The efficacy of probiotics as a therapeutic alternative for attention-deficit hyperactivity disorder (ADHD) remain unclear.

To investigate the effectiveness of probiotics for symptoms of ADHD and identify possible factors affecting their efficacy.

Randomised placebo-controlled trials were identified through searching major databases from inception to April 2023, using the main keywords ‘probiotics’ and ‘ADHD’ without limitation on languages or geographic locations. The outcome of interest included improvement in total symptoms of ADHD, symptoms of inattention and hyperactivity/impulsivity, and drop-out rate. Continuous and categorical data were expressed as effect sizes based on standardised mean differences (SMDs) and odds ratios, respectively, with 95% confidence intervals.

Meta-analysis of seven trials involving 379 participants (mean age 10.37 years, range 4–18 years) showed no significant improvement in total symptoms of ADHD (SMD = 0.25; P = 0.12), symptoms of inattention (SMD = 0.14; P = 0.3) or hyperactivity/impulsivity (SMD = 0.08; P = 0.54) between the probiotic and placebo groups. Despite non-significance on subgroup analyses, there was a large difference in effect size between studies using probiotics as an adjunct to methylphenidate and those using probiotics as supplementation (SMD = 0.84 v. 0.07; P = 0.16), and a moderate difference in effect size between studies using multiple strains of probiotics and those using single-strain regimens (SMD = 0.45 v. 0.03; P = 0.19).

Current evidence shows no significant difference in therapeutic efficacy between probiotics and placebos for treatment of ADHD symptoms. However, albeit statistically non-significant, higher therapeutic efficacies associated with multiple-strain probiotics or combining probiotics with methylphenidate may provide direction for further research.

The risk of antipsychotic-associated cardiovascular and metabolic events may differ among countries, and limited real-world evidence has been available comparing the corresponding risks among children and young adults. We, therefore, evaluated the risks of cardiovascular and metabolic events in children and young adults receiving antipsychotics.

We conducted a multinational self-controlled case series (SCCS) study and included patients aged 6–30 years old who had both exposure to antipsychotics and study outcomes from four nationwide databases of Taiwan (2004–2012), Korea (2010–2016), Hong Kong (2001–2014) and the UK (1997–2016) that covers a total of approximately 100 million individuals. We investigated three antipsychotics exposure windows (i.e., 90 days pre-exposure, 1–30 days, 30–90 days and 90 + days of exposure). The outcomes were cardiovascular events (stroke, ischaemic heart disease and acute myocardial infarction), or metabolic events (hypertension, type 2 diabetes mellitus and dyslipidaemia).

We included a total of 48 515 individuals in the SCCS analysis. We found an increased risk of metabolic events only in the risk window with more than 90-day exposure, with a pooled IRR of 1.29 (95% CI 1.20–1.38). The pooled IRR was 0.98 (0.90–1.06) for 1–30 days and 0.88 (0.76–1.02) for 31–90 days. We found no association in any exposure window for cardiovascular events. The pooled IRR was 1.86 (0.74–4.64) for 1–30 days, 1.35 (0.74–2.47) for 31–90 days and 1.29 (0.98–1.70) for 90 + days.

Long-term exposure to antipsychotics was associated with an increased risk of metabolic events but did not trigger cardiovascular events in children and young adults.

The association between autoimmune diseases and schizophrenia has rarely been systematically investigated.

To investigate the association between schizophrenia and a variety of autoimmune diseases and to explore possible gender variation in any such association.

Taiwan's National Health Insurance Research Database was used to identify 10 811 hospital in-patients with schizophrenia and 108 110 age-matched controls. Univariate and multiple logistic regression analyses were performed, separately, to evaluate the association between autoimmune diseases and schizophrenia. We applied the false discovery rate to correct for multiple testing.

When compared with the control group, the in-patients with schizophrenia had an increased risk of Graves' disease (odds ratio (OR) = 1.32, 95% CI 1.04–1.67), psoriasis (OR = 1.48, 95% CI 1.07–2.04), pernicious anaemia (OR = 1.71, 95% CI 1.04–2.80), celiac disease (OR = 2.43, 95% CI 1.12–5.27) and hypersensitivity vasculitis (OR = 5.00, 95% CI 1.64–15.26), whereas a reverse association with rheumatoid arthritis (OR = 0.52, 95% CI 0.35–0.76) was also observed. Gender-specific variation was found for Sjögren syndrome, hereditary haemolytic anaemia, myasthenia gravis, polymyalgia rheumatica and dermatomyositis.

Schizophrenia was associated with a greater variety of autoimmune diseases than was anticipated. Further investigation is needed to gain a better understanding of the aetiology of schizophrenia and autoimmune diseases.