This meta-analysis pooled 23 controlled studies of adults with a formal ADHD diagnosis. The studies were published from 2002 to 2021 and included 3,835 people.
Twelve studies tested medicines. Eleven tested psychosocial approaches, such as cognitive behavioral therapy (CBT), mindfulness and cognitive training. Each compared a treatment group with a control group and measured something tied to work, such as time management, organization or job functioning.
Overall, both kinds of treatment helped more than the control conditions, but the gain was small. The psychosocial treatments showed a moderate effect that was clearly larger than the effect of medicines. Among medicines, only methylphenidate showed a firm effect, and it was small. Atomoxetine, vortioxetine, modafinil and guanfacine did not show a reliable effect. Among psychosocial treatments, CBT showed the firmest effect. Mindfulness looked large, but its studies disagreed and the result could have been down to chance.
The authors flag limits. Only six of the 23 studies were rated low risk of bias. A check suggested seven studies may be missing, and adjusting for them shrank the overall effect. Nine of the work measures relied on people rating themselves. Medicine and therapy studies also used different control groups.
The authors conclude that methylphenidate should be combined with a psychosocial treatment such as CBT for adults with ADHD, though the study did not test that combination.
Talk to a clinician about your own situation.
The numbers Stats and what the terms mean
The statistics behind the plain-language claims above.
- Lauder, 2024: overall effect of all treatments versus control on work-relevant outcomes, d = .30 (p < .001, 95% CI .20 to .40), heterogeneous (I2 of 57.73%).
- Lauder, 2024: medicines versus control, d = .19 (95% CI [.10 to .29], p < .001, I2 = 49.71%).
- Lauder, 2024: methylphenidate versus control, d = .31 (95% CI [.21 to .42], p < .001, I2 = 0%, small number of studies).
- Lauder, 2024: the other medicines versus control, Atomoxetine d = .13, Vortioxetine d = .01, Modafinil d = .07 and Guanfacine (single study) d = .23, all nonsignificant.
- Lauder, 2024: psychosocial treatments versus control, d = .56 (95% CI [.40 to .73], p < .001).
- Lauder, 2024: psychosocial pooled effect compared with the medicine pooled effect, Qw(1) = 14.43, p < .001.
- Lauder, 2024: CBT versus control, d = .65 (95% CI [.46 to .83], p < .001, I2 = 0%).
- Lauder, 2024: mindfulness versus control, d = .79, nonsignificant (p = .07, I2 = 78.66%).
- Lauder, 2024: overall effect after adjusting for seven possibly missing studies (trim-and-fill), from d = .30 to d = .21 (95% CI [.11 to .32]).
- Lauder, 2024: overall effect when each study was left out in turn (leave-one-out), d = .28 to d = .32. What the terms mean:
- d An effect size: a number that shows how big a difference is, separate from whether it was statistically significant. Roughly, 0.2 is small, 0.5 is moderate and 0.8 is large. A minus sign shows the direction of the change, not its size.
- 95% CI A confidence interval: a range of values around the result that fit the data reasonably well, where a narrow range means a more precise result and a wide one a less precise result.
- p A p-value: a number that shows how unusual a result would be if there were really no effect, and it depends on the study's own assumptions being right.
- Nonsignificant The result's p-value did not fall below the cutoff set in advance, most often 0.05, which is a statistical label rather than a judgment about importance.
- I2 How much the pooled studies disagreed with each other, from 0% (no disagreement) upward.
- Qw(1) A test of whether two pooled groups of studies differ, used here to compare psychosocial treatments with medicines.
- Trim-and-fill A method that estimates how many studies may be missing from the published record and adjusts the pooled result for them.
- Leave-one-out A check that reruns the pooled result with each study removed in turn, to see whether any single study drives it.
Source: Lauder K, McDowall A, Tenenbaum H. A meta-analysis of pharmacological and psychosocial interventions aiming to improve work-relevant outcomes for adults with ADHD. Neurodiversity (2024). doi:10.1177/27546330241292984