Cookie Preferences
By clicking, you agree to store cookies on your device to enhance navigation, analyze usage, and support marketing. More Info
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
January 2, 2025
The Background:
Caffeine and its primary breakdown compounds – paraxanthine, theophylline, and theobromine (also the primary alkaloid in chocolate) – readily cross the placenta and tend to linger due to the underdevelopment of fetal caffeine metabolizing enzymes.
Could accumulating caffeine metabolites harm the developing fetal brain? Studies to date have come to contradictory conclusions, some finding an association, others finding none.
Norway has a national single-payer health insurance system with a national health registry comprehensively tracking virtually all residents.
The Methods:
An international study team used Norways’s registry, and its Mother, Father, and Child Cohort Study (MoBa) to explore the relationship between maternal and paternal coffee consumption and offspring neurodevelopmental difficulties (ND), including symptoms of ADHD as determined from validated rating scales, using 58,964 mother-child pairs and 22,576 father-child pairs, covering the decade 1999-2008.
Unadjusted results indicated a strong association between maternal coffee consumption before and during pregnancy with offspring ADHD, but none with paternal coffee consumption. Superficially, that might seem like compelling evidence.
Yet after adjusting for education, income, smoking, and alcohol, “previously significant effects attenuated towards zero and some associations switched from being positive to negative.”
Then, after further adjusting for genetic confounding, ADHD symptoms at 1.5 years, 3 years, 5 years (on both of two different rating scales), and 8 years (including separate measures of inattention and hyperactivity symptoms) all became indistinguishable between offspring from mothers consuming coffee during pregnancy and offspring of mothers abstaining from coffee during pregnancy.
Conclusion:
The team concluded, “This study applied several conventional and genetic epidemiological approaches to investigate the potential relationship between maternal coffee consumption during pregnancy and offspring NDs [neurodevelopmental difficulties]. When considering the results of the conventional and genetic epidemiological analyses, as well as the broader literature, we conclude that there is little evidence that maternal coffee consumption during pregnancy is strongly causally related to offspring NDs.”
More specifically, the team found no evidence of any causal relationship with offspring ADHD.
Shannon D’Urso, Robyn E Wootton, Helga Ask, Caroline Brito Nunes, Ole A Andreassen, Liang-Dar Hwang, Gunn-Helen Moen, David M Evans, and Alexandra Havdahl, “Mendelian randomization analysis of maternal coffee consumption during pregnancy on offspring neurodevelopmental difficulties in the Norwegian Mother, Father and Child Cohort Study (MoBa),” Psychological Medicine (2024), Vol. 54, Issue 12, 3548-3561, https://doi.org/10.1017/S0033291724002216.
Antidepressants are the primary drug treatment for depressive disorders, which affect 15–20% of pregnant women. They are among the most widely prescribed medications worldwide, and their use has increased in recent decades. Understanding their reproductive safety is critical to support informed, evidence-based prescribing during pregnancy.
A new meta-analysis sheds important light on one of the most debated concerns: whether children born to mothers who took antidepressants during pregnancy face a higher risk of ADHD.
The Study:
Pooling 14 studies covering more than 14 million participants, the analysis found that prenatal antidepressant exposure was associated with a 35% higher rate of ADHD in offspring compared to no exposure. A separate look at SSRIs (the most widely prescribed class of antidepressants, including Prozac and Zoloft) across 11 studies and over four million pregnancies found an even higher apparent risk (44%) after correcting for publication bias. On the surface, these are striking numbers.
Both associations came with an important caveat: enormous variation between individual studies, a statistical red flag suggesting the results may not reflect a true underlying effect. More tellingly, the apparent risk evaporated entirely when researchers applied a more rigorous method — comparing siblings within the same family, where one child was exposed to antidepressants in the womb, and another was not.
This sibling-comparison design is particularly powerful because it automatically controls for factors that run in families: shared genes, household environment, parenting, and socioeconomic conditions. When those influences are held constant, the link between antidepressant exposure and ADHD disappears. The same pattern held for SSRIs specifically.
Two other antidepressant classes, SNRIs (serotonin norepinephrine reuptake inhibitors) and tricyclics, showed no significant association in any analysis.
“Confounding by Indication”:
The probable driver of the initial association is what researchers call confounding by indication. The very condition being treated (depression) is itself a risk factor for ADHD in offspring, independently of any medication. Mothers with more severe depression are also more likely to be prescribed antidepressants, meaning the drug and the underlying illness are difficult to disentangle in standard analyses. Sibling studies cut through this problem cleanly.
The Take-Away:
The authors concluded that the association between antidepressants and ADHD risk was non-significant across all analyses designed to account for these confounding factors. This doesn’t mean antidepressants are without any reproductive considerations, but it does suggest that ADHD risk, at least, is driven by heritable and family-level factors rather than medication exposure itself.
For clinicians and patients weighing the risks of treating or not treating depression during pregnancy, this distinction matters considerably.
Executive functions are the mental processes that allow us to plan, adapt, and follow through. This encompasses working memory, inhibitory control, cognitive flexibility, goal-directed planning, and problem-solving. In people with ADHD, weaknesses in these areas compound the disorder's core symptoms, making it substantially harder to manage complex, real-world demands.
Background:
Medication remains the frontline clinical response. Stimulant medications can meaningfully reduce both executive function deficits and ADHD symptoms, and are often combined with behavioral or psychological therapies for better overall outcomes.
Medication, however, is not entirely without risk of side effects. These risks have spurred interest in new, non-pharmacological alternatives that target the same neural pathways. One of these new therapies is Computerized Cognitive Remediation Therapy (CCRT). This therapy uses digital programs delivered via computer, tablet, or smartphone that train attention, memory, and inhibitory control through structured cognitive exercises. A key feature of many CCRT platforms is adaptive difficulty: tasks adjust in real time to match the child’s current ability, keeping training both challenging and engaging.
The Study:
Despite this promise, the evidence base in younger populations has been limited. This meta-analysis pooled results from randomized controlled trials enrolling participants under 18 who either carried an ADHD diagnosis or scored above the threshold on a validated rating scale. Comparators included no treatment (waitlist), placebo (pharmacological or psychological), or treatment as usual. The primary outcomes (overall executive function and clinical symptom severity) were assessed via questionnaires and neuropsychological testing. Studies including participants with comorbid autism, tic disorders, epilepsy, or other psychiatric conditions were excluded.
The findings were informative, but overall results were mixed. CCRT produced a small but statistically meaningful reduction in inattention symptoms across 13 studies (885 participants), with consistent results across individual trials and no evidence of publication bias. However, it had no detectable effect on hyperactivity and impulsivity (12 studies, 833 participants) or on total ADHD symptom burden (10 studies, 731 participants).
The picture was more encouraging for executive function. Nine studies (500 participants) showed small overall improvements, with specific gains in working memory (454 participants), inhibitory control (428 participants), and planning (6 studies, 335 participants). Emotional control showed no significant change (5 studies, 265 participants), nor did cognitive flexibility (4 studies, 189 participants).
The Take-Away:
Taken together, these results are modest rather than transformative, but context matters. CCRT is low-cost, digitally scalable, and carries negligible side effects. For a population where medication often comes with a significant burden of adverse reactions, even small, reliable improvements in executive function represent a meaningful clinical option.
The evidence positions CCRT not as a replacement for established treatments, but as a practical and well-tolerated addition to the therapeutic toolkit for children and adolescents with ADHD.
Methylphenidate is an effective treatment for ADHD in adults who also have bipolar disorder (BD), but it carries a potential risk of triggering manic episodes. Current guidelines therefore recommend using it only alongside mood-stabilizing medication. A new study using French nationwide claims data sought to test and extend those recommendations with greater statistical power than previous research.
The study built on findings by Viktorin et al. (2017), who observed that adults with BD not taking mood stabilizers had more than a sixfold higher risk of manic events (defined as hospitalization for mania or a new antimanic prescription) within six months of starting methylphenidate. Patients on mood-stabilizing treatment, by contrast, showed nearly half the baseline risk in the first three months. Those findings were limited, however, by small event counts (fewer than 61 manic episodes) and an effect that did not persist beyond the initial three-month window.
To build on this, researchers drew on the French National Health Data System (which is a claims database covering more than 60 million people) spanning 2008 to 2024. The final sample included 6,022 adults with BD (56% women) who had started methylphenidate. Using a self-controlled design, the study compared each patient's rate of manic events in the six months before their first methylphenidate prescription with the rate in the six months after, effectively eliminating stable individual differences as a confound.
Patients were classified as receiving continuous mood-stabilizing treatment if they had been dispensed at least two courses of specific antipsychotics (aripiprazole, olanzapine, or quetiapine) or mood stabilizers (lithium or valproate) in the nine months before starting methylphenidate, including at least one dispensation in the final six months of that window.
The results largely confirmed the earlier findings. Among the 2,745 patients not on mood stabilizers, the rate of inpatient mania diagnosis was 5.1 times higher in the first three months after starting methylphenidate, though this elevation fell to a non-significant level over the subsequent three months. Patients receiving continuous mood-stabilizing treatment showed no statistically significant change in mania risk across the full six-month post-initiation period. A formulation-specific pattern also emerged: patients without mood-stabilizing treatment had a 2.5-fold higher risk associated with extended-release methylphenidate, while no significant risk increase was seen with the immediate-release formulation or in treated patients regardless of formulation.
The authors conclude that methylphenidate at doses below 30 mg does not appear to elevate manic relapse risk when prescribed alongside mood stabilizers. The elevated risk seen in untreated patients, particularly with extended-release formulations, must be interpreted cautiously, given limited statistical power and the likelihood that it partly reflects the natural fluctuation of manic relapse over time. The authors flag this as an inherent limitation of self-controlled survival analyses when studying drug-induced mania, where temporal trends in the underlying condition can be difficult to disentangle from treatment effects.
_logo%201.svg)
