Abstract
ADHD is one of the most common psychiatric disorders of childhood that persists through adulthood in around two third of the individuals (Faraone et al., 2006; Lara et al., 2009). The prevalence of ADHD in adults is estimated between 2% and 4% (Fayyad et al., 2007; Kessler, Chiu, Demler, Merikangas, & Walters, 2005; J. J. Kooij et al., 2005; Simon, Czobor, Balint, Meszaros, & Bitter, 2009). Expression of ADHD can change with age; hyperactive symptoms appear to be typical of young children, whereas inattention represents a relatively pervasive developmental characteristic (S. J. Kooij et al., 2010). Longitudinal studies have shown that ADHD has long-term disabling consequences, for example, school failure, relational problems such as rejection, and familial conflicts (Biederman et al., 1995; Taylor, Chadwick, Heptinstall, & Danckaerts, 1996). Furthermore, it is a predictor of adult mental health problems (Mannuzza, Klein, & Moulton, 2008). In adulthood, ADHD is associated with work difficulties (Kessler, Adler, et al., 2005), psychosocial impairment (Biederman et al., 1993), and an increased risk of mood, anxiety, and substance use disorders (SUD; Barkley, 2002; Biederman, Petty, Clarke, Lomedico, & Faraone, 2011; Young, Toone, & Tyson, 2003). Comorbidity is a frequent condition in ADHD; 75% of patients develop a comorbid disorder across life span (J. J. Kooij et al., 2005), SUD being one of the most prevalent. Several studies found that more than 30% of ADHD patients present a SUD across life span (Mannuzza, Klein, Truong, et al., 2008; Wilens et al., 2011). One study found that 32% to 53% of adults with ADHD have alcohol use disorders and 8% to 32% of the patients have problems with illegal drugs (Barkley & Gordon, 2002). The most frequent illegal substance abuse for the ADHD patients is cannabis followed by cocaine (Sullivan & Rudnik-Levin, 2001).
On the other hand, the lifetime prevalence rate of SUD reported in the general population is about 15% to 18% (Kessler, Berglund, et al., 2005). The economic impact of SUD demonstrates significant cost implications on the individual, family, and society (Demyttenaere et al., 2004; Khalsa, Treisman, McCance-Katz, & Tedaldi, 2008). In this regard, the economic burden of alcohol was estimated to equate to 0.45% to 5.44% of gross domestic product (Thavorncharoensap, Teerawattananon, Yothasamut, Lertpitakpong, & Chaikledkaew, 2009). Clinical and epidemiological studies have revealed a high co-occurrence of substance abuse and psychiatric disorders, a troublesome phenomenon given the consequences at medical, social, and economical levels (Kessler, Berglund, et al., 2005; Wilens et al., 2011). A reliable and valid diagnosis of a psychiatric disorder in substance abusers is far from being simple (easy) mainly due to the fact that (a) the effects of substance abuse (chronic or acute) can mimic symptoms of many other psychiatric disorders and (b) psychiatric diagnoses are not defined by direct biological markers but by a constellation of symptoms (Torrens, Serrano, Astals, Pérez-Domínguez, & Martín-Santos, 2004).
It is also important to take into account the high rates of ADHD within SUD populations, as reported by several studies that estimate prevalence ranges from 11% (Modestin, Matutat, & Wurmle, 2001) to 54% (Ohlmeier et al., 2008). The prevalence of ADHD can be different regarding the substance of abuse. In this way, patients with alcohol use disorder present a rate of 35% to 71%, patients with cocaine use disorder present a rate of 10% to 35%, and patients in treatment in methadone maintenance programs show a 17% frequency of ADHD (Levin, Evans, & Kleber, 1998; Wilens, 2007).
The more likely reasons for such comorbidity are complex, including common genetic risk factors (Fernàndez-Castillo et al., 2010; Franke et al., 2010), altered reward processing (Frodl, 2010; Volkow et al., 2009), comorbid conduct disorder (Serra-Pinheiro et al., 2012; Tuithof, Ten Have, van den Brink, Vollebergh, & de Graaf, 2011), and increased exposure to psychosocial risk factors and self-treatment (Hennessey, Stein, Rosengard, Rose, & Clarke, 2010; Wilens, 2007). The presence of ADHD in SUD patients has a special importance because it often presents with severe forms of SUD characterized by early onset, extended duration of symptoms, greater impairment, higher risk of suicide, and a shorter transition from substance use to dependence (Kelly, Cornelius, & Clark, 2004; Kolpe & Carlson, 2007; Wilens, 2007).
Diagnosis of adult ADHD is based on a careful and systematic assessment of lifetime history of symptoms and impairment. Central to this process is the assessment of childhood onset, current symptoms, and the presence of impairment in at least two domains (school, work, home, interpersonal contacts), according to the Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; American Psychiatric Association, 2000) criteria (S. J. Kooij et al., 2010; Ramos-Quiroga et al., 2006).
Conners’ Adult ADHD Diagnostic Interview for DSM-IV-TR (CAADID) is one of the most frequently used semistructured diagnostic interviews for the assessment of adult ADHD (Arcos-Burgos et al., 2010; Daigre Blanco et al., 2009; Ramos-Quiroga, 2012; Epstein & Kollins, 2006; Medori et al., 2008; Ribasés et al., 2009). The CAADID is useful in clinical and research settings because it is scorable, provides categorical information helpful for defining research groups, and offers a variety of content information that can be used for qualitative analyses (Ramos-Quiroga, 2012). Nevertheless, the CAADID doesn’t systematically evaluate SUD, as it only focuses on ADHD symptoms. In research or clinical settings, other structured interviews (e.g., Structured Clinical Interview for DSM Disorders [SCID-I]) are required to assess SUD.
In the absence of an ADHD-specific measure to validate comorbid psychiatric diagnosis, the adult ADHD section of the psychiatric research interview for substance and mental disorders–IV (PRISM-IV) could serve as a valid alternative. PRISM-IV is a structured interview based on DSM-IV-TR diagnostic criteria and is specifically designed to study comorbidity in substance use subjects. The current version assesses 20 Axis I and 2 Axis II disorders, although other disorders (e.g., ADHD) are in the scope. PRISM showed high test–retest reliability for the majority of dependence, depressive, anxiety, and psychotic disorders, as well as for antisocial and borderline personality disorders (Hasin et al., 2006). In addition, in a validity study diagnoses obtained by PRISM using as “gold standard” the diagnoses obtained through the longitudinal evaluation performed by an expert, using all data available (the longitudinal, expert, all data [LEAD] procedure) showed a good correlation (Kappa index) between diagnoses obtained either by the PRISM or the LEAD procedure in the majority of disorders (Torrens et al., 2004). The concordance for any current affective disorders was fair between PRISM and LEAD (κ = .53), and poor between SCID and LEAD (κ = .36). Concordances for past affective disorders were fair between PRISM–LEAD (κ = .63) and SCID–LEAD pairs (κ = .53) although concordance was better between PRISM and LEAD than between SCID and LEAD in cases of substance-induced depression (κ = .40 vs. κ = .08). For current and past psychotic disorders, the concordance between PRISM and LEAD was excellent to good (κ = .85 and κ = .74), and significantly better than concordance between SCID and LEAD. For panic with or without agoraphobia and any anxiety disorder, the concordance PRISM–LEAD was excellent to good (current κ = .81, κ = .67; past κ = .74, κ = .68). In personality disorders (antisocial personality disorder and borderline personality disorder), concordances PRISM–LEAD were good (κ = .66, κ = .63; Torrens et al., 2004). In the absence of a gold standard to validate the psychiatric diagnosis, we decided to compare PRISM adult ADHD section with the CAADID, which is the most widely used interview for the diagnosis of ADHD in adulthood.
The main objective of this study is to assess the criteria and concurrent validity of the PRISM ADHD section in adulthood. As far as we know, this is the first study aimed to validate the PRISM ADHD section in adults.
Method
A case-control study was performed to check the criteria and concurrent validity of PRISM adult ADHD section.
Participants
Study participants were recruited from the outpatient clinic and from the detoxification unit in the Psychiatry Department of Hospital Universitari Vall d’Hebron in Barcelona, Spain. A convenience sample of 84 participants was selected to participate in the study. Of these, 3 individuals refused to participate and 1 participant was excluded due to intoxication symptoms (cannabis and alcohol) at the moment of the assessment. Of this total sample, 40 participants had a previously established diagnosis of SUD and ADHD and 40 had a diagnosis of SUD with no ADHD; diagnoses where established by a clinical expert (J. Antoni Ramos-Quiroga, Carlos Roncero, Gloria Palomar). Kappa coefficients of agreement were computed between raters and two board-certified psychiatrists who listened to the audiotaped interviews conducted by the raters. The median kappa was 0.98 for ADHD. The average age at assessment was 35.79 years (SD = 11.86) for the 80 participants included in the study.
Inclusion criteria for the ADHD + SUD group were diagnosis of ADHD according to DSM-IV-TR confirmed by CAADID and diagnosis of SUD (DSM-IV-TR criteria) confirmed by SCID-I. Inclusion criteria for the SUD group were as follows: no diagnosis of ADHD according to DSM-IV-TR criteria confirmed by CAADID and diagnosis of SUD (DSM-IV-TR criteria) confirmed by SCID-I. Exclusion criteria for both groups were intelligence quotient below 70 and presence of substance intoxication or abstinence symptoms at the moment of interview. The research protocol was approved by the ethics committee of Vall d’Hebron Research Institute, and all the participants signed an informed consent.
Research Diagnostic Assessments
Research diagnosis was made using the PRISM that was developed to assess psychiatric and substance use comorbidity using DSM-IV-TR criteria. PRISM interview has a validated Spanish version for several psychiatric disorders other than ADHD, but a translation and retro-translation for the ADHD section is available. We used this new Spanish translation of ADHD section of the PRISM.
The diagnosis of ADHD was made using the Spanish validated the CAADID interview version (Ramos-Quiroga et al., 2012). The purpose of CAADID is to determine if the patient meets the first four DSM-IV-TR criteria (Criteria A-D). The first section assesses the presence of the DSM-IV-TR inattention symptoms at childhood and adulthood, followed by questions about onset of the inattention symptoms and pervasiveness of the inattention symptoms. The second section assesses the criteria for the hyperactivity-impulsive symptoms. This is followed by a section that assesses impairment for all symptoms. In the last section of this article, there is a summary sheet and a scoring algorithm of the CAADID to appraise DSM-IV-TR diagnoses. The level of impairment was measured with the Clinical Global Impression Scale, included in the CAADID Part II.
The severity of childhood ADHD symptoms was assessed by Wender Utah Rating Scale (WURS; Ward, Wender, & Reimherr, 1993), which is a self-administered scale with 61 items that have good internal consistency and temporal stability. Severity of ADHD symptoms in the last month previous to the visit was assessed by the ADHD Rating Scale (ADHD-RS; DuPaul, Power, Anastopoulos, & Reid, 1998), which assesses all the ADHD symptoms included at DSM-IV-TR criteria. Full-Scale IQ was estimated with the Vocabulary and Block Design subtests of the Weschler Adult Intelligent Scale–III. The diagnoses of SUD and antisocial personality disorder were performed by SCID-I.
Procedure
Phase 1
ADHD diagnosis was made according to the clinical protocol for the assessment of ADHD in adults that has been approved previously by the ethics committee of Vall d’Hebron Research Institute (more details at Ribasés et al., 2009). The CAADID was administered during the first visit; during the second visit, the WURS and ADHD-RS were applied, and at the third visit, SCID-I was administered to assess SUD and other comorbid disorders. At least 1 hr per visit was required.
Phase 2
An interviewer, blinded to participants’ diagnoses, administered the PRISM Spanish version to all participants.
Statistical Analysis
Criteria validity analysis of the PRISM adult ADHD section was carried out by a study of sensitivity, specificity, positive predictive value (PPV, probability that a patient has the characteristic if the probe is positive) and negative predictive value (NPV, probability that a patient doesn’t have the characteristic if the probe is negative). All these estimations were accompanied with each confidence interval (CI). All information was completed with the Kappa calculus as global measure of concordance. Analysis of the concurrent validity was performed by bivariant correlations. The statistics hypothesis was bivariant, and the confidence level was 95%.
Results
The final sample was composed of 80 participants (76.3% men, mean age = 35.79 years, age range = 18-61); other sociodemographic and clinical characteristics are depicted in Table 1.
Sociodemographic and Clinical Characteristics.
Note. SUD = substance use disorder; WURS = Wender Utah Rating Scale; SD= standard deviation.
After the administration of PRISM, 41 participants (51.3%) were classified as SUD + ADHD and 39 (48.8%) as SUD. Kappa index showed a concordance between the PRISM adult ADHD module and our gold standard criteria used for ADHD diagnosis (CAADID) of 0.78 (95% CI = [0.64-0.91]). The proportion of concordance observed was 0.89 and 0.50 for the proportion expected randomly. Sensitivity of the PRISM adult ADHD module (proportion of participants with ADHD) was 90% (95% CI = [76.9-96.0]). With regard to specificity (proportion of participants without ADHD) was 87.5% (95% CI = [73.9-94.5]). PPV was 87.8% (95% CI = 74.5-94.7), and the NPV was 89.7% (95% CI = [76.4-95.9]).
The adult ADHD symptoms detected by PRISM were statistically significant for both cluster of correlation between WURS ratings and inattention and hyperactivity reported symptoms (inattention symptoms, r = .47, p < .01; hyperactivity symptoms, r = .56, p < .01; and total symptoms, r = .62, p < .01). At the same line, PRISM showed a significant correlation between inattention and hyper-activity symptoms and the results of ADHD-RS (inattention symptoms, r = .69, p < .01; hyperactivity symptoms, r = .80, p < .01; and total symptoms, r = .77, p < .01).
Discussion
This is the first study that validates the PRISM adult ADHD section for the diagnosis of adult ADHD in patients with SUD. The results show that PRISM has good psychometric properties to detect ADHD in adult patients with SUD. The PPV and NPV were superior to 85% as well as the sens-itivity and specificity. On the other hand, the concurrent validity analysis showed that the PRISM ADHD module for adults has a significant correlation with WURS and ADHD-RS.
These results are in line with previous studies that evaluated the validity of PRISM to detect depressive disorders, anxiety disorders, psychotic disorders, or antisocial personality disorder (Astals et al., 2008, 2009; Herrero, Domingo-Salvany, Torrens, Brugal, & ITINERE Investigators, 2008; Rodríguez-Llera et al., 2006; Torrens, Gilchrist, Domingo-Salvany, & psyCoBarcelona Group, 2011). Kappa index showed a concordance between the PRISM adult ADHD module and the CAADID of .78. Previous studies showed a kappa index between the PRISM and the LEAD procedures for affective disorders of .53 to .63, for anxiety disorders of .67 to .68, for psychotic disorders of .85 to .74, and for antisocial personality disorders of .63 to .66. Following DSM-IV-TR criteria, the most widely used interviews for psychiatric diagnosis are SCID-I and PRISM. We found that PRISM has shown better reliability and validity in comparison with the combined use of SCID-I and LEAD as a gold standard in substance abusers (Torrens et al., 2004). In clinical practice is very useful to dispose of a tool that allows to diagnose SUD and other related psychiatric disorders. PRISM should be this tool given its characteristics.
The diagnosis of ADHD in adulthood requires the correct assessment of comorbid disorders, especially SUD because of the high frequency between both disorders. Current diagnostic interviews for adult ADHD do not assess this comorbidity, which makes it mandatory to include another interview (e.g., SCID-I) in the assessment. This situation is time-consuming and limits the use of these instruments in daily clinical practice. PRISM presents a good capacity to detect ADHD and also the most common psychiatric disorders associated with ADHD (i.e., SUD, depression, anxiety, and antisocial personality disorder). Because of this, it is a perfect tool for detection of adult ADHD in patients with SUD. PRISM should be incorporated in the regular clinical practice when the participant has a positive screening of adult ADHD with any other short instrument (e.g., Adult ADHD Self-Report Scale). In conclusion, we believe that because of its good psychometric properties to detect ADHD in adult patients with SUD and ease of use, PRISM should be the tool used in clinical settings to diagnose SUD related to ADHD and other psychiatric diagnoses.
The results of the study should be understood on the light of limitations. The first limitation is that the study was performed in a university hospital setting. As this study was done in clinical population, we cannot make asseverations of the psychometric properties of the adult ADHD module in general population. Nevertheless, the psychometric properties of other modules of the PRISM have good reliability in several settings.
In conclusion, the PRISM adult ADHD section has good psychometric properties to detect ADHD in adults with SUD. PRISM can be a useful tool in clinical settings, for primary settings as well as in other level of attention to assess ADHD in adults with SUD.
Footnotes
Acknowledgements
The authors would like to thank all the staff at Adult ADHD Program and patients who participated in this project. Thanks to Yolanda Santaella and Yemima Villegas for their administrative support.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the 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: Funding for this study was provided by Instituto de Salud Carlos III (ISCIII)-Fondos; Fondo Europeo de Desarrollo Regional l; Red de Trastornos Adictivos; RTA (FEDER; Red de Trastornos Adictivos [RTA] RD06/001/1009 to M.T.). The ISCIII had no further role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the article for publication.
