Introduction: Pharmacotherapy of attention-deficit hyperactivity disorder (ADHD) in adults has increased internationally. The global prevalence of persistent adult ADHD is estimated to be 2.6% (95% CI 1.5-4.5). While unexplained geographic variation in paediatric ADHD treatment rates has been reported in Australia, this has not been investigated in adults.

Aims: Describe ADHD medicine use among adults at the person level, stratified by geographic area.

Methods: We accessed linked Pharmaceutical Benefit Scheme (PBS) and demographic data via the Person Level Integrated Data Asset (PLIDA) using the Australian Bureau of Statistics (ABS) DataLab for all Australian residents during the years July 2019-June 2020 and July 2023-June 2024. We quantified annual prevalence of ADHD medicine dispensing by Statistical Area, Level 3 (SA3) of residence, calculating age-standardised rates for people aged 20-44 years using ABS estimated residential population counts as denominators.

Results: Among people aged 20-44 years, the overall national prevalence increased from 0.7% in 2019-20 to 2.4% in 2023-24. In 2023-24, treatment prevalence varied markedly by location, ranging from 5.8% for people aged 20-44 years in Fremantle (Western Australia) to 0.5% in Fairfield (New South Wales). Small-area mapping shows striking variation by: State, with 13 of the top 20 SA3 in Western Australia; rurality, with typically higher rates in major cities and across the eastern seaboard; and socioeconomic status (SES), with large disparities within major cities e.g. 10-fold difference between SA3s within inner and south-west Sydney.

Conclusions: Treatment rates in some areas exceeded the estimated international prevalence of ADHD in adults, suggesting either unexpected geographic variation in disease prevalence or overdiagnosis. Low rates in other areas, particularly remote regions and areas of low SES, suggest financial and health system barriers may be preventing people from accessing the care they need.

Keywords: Attention-deficit hyperactivity disorder, psychostimulants, drug utilisation, clinical variation, health equity, record linkage

Introduction: Previous studies have reported that treatment is associated with favorable clinical outcomes among individuals with attention-deficit/hyperactivity disorder (ADHD), including reduced psychiatric hospitalization. However, individuals often present with varying degrees of pre-treatment psychiatric burden, and whether this influences treatment benefit remains unclear.

Aims: To evaluate the association between ADHD stimulant treatment and psychiatric hospitalization across levels of pre-treatment psychiatric burden.

Methods: Using the Korean nationwide health insurance database, individuals aged 6–64 years with a new ADHD diagnosis between January 2013 and April 2023 were identified. A target trial emulation framework was applied, defining the initiation arm as those who commenced methylphenidate within three months of diagnosis. The cloning-censoring-weighting method was used to address confounding, with treatment effects estimated using weighted pooled logistic regression. The primary outcome was psychiatric hospitalization within two years after ADHD diagnosis (time zero). Pre-treatment psychiatric burden was assessed across four domains including psychiatric comorbidities, psychotropic medication use, health care utilization, and prolonged mental health service use. Burden was classified as high if two or more domains were present, and low otherwise, with analyses stratified accordingly.

Results: A total of 436,702 individuals with ADHD were included (159,324 females [36.5%]; mean age, 17.4 years [SD, 11.7]). Of these, 302,100 (69.2%) were assigned to the initiation arm. Methylphenidate initiation was associated with a lower risk of psychiatric hospitalization (adjusted incidence rate ratio [aIRR], 0.72; 95% CI, 0.69–0.77). The association varied according to pre-treatment psychiatric burden (Pinteraction <0.01), with aIRRs of 0.73 (95% CI, 0.64–0.80) among individuals with low burden and 0.68 (95% CI, 0.64–0.73) among those with high burden.

Conclusions: Methylphenidate initiation was associated with a lower risk of psychiatric hospitalization, with treatment benefit differing by the pre-treatment psychiatric burden. Further investigation into specific burden components is warranted.

Keywords: Methylphenidate; Psychiatric Hospitalization; Treatment Effect Heterogeneity

Introduction: Attention-deficit/hyperactivity disorder (ADHD) is a frequent comorbidity of epilepsy, and patients with both may receive concomitant methylphenidate (MPH) and antiseizure medications (ASMs). Although both MPH and ASMs have been independently associated with movement disorders (MDs), evidence on potential drug-drug interactions affecting MD risk remains limited.

Aims: To identify MPH-ASM drug-drug interaction (DDI) signals for MDs and assess age patterns and robustness to concomitant medications.

Methods: This disproportionality study used nervous-system individual case safety reports from VigiBase (2015-2025). Reports co-reporting MPH and ASM with movement-disorder Preferred Terms under five MedDRA High Level Terms (choreiform movements, dyskinesias and movement disorders, dystonias, Parkinson's disease and parkinsonism, and tremor) were identified. For each MPH-ASM pair with at least 5 co-reported MD cases, the Omega shrinkage (Ω) measure tested whether co-reporting exceeded each drug's individual expectation (Ω0.025>0). Interaction signals were also assessed in a pediatric subgroup (<18 years). Because antipsychotics independently cause MDs and are commonly co-prescribed, a sensitivity analysis excluded antipsychotic-exposed reports.

Results: Among 5,116,183 nervous-system reports, 9,093 involved MPH-ASM co-exposure, of which 486 reported movement disorders. Thirteen MPH-ASM pairs met the ≥5-case threshold, of which five reached the signal criterion: oxcarbazepine (n=38, Ω0.025=0.555), phenobarbital (n=7, 0.359), carbamazepine (n=26, 0.294), valproate (n=109, 0.222), and lamotrigine (n=105, 0.075). In the pediatric subgroup, only the MPH-valproate signal remained (n=43, Ω0.025=1.393). In the sensitivity analysis excluding antipsychotics, all signals disappeared, suggesting that the observed signals may be attributable to antipsychotic co-medication rather than direct MPH-ASM interactions.

Conclusions: This disproportionality analysis detected movement-disorder signals for several MPH-ASM combinations, most notably MPH-valproate in pediatric reports. However, these signals were not robust to co-reported antipsychotics. These hypothesis-generating findings require confirmation in pharmacoepidemiologic studies that can account for concomitant medications.

Keywords: drug-drug interaction; attention-deficit/hyperactivity disorder; antiseizure medication

Introduction: Benzodiazepines (BZDs) and Z-drugs are increasingly prescribed to adolescents and young adults (AYAs) in Hong Kong, despite limited guidance. Given the burden of sleep disturbance during this developmental period, comparative safety data are needed.

Aim: To compare adverse clinical outcomes after initiation of BZDs versus Z-drugs among AYAs.

Methods: Electronic health records from all public hospitals in Hong Kong were used to emulate a target trial with an active-comparator, new-user design. Individuals aged 12–25 years who initiated a BZD or Z-drug in 2009-2023 were followed from the index date until censoring (drug switch, discontinuation, epilepsy diagnosis). Outcomes were self-harm, overdose, substance use disorder (SUD), and long-term drug use, defined as continuous prescribing for ≥180 days. Individuals with a history of the outcome were excluded from the corresponding analysis. We applied inverse probability-of-censoring weights using baseline characteristics (age, sex, baseline psychiatric diagnosis) and time-varying covariates (diagnosis of depression/anxiety/psychosis and use of other psychotropics). Pooled logistic regression was used to estimate absolute risks and risk ratios (RRs) at 6 months, with 95% confidence intervals. Subgroup analysis was conducted by sex.

Results: Among 25,979 eligible AYAs, analyses included 23,940 for self-harm, 25,215 for SUD, 24,582 for overdose, and 25,819 for long-term drug use. Median follow-up was 5 and 4 months for BZD and Z-drug users. Compared with Z-drugs, BZD use was associated with higher risks of self-harm (RR 1.20, 1.02–1.49), overdose (1.52, 1.22–1.99), and long-term drug use (1.24, 1.08–1.42), but not SUD (1.04, 0.81–1.39). Similar associations were observed among females; among males, BZD use was associated only with higher long-term drug use risk (1.29, 1.05-1.59).

Conclusion: When pharmacological treatment is needed, Z-drugs may have a comparatively safer profile than BZDs in AYAs. Careful risk assessment is warranted, particularly when prescribing BZDs to females.

Keywords: Comparative safety, benzodiazepines and Z-drugs, young people

Introduction: Adult attention-deficit/hyperactivity disorder (ADHD) pharmacotherapy has expanded rapidly with growing recognition of ADHD beyond childhood. These medications are associated with adverse effects including insomnia that may trigger prescribing cascades. While cascades have been studied in elderly populations, evidence among the growing adult ADHD population remains limited.

Aims: To detect prescribing cascade signals following ADHD medication initiation among adults across multiple adverse effect domains using nationwide claims data.

Methods: A prescription sequence symmetry analysis was conducted using the Korean Health Insurance Review and Assessment Service database, covering approximately 98% of the national population (January 2015 to August 2024). Adult (aged 19 years and older) patients diagnosed with ADHD who initiated methylphenidate, atomoxetine, or clonidine were identified as incident users using a 365-day prescription-free washout period. New prescriptions of five marker drug groups were assessed within a 365-day symmetric interval: sedative-hypnotics, prokinetics, proton pump inhibitors, selective serotonin reuptake inhibitors, and second-generation antipsychotics, with an antibiotic (amoxicillin/clavulanate) as negative control. Adjusted sequence ratios and 95% confidence intervals were calculated using null-effect sequence ratios, with signals identified when the lower confidence limit exceeded 1.0.

Results: Among 257,276 adult incident ADHD medication users, annual use increased from 8,064 in 2015 to 54,858 in 2023. A prescribing cascade signal was detected for atomoxetine followed by sedative-hypnotics (adjusted sequence ratio 1.49; 95% confidence interval 1.41 to 1.57). No signals were detected for methylphenidate or clonidine followed by sedative-hypnotics. No signals were observed for the remaining marker drug groups with any index drug. The negative control demonstrated no sequence asymmetry, supporting analytical validity.

Conclusions: A sleep-related prescribing cascade signal was identified specifically for atomoxetine among Korean adults, while no signals were detected for methylphenidate or clonidine. As adult ADHD pharmacotherapy continues to expand, targeted monitoring for atomoxetine-related sleep disturbances and subsequent prescribing cascades is warranted.

Introduction: Ethnic and socioeconomic differences in ADHD diagnosis and treatment remain poorly characterised, despite substantial increases in diagnosis and medication prescribing in England.

Aims: To describe ethnic and socio-economic patterns in recorded ADHD diagnosis, medication initiation and discontinuation in English primary care (2008-2025).

Methods: Using IQVIA Medical Research Data (IMRD), we conducted a retrospective cohort study of individuals aged 3–99 years. Separate cohorts were constructed for diagnosis, initiation and discontinuation. Diagnosis analyses excluded prior recorded ADHD diagnosis, initiation analyses excluded prior ADHD medication use. Discontinuation was assessed among incident medication initiators and confirmed following ≥180 days without subsequent medication coverage. Poisson regression, adjusted for age, sex and calendar year, estimated adjusted incidence rate ratios (aIRRs) across ethnicity and Index of Multiple Deprivation (IMD) quintiles. Age- and sex-standardised annual rates were estimated using direct standardisation. Time to discontinuation was analysed using Cox regression.

Results: The diagnosis, initiation and discontinuation cohorts included 3,832,111, 3,840,470 and 8,833 individuals, respectively. Of the 8,833 individuals in the discontinuation cohort, 7,630 contributed to the primary analysis. Adjusted diagnosis and initiation rates were lower in every ethnic group than in White British patients. The largest differences were observed in South Asian (aIRR 0.20, 95% CI 0.17–0.24), East Asian (0.26, 0.21–0.31) and Black African patients (0.29, 0.26–0.33). Medication initiation followed a similar pattern, with aIRRs of 0.16 (0.13–0.20), 0.27 (0.20–0.36) and 0.21 (0.18–0.24), respectively. Adjusted diagnosis and initiation rates were similar across IMD quintiles overall, although recent increases in initiation were greater in less deprived groups, suggesting widening socioeconomic differences. Discontinuation showed no clear socioeconomic differences, while ethnic estimates were imprecise.

Conclusions: Marked ethnic differences persisted in recorded ADHD diagnosis and medication initiation despite recent increases in rates. Further research is needed to understand the underlying drivers of these differences.

Keywords: ADHD, ethnicity, socioeconomic deprivation, primary care