The Utility of the CPT in the Diagnosis of ADHD in Individuals with Substance Abuse: A Systematic Review

Attention-deficit/hyperactivity disorder (ADHD) is the most prevalent neurodevelopmental disorder of childhood, with an estimated rate of 9.4% of children in the USA [1]. The disorder is associated with adverse academic, social, emotional, and physical outcomes [2, 3] and persists into adulthood in 60% of children [4]. ADHD is strongly linked to the development of substance use disorders (SUD) [5-7], often at a very early age and with a faster transition from less severe to more severe forms of SUD [8]. The earlier onset and increased use of substances in adolescents with ADHD compared with controls has been demonstrated in various retrospective [9] and prospective studies [10, 11]. A meta-analysis of cohort studies has shown that childhood ADHD significantly increased the risk for nicotine use in middle adolescence and the risk for alcohol use disorder during young adulthood [12]. International research revealed large variance in the prevalence of comorbid ADHD in adult SUD patients [13, 14], which could only partly be explained by differences in ADHD assessment instruments [6]. An international multicenter, cross-sectional study in 10 countries (8 European countries, Australia, and the USA) found that more than 40% of the treatment-seeking SUD patients screened positive for ADHD, with DSM-IV adult ADHD prevalence rates ranging from 5.4% in Hungary to 31.3% in Norway [7]. While the mechanism associating ADHD with SUD remains unclear [15], family genetics, impulsive behavior, self-medication, and shared neuropsychological and biological pathways were offered as possible candidates [16-18].

Recent studies have established the clinical relevance of ADHD symptoms (impulsivity, disinhibition, executive function deficits, and slowed mental processing) for the prediction of SUD development and persistence [19, 20]. These studies suggested that treatment outcomes for SUD may be improved by early and accurate identification and targeting ADHD symptoms in this population. Nevertheless, overlapping symptoms between ADHD and comorbid SUD represent challenges for diagnosis and treatment. Currently, no self-report ADHD scale has been sufficiently validated in a population with ADHD and SUD, and up to 20–40% of the patients with ADHD are missed when using a single screener [21-23]. The underreporting of ADHD symptoms by patients with SUD may reflect the cognitive deficits associated with both disorders, including limited mentalization abilities, underestimation of ADHD-related impairments, and difficulties in linking actual impairment with ADHD [24]. Diagnosing ADHD in adult patients with SUD requires the collection of retrospective information. However, this is often difficult to obtain because of the lack or inadequate anamnestic data [25, 26]. Moreover, because the link between ADHD and SUD is usually investigated in treatment-seeking SUD patients, sociocultural influences on SUD treatment utilization may affect the likelihood that ADHD symptoms will be identified and reported [27-29].

Less frequent, but important, is the problem of overdiagnosing ADHD in SUD patients. Overdiagnosing can occur when substance-induced or withdrawal symptoms are erroneously perceived as ADHD symptoms or when the persistence of ADHD symptoms into adulthood is overlooked [30]. Thus, assessing ADHD symptoms using objective, easy-to-collect variables by noninvasive methods might be useful as a supportive measure in the detection and evaluation of comorbid ADHD and other neurological/psychiatric disorders in SUD patients [31].

Continuous performance tests (CPTs) are considered a key component of any ADHD assessment and are currently the most popular objective laboratory-based measures to support a clinical diagnosis of ADHD [32-34]. The majority of CPT tasks measure selective attention, sustained attention, and impulsivity [35-37] and can be used alongside clinical evaluation to inform the diagnostic process of ADHD [38, 39]. A typical CPT includes a serial presentation of visual or auditory target and nontarget stimuli (numbers, letters, or geometric figures) [35]. Response to nontarget stimuli is referred to as a “commission error” and is considered to measure impulsivity. The absence of response to target stimuli is referred to as an “omission error” and is assumed to measure inattention. Other standard measures include the number of correct responses, response time (RT), and variability in RT.

Several studies have supported the utility of the CPT in the diagnostic process of ADHD in children (see Hall et al. [39] for review). Research in adults, however, revealed that the CPT is sensitive to multiple conditions that involve central nervous system dysfunction, such as schizophrenia [40], mood disorders [41], and anxiety [42], but has low specificity for ADHD [43, 44].

Despite the popularity of the CPT in the diagnostic process of ADHD, only a few studies compared its utility to that of other laboratory-based tests in identifying ADHD. A meta-analysis by Nichols and Waschbusch [45] of the validity of laboratory cognitive tasks for ADHD, including the CPT, the Gordon Diagnostic System (GDS), the Children’s Checking Task (CCT), the Delay of Gratification Tasks, the Choice-Delay Task (C-DT), and the Stop Signal Task (SST), supported the use of the SST, the C-DT, and the CPT for children with ADHD. Similarly, a study examining test-retest reliability of the CPT and the SST in children with ADHD found that the two tests yielded reliable measurements in ADHD children [46].

Over the past 2 decades, many studies have focused on the relationship between acute and chronic substance use and cognitive functions, such as attention and response inhibition. Findings, however, were inconsistent. CPT impairment has been associated with chronic use of some substances, such as heroin [47] and cocaine [48], but not with alcohol [49] or nicotine [50, 51]. Differences between studies reflect methodological constraints (e.g., small samples) and variations in participants’ age, severity, type of substance used, the use of extreme group designs (i.e., comparing control participants and problematic or disordered substance users) [52, 53], and different task requirement [54].

Importantly, while some studies examined the effects of chronic alcohol or nicotine use, others measured the acute and immediate effects of their administration. There is evidence to suggest that chronic smoking is associated with lower CPT performance [55], whereas nicotine administration is associated with improved CPT performance in both smokers and nonsmokers [56, 57].

Although patients with SUD may show similar attentional deficits to those observed in ADHD, data on the CPT performance of ADHD patients with comorbid SUD are currently very limited. Thus, it is difficult to conclude whether such a decreased performance reflects the comorbidity of ADHD or the cognitive deficits specific to SUD. As disinhibition control, impulsivity, and executive function deficits have been associated with worse substance use outcomes [58-61], clarifying the specific cognitive profile of SUD with and without ADHD may be of significant clinical relevance.

Although research on the shared genetic, behavioral, and neurobiological correlates of ADHD and SUD is well established, there have been only a few studies comparing the cognitive profiles of the 2 disorders [62]. Several studies suggested that patients with comorbid ADHD and SUD may exhibit higher levels of impulsivity compared with either ADHD [62, 63] or SUD [64, 65] alone. However, others failed to find such differences [66]. The aim of this review is to directly examine the quantitative similarities and differences in CPT performance of adults with ADHD, SUD, and their comorbidity to determine if a distinct neurocognitive profile exists for each. To this aim, a systematic review of CPT studies involving adults with combined ADHD and SUD and a comparison group of one of the disorders alone was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [67]. Given that both ADHD and SUD are associated with impaired attention functions, it was expected to find the most severe CPT deficits in adults with comorbid ADHD and SUD compared with either disorder alone.

To locate studies for this review, PsycINFO, PubMed, and Google Scholar were searched. The search consisted of the terms “ADHD”/attention” and “Continuous Performance Test/CPT” with the following keywords: “substance use/substance use disorder/SUD,” and/or “drug,” and/or “addiction,” and/or “cannabis,” and/or “cocaine,” and/or “amphetamine” and/or “heroin” and/or “nicotine,” and/or “alcohol.” Reference lists of studies were also searched to find other relevant articles. Retained studies had to meet 3 criteria. First, studies had to include empirical data. Second, studies had to include patients with comorbid ADHD and SUD and a comparison group of at least one of the disorders alone. Finally, studies had to include CPT results. No restriction was made on the date of publication, the language of publication, and the number of participants. Also, no restriction was made regarding the formal diagnosis of ADHD or SUD. In other words, participants were not obliged to meet the formal DSM-based criteria of ADHD or SUD. The search ended in March 2020.

Figure 1 presents a PRISMA flow diagram of the study selection process. The database search identified 639 studies, searching for reference lists of reviews, and primary studies identified 3 additional studies. Contacting experts in the field revealed one more study. The screening phase involved examining the titles and abstracts of all 643 studies identified. This process resulted in 594 studies being excluded as they were deemed unsuitable for the present review.

Consequently, 49 studies were selected for the eligibility phase. Of these, 29 studies did not target the population of interest (i.e., adult patients with comorbid ADHD and SUD). These studies included children with prenatal exposure to substances, patients with ADHD that used methylphenidate (MPH) for treating ADHD-related symptoms, and nonhuman subjects. Nine additional studies addressed the influence of substance use on CPT performance, but participants were given substances as part of the experimental design. Three more studies were review papers and not empirical studies. After removing these 41 articles, 8 articles were included in the review. All studies were inspected for their year of publication, sample size, substance type, and gender distribution. Cohen’s d was chosen as the common effect size.

Eight studies met the review inclusion criteria. Table 1 provides an overview of the authors’ names, publication year, sample size, gender distribution, substance type, CPT used, and diagnostic criteria of ADHD and SUD. Cohen’s d effect sizes are reported (Table 2) for the 4 studies that provided sufficient data for these to be calculated [62, 66, 68, 69].

As seen, except for 1 study [66], all sample sizes were limited and ranged between 17 and 72 participants. All studies except 2 [68, 69] included more males than females (52–100% males). Studies significantly differed in the substances used by participants, including nicotine (4 studies), morphine (1 study), cannabis (1 study), cocaine (2 studies), alcohol (1 study), and amphetamine (1 study). Three [68, 70, 71] assessed both childhood and adult ADHD, while the rest included only adult ADHD.

Table 2 presents CPT outcomes for all included studies. Study findings were integrated across the 4 key CPT indices: commission errors, which are considered a measure of impulsivity; omission errors, which indicate deficits in sustained attention; mean hit reaction time (RT), which measures response speed; and RT variability, which measures the inconsistency in response speed and fluctuations in attention [72-75].

Four studies [68-70, 76] reported that the effect of substance use on the rate of commission errors differed between individuals with and without ADHD. Three of these studies addressed the effect of nicotine. Kollins et al. [76] found that the increase in commission errors on an abstinent day compared with a day after smoking as usual was nominally larger for smokers with ADHD than for smokers without ADHD, but the effect was not significant (p = 0.09). Similarly, McClernon et al. [68] found that smokers with ADHD made more commission errors than smokers without ADHD in the abstinent condition. Supporting the ameliorating effect of nicotine on commission errors in patients with comorbid ADHD, Rigbi et al. [69] found that nicotine administration strongly reduced the number of commission errors in ADHD patients with the highest levels of ADHD symptoms and impulsivity. There is also evidence that the presence of SUD in individuals with ADHD affected the rate of commission errors compared with patients with ADHD alone. Ginsberg et al. [70] found that inmate males with comorbidity of ADHD and SUD conducted more commission responses than psychiatric patients with ADHD alone. The 4 other studies [62, 66, 71, 77] did not find that the presence of ADHD in individuals with substance use was associated with different commission rates compared with individuals with either SUD or ADHD.

Except for the study of Ginsberg et al. [70], which found that inmate males with comorbid ADHD and SUD made more omission errors than psychiatric patients with ADHD alone, no study found differences in the rate of omission errors between patients with comorbid ADHD and SUD and patients with either ADHD [62, 66, 71] or SUD [66, 68].

Several studies reported that comorbidity of ADHD and SUD was associated with more deficits in RT variability and/or RT mean hit compared with patients with ADHD alone. Ginsberg et al. [70] found that while the ADHD group did not differ from the comorbidity group in the mean RT, RT variability was higher in the comorbidity group. Likewise, Miguel et al. [62] reported that cocaine-dependent patients with ADHD had longer RT and increased RT variability compared with individuals with ADHD alone. Finally, Slobodin et al. [66] found that patients with ADHD made more accurate and fast responses than participants with SUD and participants with both ADHD and SUD. This study also found that although patients with SUD and patients with comorbid ADHD and SUD showed impaired RT in comparison with healthy controls, no differences were found between the comorbid and the SUD-only groups. Similarly, Wallace et al. [77] reported that cannabis users demonstrated slower responses compared with controls, regardless of the level of ADHD symptoms.

The 4 studies that addressed the effects of nicotine use on RT mean and variability on individuals with and without ADHD showed inconsistent results [68, 69, 71, 76]. Levin et al. [71] demonstrated that nicotine has a facilitating effect on smokers’ RT accuracy and variability so that smokers with ADHD showed a significant nicotine-induced speeding and reduced RT variability, whereas the nonsmokers with ADHD did not. Similarly, McClernon et al. [68] reported that smokers with ADHD demonstrated higher RT variability after overnight abstinence compared with smokers without ADHD. Nevertheless, the 2 other nicotine studies showed that the facilitating effect of nicotine on the cognitive performance of smokers was independent upon ADHD symptomatology [69, 76].

Despite increasing awareness, ADHD still features as “hidden” comorbidity, often not diagnosed even in patients under psychiatric treatment for other psychiatric disorders [78]. In SUD patients, accurate identification of ADHD might be further complicated by cognitive and psychosocial problems that hinder individuals’ ability to reflect upon their difficulties and report them [24-26].

This systematic review aimed to characterize differences and similarities in CPT performance of adults with ADHD, SUD, and their comorbidity. A literature search identified 8 eligible studies that included the measurement of ADHD in adults with SUD. Eligible studies covered a wide range of substance types, substance use severities, and study populations (inpatient, outpatients, inmates, and the general population). Overall, studies were limited in their total sample sizes, particularly in the proportion of females. Synthesis of results suggests that in terms of omission and commission errors, the comorbidity of ADHD and SUD was not necessarily associated with more deficits than either disorder alone. There was only one study that reported that individuals with comorbidity of ADHD and SUD conducted more commission and omission responses than psychiatric patients with ADHD alone [70]. According to this finding, SUD had an additive negative effect on the CPT performance of patients of ADHD. However, because the study compared CPT performance of long-term inmates diagnosed with both ADHD and SUD with high-functioning outpatients diagnosed with ADHD alone, it is difficult to isolate the influence of SUD from other cognitive, social, intellectual, and functional factors.

Examining the effects of nicotine on commission errors, Rigbi et al. [69] reported that nicotine administration decreased the number of commission errors with the stronger improvements in participants with more ADHD symptoms. In addition, nicotine was associated with improved RT means and decreased RT variability [69, 76, 77], regardless of the presence of ADHD. These results provide some support for the self-medication hypothesis, which posits that individuals with ADHD may use nicotine to attenuate their symptoms [17, 20]. Previous research suggested that as the typical ADHD psychostimulant medication MPH, nicotine acts as an indirect dopamine agonist, improving attention and arousal and reducing impulsivity in individuals with ADHD [79, 80].

While omission and commission rates are considered reliable indicators of ADHD in children [81, 82], in adults, these variables do not provide enough variability between individuals due to a ceiling effect in performance [83, 84]. Therefore, the more typical outcome measure of most CPT variations is the variability of RT [72]. The current review identified 3 studies that reported that the comorbidity of ADHD and SUD was associated with more deficits in RT variability and/or RT mean hit compared with patients with ADHD alone [62, 66, 70]. Furthermore, 2 studies suggested that substance use negatively affected RT, regardless of ADHD [66, 77]. Together, these findings confirmed that inconsistency and variability are core cognitive symptoms associated with both ADHD [85-87] and SUD [88, 89]. However, the scarcity of research in this area does not allow us to draw a firm conclusion about whether both disorders have an additive effect on RT mean and variability.

Overall, the current review does not support the utility of the CPT in discriminating comorbid ADHD and SUD from either disorder alone. Included studies suggest that while the CPT might be sensitive to attentional deficits, such as disinhibition, impulsivity, and response speed, it lacks specificity for the diagnosis of adult ADHD, SUD, or their comorbidity. This finding is in contrast to previous studies that found that when ADHD co-occurred with other psychiatric disorders, such as autism syndrome disorder [90] or depression [91], CPT performance was worse compared with ADHD alone. One possible explanation for this discrepancy is that the neurobiological and behavioral similarities between ADHD and SUD (e.g., deficits in attention and working memory, slowed mental processing, and impulsivity) are larger than the similarities between ADHD and other psychiatric disorders. It is also possible that because many individuals with ADHD start to use substance early, it is difficult to disentangle the independent impact of SUD on cognition [92]. As ADHD is associated with developmental delays [93, 94], some of them are evident in the CPT [95, 96], the additional impact of SUD on CPT performance was difficult to detect. Finally, the lack of consistent differences between the cognitive profiles of ADHD and SUD may also be attributed to the history of ADHD diagnosis. It has been suggested that premorbid childhood ADHD diagnosis is associated with more significant attention deficits, compared with current ADHD symptoms [77]. The fact that only 2 included studies addressed childhood ADHD [70, 71] may explain why the current review failed to identify consistent differences between the cognitive profiles of patients with and without ADHD.

Finally, it should be noted that although the CPT is currently the most popular objective measure of ADHD-related symptoms, other neurocognitive paradigms, such as the Go/No-Go (GNG) task and the SST, showed sensitivity to attention and inhibition deficits in chronic substance users (e.g., [53, 68, 97]) and following drug administration (e.g., [98]). Despite observed similarities, these paradigms capture different functions of attention and inhibition [99]. However, their advantage over the CPT in identifying ADHD in individuals with SUD remains unclear [50, 68].

Interpretation of findings involving patients with ADHD, SUD, or both entails the distinction between the diagnoses of ADHD/SUD and their classifications. Diagnostic criteria are a set of signs, symptoms, and tests that reflect the heterogeneous features of a disease [100, 101]. Classification criteria, on the other hand, are standardized definitions that are primarily aimed at establishing well-defined, relatively homogenous cohorts for clinical research. If sufficient internal and external validity for diagnosis is demonstrated in a given population, classification criteria can be diagnostic. However, because disease features are not typically identical among patients with a given disease, meeting classification criteria is not equivalent to carrying a given diagnosis [101, 102].

The findings of this systematic review should be considered under certain limitations. First, the studies described here were not always comparable because of important differences in methodologies. For instance, while some studies included patients with a formal DSM-based diagnosis of ADHD and SUD, others used self-report measures. Second, most studies have been performed in small samples, almost all of them under 100 participants. Third, due to the cross-sectional nature of all studies, it was impossible to determine the timing and causality of the ADHD symptomology, SUD, and cognitive performance. Prospective CPT studies will enable us to examine whether and how ADHD symptoms predispose SUD and will enhance the development of specific interventions. Fourth, due to the shortage of studies and insufficient information, it was impossible to analyze findings regarding ADHD presentations. As ADHD presentations were found to be associated with both CPT performance [103] and substance abuse [104], further investigation is warranted.

There is also a need to explore the utility of other CPT measures, including distractibility and hyperactivity, in discriminating ADHD, SUD, and their comorbidity. There is evidence that the number of hyperactive responses during CPT performance distinguished between individuals with ADHD (with or without comorbid SUD) and individuals with SUD alone (hyperactivity was higher in the ADHD group and the comorbid group) [66]. This finding is consistent with previous studies that identified motor activity as the most predictive marker of ADHD [105, 106], thus emphasizing the importance of extending our investigation to other cognitive functions.

This review highlights the shortage of large-scale CPT research involving patients with ADHD and SUD. It is, therefore, impossible to determine whether the presence of ADHD in individuals with SUD is associated with decreased cognitive performance or whether both disorders have an additive effect on the CPT.

Early identification of ADHD in individuals with SUD is important because ADHD has an adverse effect on the development and course of SUD. A failure to diagnose ADHD poses significant sociocultural, educational, employment, relationship, and coping deficits [107, 108] and increases the risk for additional comorbidities [109]. Diagnosis of ADHD is also important because it is often accompanied by recommendations for intervention [110]. The pharmacological treatment of ADHD has been a source of controversy [111, 112]. Several researchers showed that individuals with ADHD and SUD experience positive treatment effects regarding both ADHD and SUD-related outcomes [113, 114]. However, it has also been noted that ADHD treatment may not be as effective in individuals with active SUD [64], partly due to drug-induced changes in the brain and similarities in the mechanism of action between stimulant medication and abused substances [115].

The International Consensus Statement on Screening, Diagnosis and Treatment of SUD patients with comorbid ADHD recommended that the diagnostic process of ADHD in SUD patients should be started as soon as possible, given that are no serious withdrawal symptoms or intoxication [30]. A prospective test-retest study of adults with ADHD and SUD [116] showed that most of them (95.3%) still fulfilled DSM-IV adult ADHD criteria at re-evaluation 2–3 months later, indicating that active substance use should not prevent a diagnostic evaluation.

An important direction for future CPT studies may be to explore whether specific attentional deficits among patients with SUD accounts for their increased risk for treatment failure. Identifying which cognitive deficits are most relevant to SUD may contribute to the development of possible intervention approaches.

Ethics approval was not required for the current literature review (in consultation with the ethics committee of the Department of Education, Ben-Gurion University).

The author declares no conflicts of interest.

The author did not receive any funding.

O.S. was the only author of this manuscript and was responsible for data search, analysis, and writing the manuscript.