Abstract
Keywords
Background
Attention deficit/hyperactivity disorder (ADHD) is a common neurodevelopmental disorder frequently diagnosed between the ages of 7 and 10, lasting into adulthood in 40% to 60% of patients (Weibel et al., 2020). The estimated worldwide prevalence among the pediatric population varies between 3% and 7% (Mohammadi et al., 2021). After changes in diagnostic criteria (from the Diagnostic and Statistical Manual of Mental Disorders 4th edition, Text Revision (DSM-IV) to the DSM-5) (APA, 2013), discussions about its prevalence and severity have increased, and there is an ongoing debate regarding the actual prevalence rates of ADHD and whether ADHD is a misdiagnosed or perhaps over-diagnosed disorder (Cho et al., 2017; Paris et al., 2015; Polyzoi et al., 2018; Rigler et al., 2016).
Overall, we can observe different classifications of ADHD. The DSM-5 differentiates between three types of ADHD: predominantly inattentive presentation (314.00), predominantly hyperactive/impulsive presentation (314.01) and combined presentation (314.01). Unlike the DSM-IV, where the symptoms had to be present prior to age 7, the DSM-5 increased the threshold age to 12 (APA, 1994, 2013). This might be one of the causes for the increased reported prevalence of ADHD. Other changes include acknowledgment of ADHD as a problem that is persistent across the lifespan as well as recognition of the importance of various life settings (such as social, work or school settings) (Rigler et al., 2016). On the other hand, the ICD 10, which is the most commonly used classification system in the Czech Republic, does not have an ADHD diagnosis per se but includes a hyperkinetic disorder (“F 90”). Although similar symptoms are listed, the International Classification of Diseases (ICD) states that the disorder usually arises within the first 5 years of life (similar to the DSM-IV) (Organization, 2018). Although there is an overlap, the DSM-5 criteria allow for a broader group of individuals to be diagnosed with ADHD, and studies have found an increase of up to 65% when changing from the DSM-IV to DSM-5 criteria (Rigler et al., 2016). The overall prevalence of ADHD varies to a great extent. The DSM-5 reports a prevalence of 5% in children and 2.5% in adults.
Since ADHD had been accepted as only a childhood disorder for a long time, considerably fewer papers address adult ADHD (Ramos-Quiroga et al., 2014), and there is a lack of consistency in the diagnostic criteria for ADHD as a lifelong disorder (Matte et al., 2012). The ADHD prevalence among adults ranges between 2.9% and 7.11% (Hesson & Fowler, 2018; Van Wijk, 2020; Yallop et al., 2015). There has been an observed increase in the ADHD incidence as well as diagnosis among adults (Polyzoi et al., 2018; Vasiliadis et al., 2017; Zhu et al., 2018). Some studies, on the other hand, have argued that the observed imbalance of prevalence in childhood is no longer present in adulthood (Moffitt et al., 2015), and similar results were found in France with a prevalence of 2.99% without significant sex or age differences (Caci et al., 2014). Certain studies report more males than females being diagnosed (Montejano et al., 2011). In the Netherlands, researchers found significant age differences in the ADHD prevalence between the young elderly (60–70) and older participants (>70), with the young elderly individuals demonstrating higher levels of ADHD (Michielsen et al., 2012). Furthermore, in a German study, the reported prevalence was 4.7%, and significant associations with lower age, low educational level, unemployment, marital status (never married and divorced), and rural residency were found. There was no association with regard to sex (De Zwaan et al., 2012). Similar results were reported by Kessler and colleagues, who found the prevalence of adult ADHD to be 4.4%. The correlates in that study included sex and, similar to De Zwaan (2012), marital status and employment status (Kessler et al., 2010).
Currently, there is an interesting ongoing debate about the adult-onset of ADHD. The question of whether ADHD can arise in adulthood has been raised and supported by certain research (Moffitt et al., 2015). This view would change the perception of ADHD as a neurodevelopmental disorder. Although further support is needed, we may look more deeply into the etiology of ADHD if diagnosed in adulthood. Although this has been disputed by others (Solanto, 2019) since the majority of adults meet at least some diagnostic criteria for childhood ADHD, it is necessary to consider at least adolescent/late adolescent onset.
There is a high prevalence of ADHD among psychiatric patients, with Duran and colleagues reporting a 15.9% ADHD prevalence rate (Duran et al., 2014). One of the possibilities as to why ADHD often remains untreated is that the high number of psychiatric comorbidities, such as depression and substance use (Chen et al., 2016) or autism spectrum disorder (Nylander et al., 2013), may mask core ADHD symptoms. Furthermore, the treatment options are often limited (Ramos-Quiroga et al., 2013). A high prevalence of ADHD was also found among the adult incarcerated population in an all-female study, where 41% of prisoners were eligible for a diagnosis of ADHD (Farooq et al., 2016).
ADHD symptoms can be measured using various tools. Some of the most common scales are ADHD rating scales (Adshaw & Mal, 2017) or direct tests of attention, such as the continuous performance test (CPT) (Sims & Lonigan, 2012), clinical interviews or various forms of self-report questionnaires (Gray et al., 2014; Weissenberger et al., 2018). One of the self-reporting questionnaires is the Adult ADHD self-report questionnaire v 1.1 (ASRS) This screening tool is a 6-question scale developed by a WHO working group, and its aim is to screen for adult ADHD. Its longer version comprises 18 questions. ASRS was developed in accordance with DSM-IV criteria. For use in the general population, the short screener tool has been shown to be not only more specific but also more sensitive than the longer 18-question version. The reported specificity was 99.5%, and the sensitivity was 68.7% (Kessler et al., 2005, 2010). For the 0 to 6 scoring approach, the sensitivity was 39.1 %, and the specificity was 88.3 % (Kessler et al., 2007). The WHO working group led by Kessler also tested its validity and found that the internal consistency reliability was in the range of 0.63 to 0.72, and the test-retest reliability (Pearson correlations) was in the range of 0.58 to 0.77 (Kessler et al., 2007). Furthermore, they demonstrated strong concordance between the screening tool and clinical diagnosis. Strong test-retest reliability was also shown in individuals without ADHD (Silverstein et al., 2018).
Self-reporting can serve as a good starting point for steering patients toward diagnostic services and psychiatric evaluation (Van Wijk, 2020). The use of different diagnostic tools as well reliance on one method, such as self-reports or neurological tests, may represent another contributing factor to differences in the reported prevalence of ADHD across studies.
To date, most studies have closely examined the prevalence of ADHD symptomatology in childhood and produced conclusive results. However, mixed results regarding the prevalence and predictors of symptomatology in adulthood are apparent (Matte et al., 2012). The relationship with sex is not unambiguous, but some studies have reported an association with sex, thus suggesting that it remains the same as in childhood (Caci et al., 2014; Montejano et al., 2011). Other studies claim that the differences disappear with age. ADHD symptomatology does appear to decrease with age, as has been found in most studies (Caci et al., 2013). The question of an association with education also remains uncertain. Our study aims to deepen the understanding of ADHD as a lifelong disorder and tries to map the aforementioned associations with sex, age and education.
The main purpose of this paper was to explore the prevalence of ADHD symptomatology within the Czech population. Our hypotheses were as follows. There will be a higher level of ADHD symptomatology among males than females. Furthermore, ADHD symptomatology will have an inverse relationship with age, with the younger population having a higher level of symptomatology. Lastly, there will be an inverse relationship with education: the lower the symptomatology, the higher the education that will be achieved.
Materials and Methods
This study was approved by the ethics committee of the General Teaching Hospital in Prague. Data was collected in cooperation with the research company STEM/MARK. The computer-assisted web interviewing (CAWI) method was used for online questioning, and the participants were chosen from a panel of respondents (namely, the European National Panel). Due to the sensitivity of the scope of the topic and a target group aged 18 to 65, this method of data collection was deemed the most appropriate for a cross-sectional type of research design.
Voluntary participation in the study was ensured by the ethical consent form, and respondents were informed about their right to not answer questions and their right to withdraw from the study at any time.
Quantitative data collection was conducted between January 2nd, 2019, and January 14th, 2019. The current study used the following exclusion criteria: 1. the presence of a severe neuropsychiatric disorder (especially intellectual disability, schizophrenia, psychosis, severe forms of mood disorders, dementia, substance dependence, behavioral addiction or neurodegenerative diseases); 2. severe somatic disorders with a direct effect on cognitive/executive functions (especially cardiovascular, cerebrovascular or endocrinological diseases) and 3. use of drugs that potentially affect cognitive functioning. This was controlled by prior contact with the participants through informed consent before they agreed to participate in the research.
In the first wave we collected 1,518 completed questionnaires. Furthermore, the respondents were asked whether they would be interested in joining the second, qualitative round of data collection; in total, 899 respondents consented to participating in the next part of the study. The second part of the data collection will be a follow-up study that will incorporate qualitative methodology, and participants will be asked in depth about the impact of their ADHD diagnosis on their daily lives.
The European panel of respondents provides a representative sample that covers age, sex, education and place of residence as variables corresponding to the Czech population. Although the cross-sectional design has certain limitations, it was the best choice for this type of research. The respondents completed a demographic questionnaire focusing on ADHD history, risky behaviors and lifestyle. The test battery also included 3 standardized questionnaires: the Wender Utah Rating Scale (WURS) created by Ward, Wender and Reimherr for ADHD symptomatology in childhood (Ward et al., 1993); the Adult ADHD Self-Report Scale (ASRS) by the WHO working group (Kessler et al., 2005) for ADHD symptomatology in adulthood and the Zimbardo Time Perspective Inventory (ZTPI) by Zimabrdo and colleagues for time perception styles (Zimbardo & Boyd, 1999). Data from the WURS and ZTPI has been reported in other articles (Ptáček, in preparation and Weissenberger, in print) and were not analyzed for the current study.
The data was analyzed using the statistical software R (R Core Team, 2019). We tested the differences between demographic variables using a t-test (sex) or one-way analysis of variance (ANOVA) (education and age categories). The effect size was expressed using Cohen’s d for t-tests and by
To answer our research question regarding the prevalence of ADHD symptomatology within the Czech population, data from the ASRS questionnaires on current ADHD symptomatology were analyzed. The screening part of the ASRS 1.1 questionnaire comprised six five-point Likert questions. For each question, a score of 0 or 1 is computed based on the following criteria. For the first two questions, the respondent scores 1 for a response of at least “Sometimes”, while for the remaining four questions, the respondent scores 1 for a response of at least “Often”. For the first part of the data analysis, we treated the ASRS results as a categorical variable according to the “traditional” approach to the ASRS. The traditional approach means that the first part of the questionnaire–the screening part–was assessed, and the participants were divided into two groups: “no ADHD” (score 0–3) and “suspected ADHD” (score 4–6). To test the significance of the explored variables, we also decided to treat the ASRS scores as a continuous variable. To do this, we computed the sum of the first 6 questions used as screening tools, and we obtained a number ranging from 0 to 6.
Results
The total number of respondents was N = 1,518 (766 males), with a mean age of 41.56 years (SD = 13.64). From the sample, 3% of the respondents reported having been diagnosed with ADHD/hyperkinetic disorder in their lifetime. Across the sexes, this was 4% of the men and 2% of the women. The average age at which they were diagnosed was 9.85 years (SD = 7.29); the age distribution is presented in Figure 1. We can see that the majority of diagnoses were before the age of 10, and most cases were diagnoses prior to the age of 15. However, our data shows rare cases of ADHD being diagnosed in later life, suggesting that it is already being perceived by some experts as a lifelong disorder. In 34.5% of the cases, the diagnosis was made by an expert in pedagogical-psychological counseling, followed by a child psychiatrist (24.1%) and a general practitioner (20.7%). Only one case was diagnosed by a clinical psychologist. We believe these results suggest high heterogeneity in the diagnosis of ADHD. There are several reasons for the inconsistency in the diagnosis of ADHD across different professions in the Czech Republic. First, GPs have very limited training in psychiatry, and they unfortunately lack skills in using diagnostic manuals, clinical scales and questionnaires. They are not accustomed to diagnosing mental disorders. Second, the diagnosis of ADHD in adults is a new issue in the Czech Republic. Thus, not only GPs but also a number of adult psychiatrists do not recognize ADHD as a lifelong disorder and are not considering ADHD in their patients. Third, both GPs and pediatricians in the Czech Republic are not allowed to prescribe the first- and the second-line treatment for ADHD (stimulants and atomoxetine). Lastly, a total of 10.7% of the respondents reported that a close family member had been diagnosed with ADHD/hyperkinetic disorder

Age, when ADHD was first diagnosed.
As seen in Table 1 according to ASRS scoring, a total of 119 respondents were classified as suspected ADHD. This corresponded to 7.84% of the whole sample. Overall, more males than females had ADHD symptomatology. The difference in ASRS scores between the sexes was small but significant (t(1,516) = −1.98, p = .048, d = –0.1), with slightly higher ASRS scores for males (M = 1.33, SD = 1.40) than for females (M = 1.19, SD = 1.34).
ADHD Symptomatology as Assessed by ASRS between Sexes.
ADHD = Attention Deficit, Hyperactivity Disorder.
Regarding age, we found significant differences across age ranges (F(3, 1514) = 22.72, p < .001,
ADHD Symptomatology Across Age Categories Performed by Traditional Approach to ASRS Questionnaire.
ADHD = Attention Deficit, Hyperactivity Disorder.
Regarding education, the differences in ADHD symptomatology were not significant (F(3, 1514) = 1.57, p = 0.194,
ADHD Symptomatology Across Education Categories.
ADHD = Attention Deficit, Hyperactivity Disorder.
The ASRS scores were significantly lower in respondents who were not diagnosed with ADHD in childhood (M = 1.21, SD = 1.34) compared with participants who were diagnosed with ADHD in childhood (M = 1.21, SD = 1.34; t(1516) = –7.74, p < .001, d = 1.16).
Discussion
Overall, 3% of our sample reported being diagnosed with ADHD/hyperkinetic disorder in their lifetime. Kessler reported that with the 0 to 6 scoring approach, which we adapted, the sensitivity is 39.1% and the specificity is 88.3%. This would mean that our prevalence is underreported; according to Kessler, the predicted prevalence is as high as 14% (Kessler et al., 2007). Due to the way our data was collected, we cannot establish empirical evidence for our population because clinical diagnosis was not performed. More males than females reported a previous ADHD diagnosis. These results are consistent with the findings by Kessler (Kessler et al., 2010). When we looked at the ASRS results, we found significant differences between the sexes, unlike in other studies such as De Zwaan (2012). Furthermore, we also found age differences, with younger people demonstrating higher ADHD symptomatology compared with older individuals. These results corresponded to those by Michielsen and De Zwaan (De Zwaan et al., 2012; Michielsen et al., 2012). However, it remains unknown whether these cases were present in childhood as well or whether we mapped the prevalence of symptomatology that appeared only in adulthood. For this reason, we cannot confirm the previously suggested hypothesis that ADHD can be adult-onset only (Moffitt et al., 2015). Further longitudinal research is needed. The observed differences may be due to the lack of awareness about ADHD as a lifelong disorder. This may mean that adults may overlook their symptoms or that they do not necessarily view them as an indication of difficulty because they have experienced and managed them their whole life. This might also mean that adults who have ADHD but have never been diagnosed have unknowingly developed coping strategies that help them to navigate their daily life. This hypothesis will be tested in our second round of study, the qualitative part of the study. Furthermore, we can assume that the severity of symptoms decreases over time and therefore is more manageable in adulthood. Lastly, we can also assume that the severity of the symptoms might not have been dominant, and therefore these individuals remained undiagnosed into adulthood. Education differences yielded no significant results; however, similar to De Zwaan (2012), we observed a pattern of decreasing ADHD symptomatology with increasing education levels. These results do not mean that reaching a postgraduate level of education is a protective factor for ADHD. However, research shows only a modest association between IQ and attention deficits (Jepsen et al., 2009), and the ADHD diagnosis is equally valid among children with high IQ as those with average IQ (Antshel et al., 2007). A possible explanation is that individuals who complete a postgraduate level of education with ADHD diagnosis compensate for this with a higher IQ than their peers who do not have ADHD symptomatology. However, this hypothesis needs to be further explored because ADHD adults with high IQ suffer more from executive function impairments than their peers (Antshel et al., 2010; Brown et al., 2009).
The ASRS results showed that 7.84% of the population would be eligible for an ADHD diagnosis according to this screening tool, which corresponded to the results for the prevalence rates found by Yallop (Yallop et al., 2015). Because our study was based on self-reporting, our results demonstrate the need for increased awareness of adult AHD; as Van Wijk suggests, the next step should be steering suspected ADHD participants into psychiatric examination (Van Wijk, 2020). Although the ASRS has high concordance with actual clinical diagnosis, the sensitivity and specificity suggest that our results are underreported.
Our study also has certain limitations. We are aware that the cross-sectional design has general weaknesses; however, because we were not attempting to explore causal relationships, we believe the design was well justified for our study. Nonetheless, due to the nature of the design, we were not able to establish whether the observed ADHD symptomatology was truly lifelong or whether we also engaged with individuals who experienced difficulties only in adulthood. We are aware that although ASRS has shown a strong concordance with clinical diagnosis, ASRS is a screening tool and not a diagnostic tool (Kessler et al., 2005). Due to the large scale of our study, we were unable to use clinical interviews, and this decreased the precision of symptomatology estimation. Therefore, we were careful to present our results as ADHD symptomatology rather than as an ADHD diagnosis. The data also clarify the further need for the translation and validation of screening tools for ADHD, as it is clear that ADHD is a lifelong disorder.
Conclusion
With this study, our goal was to examine the point that has been made by the American Psychiatric Association that ADHD is a disorder prevalent in adulthood; this was achieved by examining the prevalence of adult ADHD in a representative sample of the Czech Republic according to several demographic variables. We are confident that with more empirical research, such as ours, mental health professionals in the Czech Republic and elsewhere will take the issue of adult ADHD as seriously as its childhood counterpart. Our study shows that there is a prevalence of ADHD symptomatology among the adult Czech population that is similar to other studies that have been conducted internationally. Variables such as sex and age play a significant role in ADHD symptomatology, as we had expected. Males and the younger population showed increased symptomatology compared with females and the older population, thus shining light on the issue of sex disparity. In our sample, education was not a significant predictor; nonetheless, we observed an interesting pattern. Among those with higher levels of school education, there were fewer people with ADHD symptomatology. We hope that our research will also shed light on the methods and tools used for diagnosis so that clinicians will not misdiagnose the condition. Although using a self-report screening tool is a starting point, we need ensure that the diagnosis shifts from GPs to psychiatrists and clinical psychologists only. Future research should also focus on the question of whether ADHD symptomatology can be prevalent in adulthood without any prior presentation because we hypothesize that some individuals may remain undiagnosed due to milder symptomatology.
Footnotes
Authors’ Note
Work was done in the Department of Psychiatry, First Faculty of Medicine Charles University in Prague and General University Hospital in Prague, Prague, Czech Republic.
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 by the GAČR – 1811247S, Progres Q 06 1LF, RVO 68081740.
