± standard deviation. SAR = sickness absence recommendation.

The four classes predicted by Model 2 where SAR was included as a covariate comprised between 3.4% and 43% of the participants (Table 2, right side). In the smallest class, there were 18 individuals, all with a diagnosis of personality syndrome who were women, the mean age was 26 years all had received SAR. The second-smallest class comprised 99 individuals, neurodevelopmental disorders were common, (ASD 87% and intellectual disability 23%). Most (72%) were males and mean age was 23 years where a majority received SAR (65%). The second-largest (34%) class comprised the same individuals as the second-largest in Model 1, that is, everyone had anxiety disorder, 47% were males, and mean age was 24 years, and less than half received SAR (46%). In the largest class (43%), mood disorder was the most common comorbidity (64%), 35% had no other diagnoses than ADHD, and 49% were males with 15 % receiving SAR.

When looking at the total group receiving SAR, most belonged to the anxiety class in both models (Figure 2), although the rate was proportionally higher in the personality and neurodevelopment classes (Table 2).

Figure 2.

The proportion of sickness absence recommendation (SAR) by latent class analyses (n = 197). On the left: not conditioning on SAR (model 1), and on the right: conditioned on SAR (model 2).

Discussion

The current study found that SAR was common for ADHD patients treated at the outpatient psychiatric ward. Our analysis revealed that SAR rates differed between classes identified by comorbidities, sex and, to a lesser extent, age.

In these cases of severe ADHD as either primary or secondary diagnosis, rates of SAR were generally higher than in other studies (Biederman et al., 2006) with over a third having SAR while in the general population only 2.4% receive that benefit, translating into almost 20 times higher rate ranging from five times in the low rate comorbidity class to over 300 times higher prevalence in the smallest class with personality disorders (The Swedish Social Insurance Agency, 2016).

There were several differences between the latent classes with respect to comorbid diagnoses, sex and SAR, especially depending on the model being conditioned. In Model 1, the three classes were more homogeneous with respect to SAR as well as missing a small class of women diagnosed with personality disorder where all had SAR. The importance of comorbidities is further illustrated in the difference between the anxiety class, which contained the same individuals in both models, and the larger low-comorbid class. Those two groups differ with respect to SAR rates and needs of treatment. This might suggest that a comorbidity of ADHD and anxiety disorders present a higher risk for impairment of work ability than comorbidity between ADHD and unipolar mood disorders. This is highlighted in Figure 2 where the proportion of SAR is most disproportionally high in the class with high anxiety rates.

In the two models, the class with a high rate of neurodevelopment disorders was better defined by conditioning on the SAR outcome indicating that the functional level is a part of belonging to the class as in Model 1 the group was smaller and had a lower rate of SAR. Autism spectrum syndrome is a prototypical example of male preponderance, with a male to female ratio of ~4:1 (Werling & Geschwind, 2013) but different clinical symptomatology may cause women to be underdiagnosed (Kopp & Gillberg, 2011). However, literature describing work ability or disability in patients with comorbidity of ADHD and autism, regarding differences between the sexes is scarce. These differences could be of interest in light of the results of the current study and one can speculate whether the difference in SAR rates depend only on diagnosis or whether it is also related to the patients’ sex.

The results could further be discussed from a perspective of functioning in everyday activities, including the ability to work. Bejerholm and Areberg (2014) studied factors predicting high return-to-work potential in individuals with severe mental illness and found significant associations with having fewer symptoms, rehabilitation support or participating in productive activities. This is partly in line with the findings in the larger low-comorbid classes, where a higher proportion of participants were working or studying. This could imply that either the functional impairments are not so frequent in this subgroup or that successful medication and/or a god adjustment of both work and everyday life activities has had a positive effect on the potential work ability.

Characteristics from the high rate of anxiety class, with a higher prevalence of SAR can be related to a study on such symptoms and work capacity (Bertilsson, Petersson, Ostlund, Waern, & Hensing, 2013). The results described a reduced capacity related both to task, time, context, and social interactions implying a larger need for focusing on such matters when supporting patients to return to work. Symptoms of depression have been shown to negatively influence the potential for return to work, whereas engagement in meaningful daily occupations work in the other direction, indicating that support through a constructive use of time and symptom management could be helpful (Bejerholm & Areberg, 2014).

In the classes with high rates of personality and neurodevelopmental disorders, where there was more comorbidity and an even higher proportion SAR, regardless of the two models, individuals may have more comprehensive functional deficits related not only to work ability, but to most major occupational role areas (Ek & Isaksson, 2013). It has been proposed that metacognition plays a larger role in these difficulties than behavior regulation problems. A larger focus on assessing how such impairments have an impact on occupational performance in everyday life has been proposed (Stern & Maeir, 2014). Such assessments could be a basis for a more comprehensive SAR and clarify the prognosis for the possibility to return to work. It could further be hypothesized that these two groups are in larger need for more supportive interventions both from the health care system and from community based practices.

The latent class analysis is a hypothesis-free method where a set of predictor variables are used to predict membership of underlying classes conditioning on an outcome risk, the underlying classes will then be comprised of individuals sharing predictor variable values or outcome risk. In the present study, Model 1 was not conditioned on SAR whereas Model 2 was. The classes in Model 2 give a clinically relevant categorization of patient groups and highlights the female dominant class with personality disorders and ADHD that are deemed with a high need of SAR. The variation between the different subgroups with respect to comorbidity and sickness absence indicates that impairment in ADHD is a complex phenomenon and these factors should be considered when evaluating a patient’s potential for work. No other study has presented results of SAR and ADHD with comorbid personality syndromes and only a few studies exist about work disability among persons with borderline personality syndrome, the most common subgroup. However, personality syndrome has been described as a significant source of psychiatric morbidity, accounting for more impairment in functioning than major depressive disorder alone by Skodol et al. (2002) which is in line with our results in which personality syndrome and ADHD was associated with a devastating functional impairment which should be studied further.

In our study of patients receiving specialized ADHD treatment, workforce participation was much lower than in the general population. However, it is a common hypothesis in studies or workforce participation in individuals with ADHD diagnoses that ADHD treatment can improve patient’s work ability (Biederman, 2004). As all study participants were receiving ADHD treatment and the study is cross-sectional, we cannot conclude on effects of treatment, but a likely speculation is that treatment outcomes will differ between classes with different comorbidities and longitudinal studies should be performed to evaluate any effects of ADHD treatment on workforce participation in patients with different comorbidities. Previous literature report a trend of higher risk for disability pension after having received activity compensation with more than two out of three receiving activity compensation eventually receiving disability pension at age of 30 years (The Swedish Social Insurance Agency, 2017). This gives the psychiatric outpatient health care a lot of responsibility and requires development of more qualified and effective treatments which focus on early return to work and not only to remission of the main symptoms of the diagnosis. In addition, there is a strong correlation between long term sickness absence and all-cause mortality including suicide (Bryngelson, Asberg, Nygren, Jensen, & Mittendorfer-Rutz, 2013). The natural course and consequences of disease and injury have only been studied for some diagnoses (Vingård, Alexanderson, & Norlund, 2004) and the literature is relatively poor regarding the most prominent sick-leave diagnoses, for example, psychiatric disorders. It is, therefore, important to gain deeper understanding of the different factors that may influence work ability among younger patients with ADHD.

Recent studies of adults with ADHD agree that while this is an impairing condition among people of working age it was associated with a minor, 4% to 5%, reduction in work performance (Barkley & Murphy, 2010; Biederman et al., 2006); however, adult ADHD has been associated with >120 million lost workdays in the United States each year (Kessler et al., 2005) and overall work performance evaluations are lower in employees with ADHD than in those without ADHD (Biederman et al., 2006). Our results reveal a much higher risk of being outside the workforce which varied across the latent classes, indicating that patients attending specialized care have the highest risk for SAR, which may also be the main reason for seeking care. This can be a cause for concern as in many countries the rate of sickness absence has risen, especially due to mental disorders (Kaltenbrunner Bernitz, Grees, Jakobsson Randers, Gerner, & Bergendorff, 2013).

A number of limitations of the present study should be highlighted. In Sweden, the treatment of ADHD with central stimulating medication is restricted to licensed psychiatrists and the current study is focused on patients receiving specialist outpatient treatment for ADHD. In our study, nearly all individuals within the study area receiving specialist-treatment for ADHD were included and because of the personal identification number system, there is a high degree of certainty that this study fully represents this very specific patient group giving the study a unique coverage. The psychiatric outpatient departments are responsible for patient’s treatment. However, many adults with ADHD do not receive treatment and therefore have no contact with the psychiatric outpatient department. As our study population is highly selected comprising the ADHD patients who are most in need of treatment, our results are not directly comparable to previous survey studies of work performance/participation in community samples where patient status is self-reported and comparisons are made with healthy nonpsychiatric control groups. This limits the study’s generalizability outside the study population, however, one may argue that self-reported symptoms would have lower validity if the functional level is not affected to the degree of seeking care. Another limitation of the study is the use of SAR as primary outcome rather than the Swedish National Board of Health and Welfare’s actual decision to grant sickness absence or activity compensation. However, a parallel can be drawn to register studies of medication use where prescriptions are most often used rather than actually ingested medication.

One should always bear in mind that psychiatric diagnostics are complicated and many factors can influence a diagnostic decision, especially since the implementation of the new version of the diagnostic system Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5; American Psychiatric Association, 2013), during the study period may increase risk of different disorders being diagnosed differently and could mean that syndromes are underdiagnosed or overly diagnosed as clinicians have less experience with DSM-5 but registers of diagnoses do represent the clinical use of the diagnostic codes.

This study is focused on a specific and limited group of patients purely through age, sex, and diagnoses excluding patients with bipolar disorder of type one or two as well as with schizophrenic disorders. This can entail a risk for disregarding cultural, geographical, economic, and educational or other factors which may influence patient’s work ability or disability. However, we are confident that increased knowledge of the need for sickness absence in patients with ADHD can lead to improved treatment modalities with an individualized approach.

In conclusion, treatment-seeking patients with ADHD is a heterogeneous group and subgroups with identifiable patterns have different treatment needs. It is therefore necessary to individualize treatment according to needs as well as further study treatment response in these subgroups. Such treatment plans should include both symptomatology as well as interventions related to the return to work process and the ability to perform everyday activities. Results such as ours should be taken into consideration when deciding future policies regarding ADHD patients and SAR.

Footnotes

Acknowledgements

The authors thank Peter Asplund, system administrator at the department, for helping in the extraction of data from the electronic patient charts and Katarina Harring, occupational therapist, for helping in the design of the study. Steinn Steingrimssons research is funded by the Västra Götalands regional fund ALF (ALFGBG-588611).

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 authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Steinn Steingrimsson

Author Biographies

Aikatarini Trantou, MD, is a psychiatrist at Sahlgrenska University Hospital, Gothenburg. She works as physician in charge at a general psychiatric outpatient clinic. Furthermore, she is a process leader of insurance medicine at the Sahlgrenska University Hospital.

Hanne K. Carlsen, PhD, MPH, is an environmental epidemiologist and statistician at the Department of Psychiatry. She currently works at the Section of Occupational and Environmental Medicine, Sahlgrenska Academy at University of Gothenburg.

Christina Anderson, PhD, reg OT, is a senior lecturer in the Occupational Therapy Program at the University of Gothenburg. Her present studies focuses on factors related to the sick leave process.

Steinn Steingrimsson, MD, PhD, is a psychiatrist at Sahlgrenska University Hospital. He works half-time as a clinical researcher and lecturer at Sahlgrenska Academy, Univeristy of Gothenburg.

References

Adamou

Arif

Asherson

T.-C.

Bolea

Coghill

. . . Young

(2013). Occupational issues of adults with ADHD. BMC Psychiatry, 13, Article 59.

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

Arvidsson

Bill

Eek

Hultin

Jönsson

Ljung

Nikander

(2016). Sjukfrånvaronsutveckling 2016 [Sickness absence trends 2016]. Stockholm, Sweden: Försäkringskassan.

Barkley

R. A.

Murphy

K. R.

(2010). Impairment in occupational functioning and adult ADHD: The predictive utility of executive function (EF) ratings versus EF tests. Archives of Clinical Neuropsychology, 25, 157-173.

Bejerholm

Areberg

(2014). Factors related to the return to work potential in persons with severe mental illness. Scandinavian Journal of Occupational Therapy, 21, 277-286.

Bertilsson

Petersson

E. L.

Ostlund

Waern

Hensing

(2013). Capacity to work while depressed and anxious—A phenomenological study. Disability and Rehabilitation, 35, 1705-1711.

Biederman

(2004). Impact of comorbidity in adults with attention-deficit/hyperactivity disorder. Journal of Clinical Psychiatry, 65, 3-7.

Biederman

Faraone

S. V.

Spencer

T. J.

Mick

Monuteaux

M. C.

Aleardi

M. J.

(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.

Bryngelson

Asberg

Nygren

Jensen

Mittendorfer-Rutz

(2013). All-cause and cause-specific mortality after long-term sickness absence for psychiatric disorders: A prospective cohort study. PLoS ONE, 8, e67887.

10.

Isaksson

(2013). How adults with ADHD get engaged in and perform everyday activities. Scandinavian Journal of Occupational Therapy, 20, 282-291.

11.

Gupta

Kar

B. R.

Srinivasan

(2011). Cognitive-motivational deficits in ADHD: Development of a classification system. Child Neuropsychology, 17, 67-81.

12.

Helgesson

Johansson

Nordqvist

Lundberg

Vingård

(2015). Sickness absence at a young age and later sickness absence, disability pension, death, unemployment and income in native Swedes and immigrants. European Journal of Public Health, 25, 688-692.

13.

Kaltenbrunner Bernitz

Grees

Jakobsson Randers

Gerner

Bergendorff

. (2013). Young adults on disability benefits in 7 countries. Scandinavian Journal of Public Health, 41, 3-26.

14.

Kessler

R. C.

Adler

Ames

Barkley

R. A.

Birnbaum

Greenberg

. . . Ustün

T. B.

(2005). The prevalence and effects of adult attention deficit/hyperactivity disorder on work performance in a nationally representative sample of workers. Journal of Occupational and Environmental Medicine, 47, 565-572.

15.

Kessler

R. C.

Adler

Barkley

Biederman

Conners

C. K.

Demler

. . . Zaslavsky

A. M.

(2006). The prevalence and correlates of adult ADHD in the United States: Results from the National Comorbidity Survey Replication. The American Journal of Psychiatry, 163, 716-723.

16.

Kessler

R. C.

Lane

Stang

P. E.

Van Brunt

D. L.

(2009). The prevalence and workplace costs of adult attention deficit hyperactivity disorder in a large manufacturing firm. Psychological Medicine, 39, 137-147.

17.

Kleinman

N. L.

Durkin

Melkonian

Markosyan

(2009). Incremental employee health benefit costs, absence days, and turnover among employees with ADHD and among employees with children with ADHD. Journal of Occupational and Environmental Medicine, 51, 1247-1255.

18.

Kopp

Gillberg

(2011). The Autism Spectrum Screening Questionnaire (ASSQ)-Revised Extended Version (ASSQ-REV): An instrument for better capturing the autism phenotype in girls? A preliminary study involving 191 clinical cases and community controls. Research in Developmental Disabilities, 32, 2875-2888.

19.

Linzer

Lewis

(2014). poLCA: An R package for polytomous variable latent class analysis. Retrieved from http://dlinzer.github.com/poLCA

20.

Polanczyk

de Lima

M. S.

Horta

B. L.

Biederman

Rohde

L. A.

(2007). The worldwide prevalence of ADHD: A systematic review and metaregression analysis. The American Journal of Psychiatry, 164, 942-948.

21.

Simon

Czobor

Bálint

Mészáros

Bitter

(2009). Prevalence and correlates of adult attention-deficit hyperactivity disorder: Meta-analysis. The British Journal of Psychiatry, 194, 204-211.

22.

Skodol

A. E.

Gunderson

J. G.

McGlashan

T. H.

Dyck

I. R.

Stout

R. L.

Bender

D. S.

. . . Oldham

J. M.

(2002). Functional impairment in patients with schizotypal, borderline, avoidant, or obsessive-compulsive personality disorder. American Journal of Psychiatry, 159, 276-283.

23.

Søgaard

H. J.

Bech

(2009). Psychiatric disorders in long-term sickness absence—A population-based cross-sectional study. Scandinavian Journal of Public Health, 37, 682-689.

24.

Statistics Sweden. Population statistics (2018). Available from https://www.scb.se/en/finding-statistics/statistics-by-subject-area/population/population-composition/population-statistics/

25.

Statistics Sweden. (2017). Available from www.scb.se

26.

Stern

Maeir

(2014). Validating the measurement of executive functions in an occupational context for adults with attention deficit hyperactivity disorder. The American Journal of Occupational Therapy, 68, 719-728.

27.

The Swedish Social Insurance Agency. (2016). Barns relativa ålder och funktionsnedsättning [Childrens relative age and lower level of function. Report 2016:3 in Swedish]. Inspektionen för socialförsäkringen: Stockholm. Retrieved from https://www.forsakringskassan.se/wps/wcm/connect/2af3f5c6-2959-46d9-a16b-ecd2749dfdfb/socialforsakringsrapport_2016_03.pdf?MOD=AJPERES

28.

The Swedish Social Insurance Agency. (2017). Aktivitetsersättning till trettio års ålder—vad händer sedan? [Activity compensation up to 30 years of age—what happens then? Report 2017:5 in Swedish]. Inspektionen för socialförsäkringen: Stockholm. Retrieved from http://www.inspsf.se/digitalAssets/6/6128_3rapport_2017-5_ru_webb.pdf

29.

Vingård

Alexanderson

Norlund

(2004). Swedish Council on Technology Assessment in Health Care (SBU). Chapter 9. Consequences of being on sick leave. Scandinavian Journal of Public Health, 63, 207-215.

30.

Werling

D. M.

Geschwind

D. H.

(2013). Sex differences in autism spectrum disorders. Current Opinion in Neurology, 26, 146-153.

31.

Young

Morris

Toone

Tyson

(2006). Spatial working memory and strategy formation in adults diagnosed with attention deficit hyperactivity disorder. Personality and Individual Differences, 41, 653-661.

32.

Young

Toone

Tyson

(2003). Comorbidity and psychosocial profile of adults with Attention Deficit Hyperactivity Disorder. Personality and Individual Differences, 35, 743-755.