It is estimated that 10 to 47% of adults with a diagnosis of schizophrenia also experience ADHD symptoms (Arican et al., 2018), and children with ADHD have an increased risk of developing schizophrenia later in life (Dalsgaard et al., 2020). People with a dual diagnosis of schizophrenia spectrum disorder (SSD) and ADHD are more likely to have difficulties in education and at work (Levy et al., 2015) and experience more severe symptoms, poorer cognitive function and increased social difficulties. Despite this, there is a lack of research on the experience of individuals with SSD who also have ADHD symptoms.
ADHD and schizophrenia can have overlapping symptoms such as difficulties with attention, planning, learning and restlessness or impulsive behaviour. This can make it harder for clinicians to clearly distinguish the two conditions and make accurate diagnoses. Misdiagnosis between ADHD and schizophrenia spectrum disorders (SSD) and under-diagnosis of the dual conditions is a significant concern.
While ADHD medications such as psychostimulants are associated with reduced mortality (Vasiliadis et al., 2024), the picture is less clear for people with SSD and ADHD. Many clinicians are cautious about prescribing psychostimulants because of concerns they could worsen psychotic symptoms in some people.
To address this concern, Luykx et al. (2025) used data from several Swedish health and population registers to study whether ADHD medication use in individuals with schizophrenia improved or worsened their health outcomes over 9 years.
Clinicians are cautious in prescribing ADHD medication in people with psychotic disorders, but is that backed up by evidence of risk?
Methods
Research design
The authors analysed data of 9,416 people with SSD who used ADHD medication. SSD included diagnoses of schizophrenia, schizotypal disorder, delusional disorders, brief psychotic disorder, shared psychotic disorder, schizoaffective disorder and unspecified psychosis according to ICD-10. ADHD medications studied included methylphenidate, amphetamine, dexamphetamine, lisdexamphetamine, atomoxetine and modafinil.
They examined whether individuals who used ADHD medication had reduced hospitalisation or mortality rates and looked in more detail at specific hospitalisations such as for psychosis, somatic conditions (e.g. pneumonia or epilepsy) or cardiovascular illness.
Statistical methods
Analyses were conducted using a within-individual design with Cox regression models comparing periods of ADHD medication use versus non-use within the same individual for all outcomes. Analyses were adjusted for time since cohort entry, temporal order of ADHD medications, and concomitant use of psychotropic drugs (antipsychotics, antidepressants, benzodiazepines, mood stabilisers, drugs for addictive disorders).
They also used between-individual Cox models comparing people using ADHD medication to those not using medication, adjusted for age, sex, disability pension, number of previous hospitalisations for psychosis, diagnosis of ADHD, substance use disorder, previous suicide attempts and previous clozapine use.
They compared whether antipsychotic use during ADHD medication increased or decreased risk of hospitalisation and also studied dosages of the two most used ADHD medications (methylphenidate and lisdexamphetamine) and their effects on outcomes.
The study compared periods of ADHD medication use and non-use within and between people with SSD – and assessed risks of hospitalisation and mortality, after controlling for a number of demographic variables.
Results
All-cause hospitalisation/mortality
Over the 9 years, lisdexamphetamine was associated with reduced risk of being admitted to hospital (for any reason) or death (from any cause) while high-dose methylphenidate was associated with a slight increased risk specifically in those not taking antipsychotics. Moderate-dose methylphenidate was associated with a reduced risk.
Hospitalisation for psychosis and somatic conditions
Atomoxetine was associated with reduced risk for psychosis-related hospitalisation. Lisdexamphetamine and moderate-dose methylphenidate was associated with reduced risk for hospitalisation for psychosis while ADHD polytherapy increased risk of somatic hospitalisation.
Hospitalisation for cardiovascular disease
The study found no significant associations between any ADHD medication exposure and cardiovascular hospitalisation.
Dosage matters
The overall pattern suggested that, in general, moderate dosages of lisdexamphetamine and methylphenidate are generally safer to use in SSDs. On the contrary, very high dosages of methylphenidate (over 95mg/day) carried an increased risk for hospitalisation and mortality, but this was not matched for lisdexamfetamine.
This research suggests that ADHD medication (especially lisdexamphetamine (at any dose) and low to medium doses of long-acting methylphenidate) may be safer for people with schizophrenia than many people think.
Conclusion
- Lisdexamphetamine consistently showed protective effects across doses, showing the strongest evidence for long-term safety in individuals with SSDs.
- Atomoxetine also showed benefit.
- Methylphenidate had a U-shaped risk pattern:
- Moderate doses were beneficial
- Very high doses (≥95 mg/day) were associated with increased risk.
- There was no significant effect for many of the ADHD medications.
The study offers reassuring findings, especially for lisdexamphetamine in patients with SSD, but it does not confirm long-term safety for all ADHD medications. Only lisdexamphetamine use was associated with a reduced risk of both all-cause hopitalisation/mortality and somatic hopitalisations. Some medications show potential risk such as high-dose methylphenidate and ADHD medication polytherapy i.e. taking more than 1 ADHD medication.
As there was no significant effect for many of the ADHD medications, we don’t have enough evidence to say if the medications increase or decrease the risk of hospitalisation.
In people with SSDs taking ADHD medication – lisdexamfetamine showed the best safety profile – whereas methylphenidate appears to warrant caution at high doses.
Strengths and limitations
The study’s long-term follow-up period and large total sample size are key strengths; however it is unclear whether the subset sample sizes for each ADHD medication are large enough to be a high-powered study, particularly as it analyses retrospective data alone.
The study’s definiton of “SSDs” is expansive, including a variety of conditions with some psychotic features – however this can risk obscuring confounding factors in this study that may be particular to a disorder e.g. schizoaffective disorder, where pervasive mood changes and also antidepressant medications could separately affect hospitalisations and mortality risks. Further, the study statistically adjusts for a number of demographic features and also other pharmacological interventions such as previous clozapine use, it does not explicitly analyse the effects of such adjustments being made – these may have unknown confounding modulating effects on the suitability or efficacy of ADHD medication in SSDs, beyond what can be gleaned from the paper.
Another limitation is the broad categorisation of safety outcomes (e.g., “hospitalisation for psychosis”). The study did not explore the effect of ADHD medication on specific SSD-related symptoms such as the severity of psychotic symptoms, particular discrepancies in positive/negative symptoms, functional impairments, or quality of life.
The geographical confinement of this study to Swedish health systems could further be a limitation to generalisability of results.
Hence, although the authors report that they found no evidence of increased risk of adverse outcomes for ADHD medications in individuals with a condition under the schizophrenia spectrum disorders (SSDs), especially when used with antipsychotics; these findings should be interpreted as broadly reassuring, rather than conclusive. Further studies that are prospectively designed and better powered to assess each ADHD medication individually, will be useful to form conclusions on their long-term safety.
The results offer pragmatic reassurance – this is a robust proof-of-concept for further research, prospectively evaluating individual ADHD medications in people with individual SSDs.
Implications for practice
This research suggests that use of the ADHD medications lisdexamphetamine, and long-acting methylphenidate at low to medium doses, appears safer than previously assumed in individuals with SSDs. However, the authors advise caution with methylphenidate: it should generally be avoided in patients with SSDs who are not using an antipsychotic, and high-dose methylphenidate (≥95 mg daily) should be avoided in all patients with SSDs.
For the other ADHD medications examined, amphetamine, dexamphetamine, atomoxetine and modafinil, the study provides insufficient evidence to draw firm conclusions about long-term safety.
Given the lack of clear risk data for many medications, the benefits of effectively treating comorbid ADHD need to be carefully weighed against potential risks. Treatment decisions should be made using a personalised, shared decision-making approach, taking into account the individual’s psychiatric history, current medications, and risk factors, with the aim of improving recovery and functional outcomes.
There may also be a policy impact to this paper – as the availability of lisdexamfetamine in other jurisdictions, in particular Anglophone countries, can be limited by practice guidelines that list it as a second-line medication for ADHD. Providing its comparable safety across doses, relative to methylphenidate that is often prescribed as a ‘first-line’ medication, data from this study should be used to consider whether guideline and policy reform is warranted, at least in cases where a person needing ADHD medication also experiences an SSD.
Growing evidence of safety of ADHD medications in SSDs, especially lisdexamfetamine which is less favoured in some countries – should inform not only individual clinical practice but also reform in prescribing guidelines and policy.
Statement of interests
None to declare.
References
Primary paper
Luykx, J.J., Corbeil, O., Kärkkäinen, O. et al. Long term safety of ADHD medication in patients with schizophrenia spectrum disorders. Mol Psychiatry 30, 4859–4867 (2025). https://doi.org/10.1038/s41380-025-03080-3
Other references
Arican, I., Bass, N., Neelam, K., Wolfe, K., McQuillin, A., Giaroli, G., & others. (2019).
Prevalence of attention deficit hyperactivity disorder symptoms in patients with schizophrenia. Acta Psychiatrica Scandinavica, 139(1), 89–96. https://doi.org/10.1111/acps.12948
Dalsgaard, S., Mortensen, P. B., Frydenberg, M., Maibing, C. M., Nordentoft, M., & Thomsen, P. H. (2020). Association between attention-deficit hyperactivity disorder in childhood and schizophrenia later in adulthood. European Psychiatry, 29(4), 259–263. https://doi.org/10.1016/j.eurpsy.2013.06.004
Gough, A., & Morrison, J. (2016). Managing the comorbidity of schizophrenia and ADHD. Journal of Psychiatry & Neuroscience, 41(5), E79–E80. https://doi.org/10.1503/jpn.150251
Levy E, Traicu A, Iyer S, Malla A, Joober R. Psychotic disorders comorbid with attention-deficit hyperactivity disorder: an important knowledge gap. Can J Psychiatry. 2015 Mar;60(3 Suppl 2):S48-52. PMID: 25886680; PMCID: PMC4418622.
Vasiliadis HM, Lunghi C, Rahme E, Rochette L, Gignac M, Massamba V, Diallo FB, Fansi A, Cortese S, Lesage A. ADHD medications use and risk of mortality and unintentional injuries: a population-based cohort study. Transl Psychiatry. 2024 Feb 28;14(1):128. doi: 10.1038/s41398-024-02825-y. PMID: 38418443; PMCID: PMC10901868.