This study was funded by Shire, now part of Takeda
ADHD can negatively affect social and emotional well-being in adulthood and can impact on professional development and financial security. Signs and symptoms of ADHD in adults are often attributed to other psychiatric disorders which can delay diagnosis of ADHD; therefore there is a need for increased knowledge and recognition of ADHD as an adult psychiatric disorder.
This was an observational, cross-sectional, retrospective chart review of adults with a new diagnosis of ADHD enrolled from one of two specialist neuropsychiatric outpatient clinics in Stockholm, Sweden, between 2013 and 2015. Patients* (aged ≥18 years) enrolled in the study underwent an extensive diagnostic† and neuropsychiatric investigation.‡ Several different self-report scales, psychological tests and structured interviews (including the Diagnostic Interview for ADHD in Adults 2.0 [DIVA-2.0] and the Adult ADHD Self-Report Scale [ASRS-v.1.1]) Screener and the 18-item symptom checklist were used to provide a comprehensive neuropsychiatric evaluation and to confirm the diagnosis of ADHD.
Anonymised data from patients with suspected ADHD were gathered and validated by a clinical expert in ADHD for inclusion in this study. Patient data included: social and clinical characteristics; patient and family histories; information relating to prescribed medication; indications of drug/alcohol abuse and depression; and DIVA-2.0 and ASRS-v.1.1 scores. Additionally, full details of any psychiatric diagnoses that had been made using the Mini International Neuropsychiatric Interview (MINI) Swedish version 6.0.0§ were extracted from the databases, and the Montgomery-Åsberg Depression Rating Scale – Self-reported (MADRS-S) screening instrument was used to evaluate the presence and severity of depression. Health-related quality of life (HRQoL) was assessed using the 5-dimesion EuroQol questionnaire (EQ-5D). A linear regression model of EQ-5D index scores, adjusted for age, gender, education and main source of income, was used to identify potential predictors of HRQoL, and subgroup analyses using the same covariates were used to predict difference in EQ-5D index score between subgroups of patients with and without comorbid psychiatric diagnoses. A multiple imputation model was also used to estimate values for missing data using five iterations per value.¥
Of the 189 adult patients (mean age 35.2 years, standard deviation [SD] 12.3; 57% female) enrolled in this study, 60% (n=114) had an ADHD diagnosis of ‘combined type’, 26% (n=49) were identified as having ‘predominantly inattentive type’ and 14% (n=26) had ‘unspecified type’ ADHD. Only 13% of patients had a university education, and less than half of patients listed employment as their main source of income, with 37% in full- or part-time employment and 10% self-employed.
The results of the study were as follows:
- Data extracted from the MINIs showed that 49% (n=92) of patients had a psychiatric comorbidity, with anxiety (34%; n=65) and depression (20%; n=37) the most common. Both anxiety and depression were frequently registered alongside the other MINI-identified diagnoses but were diagnosed as a single comorbidity in only 13% (n=25) and 6% (n=11) of patients, respectively.
- Autism was diagnosed in 16% (n=30) of patients.
- More than half of patients (57%; n=108) had been prescribed pharmacological treatments for psychiatric disorders, including antidepressants, hypnotics, anxiolytics and antipsychotics.
- A total of 77% (n=146) of patients had been recommended pharmacological treatment for ADHD after diagnosis. The most commonly prescribed pharmacological treatments for ADHD were osmotic-release oral system methylphenidate or other methylphenidate formulations (69%; n=130), followed by atoxometine (4%; n=8) and lisdexamfetamine dimesylate (3%; n=5).
- Somatic comorbidities affected 77% (n=145) of patients after enrolment; the most frequently reported conditions were pain, gastrointestinal disorders and allergy, with 37% (n=69) of patients having been prescribed medication for these comorbidities.
- The mean ASRS-v.1.1 symptom checklist score in the study population was 47.4 (SD 13.0), indicating that the majority of patients experienced multiple symptoms of ADHD. The mean MADRS-S and EQ-5D index scores were 19.8 (SD 9.2) and 0.63 (SD 0.28), respectively, indicating that depressive symptoms were common and HRQoL was poor in this study population.
- Subgroup analyses showed that low EQ-5D index scores (indicating poor HRQoL) were associated with patient-rated severe depressive symptoms (MADRS-S score ≥30), an ASRS-v1.1 score ≥60 and the presence of multiple psychiatric comorbidities. Older patients (aged ≥50 years) tended to have a lower EQ-5D index score compared with younger patients.
- In the multivariate regression model, a MADRS-S score ≥30 was the strongest predictor of low EQ-5D index scores (adjusted mean, -0.40; p<0.001 versus scores of 1–11), in addition to having MINI-identified anxiety and depression (p<0.001) or having ≥3 MINI-identified diagnoses (p=0.002) compared with having no MINI-identified diagnoses.
- Adjusted mean EQ-5D index scores were significantly higher in men than women (p=0.049) and in patients with secondary education than those without (p=0.004). Patients in full- or part-time employment also had higher EQ-5D index scores than those who reported other sources of income (p=0.03), whereas adult ASRS-v.1.1. symptom checklist or DIVA-2.0 scores were not significant predictors of EQ-5D index score (p≥0.05).
There were several limitations to this study. Data for some parameters were missing for a substantial proportion of patients, and although a multiple imputation model to estimate missing data was used to address this, this may have introduced bias. For example, ASRS-v.1.1 score data were missing for 88% of patients in one of the outpatient clinics, thereby preventing any reliable conclusions from being drawn regarding the association between ASRS-v.1.1 score and HRQoL in adults with ADHD. Additionally, the covariates used in these models were based on the available data, but it is possible that other factors not captured in this study may influence HRQoL. Furthermore, diagnoses of all comorbid psychiatric disorders were based on the MINI, but autism was assessed as a separate covariate, which may have led to an underestimation of the impact of psychiatric comorbidities on patients’ HRQoL. Finally, a comparative control group was not included in this study and the study population was based solely on a Swedish population from two outpatient clinics in Stockholm, and therefore there results may not be applicable to all study populations.
The authors concluded that these data suggest that adults with ADHD experience reduced HRQoL, and support the theory that ADHD is associated with considerable disease burden. The authors stated that the EQ-5D scores reported in this study are similar to those reported in adults with ADHD in other European countries. Moreover, given that comorbid anxiety and depression, as well as patient-rated depressive symptoms, were the strongest predictors of poor HRQoL in this study, and since these are common comorbid diagnoses with ADHD, the potential negative impact of these disorders, in addition to somatic disorders, should be considered as a means to deliver the optimum care for adults newly diagnosed with ADHD.
*The study population was identified using electronic medical records from the two study sites. Only patients aged ≥18 years who underwent neuropsychiatric investigation and received a new and confirmed diagnosis of ADHD based on the International Statistical Classification of Diseases and Related Health Problems – 10th Revision, Swedish Modification (ICD-10-SE; diagnosis code F90) were included in the study
†Patients underwent a diagnostic interview, accompanied by a family member or companion, during which their medical history, developmental history, presence of coexisting disorders; and current condition were documented. Each patient also underwent a medical examination including blood tests and drug screening
‡The neuropsychiatric investigation recorded the frequency and duration of the patients’ psychiatric symptoms, as well as their functional impairments and cognitive and executive functioning. Routine diagnostic testing for psychiatric disorders was also completed. Diagnoses of autism were made during the neuropsychiatric investigation
§The MINI is a short structured interview in which signs and symptoms are assessed against the Diagnostic Manual of Mental Disorders – 4th Edition (DSM-IV) diagnostic criteria. Depression as identified by the MINI includes depressive episodes and recurrent depression (ICD-10-SE diagnosis codes F32–F33). MINI-identified anxiety includes phobias, other anxiety disorders and obsessive-compulsive disorders (ICD-10-SE diagnosis codes F40–F42)
¥Covariates for the imputation model included: age, gender, study site, education, source of income, MINI-identified comorbid diagnoses, previous psychiatric diagnoses, previous suicide attempts, somatic diagnoses, current psychiatric medications, recommended ADHD medication after neuropsychiatric investigation, DIVA-2.0 scores, ASRS-v.1.1 scores, MADRS-S scores, and both EQ-5D index scores and visual analogue scale scores. Missing EQ-5D index scores were replaced for 27 patients
Ahnemark E, Di Schiena M, Fredman A-C, et al. Health-related quality of life and burden of illness in adults with newly diagnosed attention-deficit/hyperactivity disorder in Sweden. BMC Psychiatry 2018; 18: 223.