A higher proportion of people with ADHD also exhibit substance-use disorders (SUD) compared with the general population, although diagnosis of ADHD in adults with SUD is complicated because of overlapping symptoms with other syndromes. A frequently used screening tool for ADHD in adults is the World Health Organization Adult ADHD Self Report Scale (ASRS) screener, a 6-item questionnaire that is scored on a 0–4 scale with some questions given more weight than others. The scale can be scored using a categorical rating of 0–6 positive items or a dimensional 0–24 scoring system. The ASRS screener has recently been revised in line with the Diagnostic and Statistical Manual of Mental Disorders – 5th Edition (DSM-5TM) guidelines for diagnosis of adults with ADHD; the number of criteria required for diagnosis of ADHD has been decreased and the cut-off age for the presence of initial symptomatology has been increased. The use of the ASRS screener to detect comorbidity of ADHD has been studied in patients with addiction, and although some studies indicated acceptable validity of the ASRS in this population, other studies have published less-favourable results.
This study compared the validity of the ASRS* screener based upon DSM-IV (ASRS-IV) or DSM-5TM (ASRS-5) criteria to identify patients with ADHD in a dually-diagnosed (SUD and psychiatric) correctional patient population. Patients (N=140) with a Clinical Global Impressions-Severity (CGI) score of ≥3, indicating at least a mild level of ADHD symptomatology, were included in the study. Patients were randomised according to whether they completed the ASRS-IV or ASRS-5 first, and according to whether they completed the ASRS screeners before or after a psychiatric assessment.
All of the patients enrolled in the study had an SUD, and half (n=70; mean age=33.5±8.3 years; 45 males) had a diagnosis of ADHD and half (n=70; mean age=36.9±9.1 years; 48 males) did not. Patients with ADHD and those without ADHD, respectively, exhibited the following comorbid psychiatric conditions: adjustment disorder (n=5 vs n=16); anxiety disorder (n=22 vs n=21); bipolar disorder (n=16 vs n=11); depressive disorder (n=22 vs n=19); eating disorder (n=1 vs n=1); intermittent explosive disorder (n=1 vs n=4); obsessive compulsive disorder (n=2 vs n=0); post-traumatic stress disorder (n=17 vs n=9); and psychotic disorder (n=1 vs n=2). Of the 70 patients with ADHD, 34 had received pharmacotherapy for ADHD in childhood compared with 2 patients without ADHD. Patients with ADHD had an increased incidence of prior inpatient rehabilitation and psychiatric treatment as well as history of abuse/trauma and prior suicide attempts. A comparison of the ASRS-IV and ASRS-5 showed that both screeners performed equally, without significant difference in sensitivity, specificity, or positive and negative predictive values, regardless of whether a categorical or dimensional scoring system was used (p-value not recorded). Only the dimensional scoring system with a cut-off score of 12/24 had an acceptable sensitivity and negative predictive value, i.e. >80%; sensitivity was 83% and 81% for the ASRS-IV and ASRS-5, respectively, and negative predictive value was 82% and 79%, respectively, which are comparable with results from other dually-diagnosed populations. These data indicate that both the ASRS-IV and ASRS-5 performed equally well in this dually-diagnosed correctional patient population.
As put forward by the authors, a limitation of this study may be the fact that diagnoses of ADHD were made by clinical interview which, while potentially enhancing the ecological validity, may have decreased the internal validity of the study. In addition, as the study included a dually-diagnosed correctional patient population, this may have limited the generalisability of the results to other less-complicated practices.
The authors concluded that both the ASRS-IV and ASRS-5 screeners performed equally well in identifying patients with ADHD in this dually-diagnosed correctional patient population. However, the only satisfactory result was obtained using the dimensional scoring system with a cut-off score of 12, providing a sensitivity and negative predictive value above 80%, which is a lower cut-off than that recommended in community and clinic sample populations. The authors suggested that the ASRS-5 cut-off may need to be adjusted according to the circumstances in which it is used.
*Positive diagnosis for ADHD was classified as an unweighted categorical score of 3/6 (ASRS-IV) or 4/6 (ASRS-5) or an unweighted dimensional score of 12/24 (ASRS-IV) or 14/24 (ASRS-5)
Bastiaens L, Galus J. Comparison of the Adult ADHD Self Report Scale screener for DSM-IV and DSM-5TM in a dually diagnosed correctional population. Psychiatr Q 2017; Epub ahead of print.