At least one third (1/3) of patients with major depressive disorder (MDD) would not qualify for inclusion in many randomized controlled trials (RCTs) that help guide the use of antidepressants, new data showed.
The analysis highlights “a major gap” in the applicability of research findings to the real-world setting, said lead author Aleksi Hamina, PhD, Pharmacoepidemiologist at Niuvanniemi Hospital in Kuopio, Finland. “We argue that current trials have limitations for capturing the full spectrum of people with MDD in routine clinical practice,” he said. “More inclusive trial designs and integration of real-world evidence are needed to inform treatment guidelines that are both evidence-based and representative.”
The findings were presented on 11 OCT 2025 at the 38th European College of Neuropsychopharmacology Congress (ECNP) 2025.
Findings Highlight ‘Major Gap’
Using nationwide registry data from Finland (2004 – 2018) and Sweden (2007 – 2021), the researchers identified 73,720 Finnish and 135,092 Swedish adults aged 18 – 65 years with nonpsychotic MDD who were stabilized on antidepressant monotherapy for at least 105 days. The primary outcome was a composite of all-cause mortality and hospitalization due to psychiatric reasons or suicide attempt, followed up for 6 months.
They then compiled a list of typical exclusion criteria based on an analysis of studies on antidepressant maintenance therapy, reaching a consensus on the following: substance use disorder (SUD), intellectual disability, other psychiatric disorders, use of electroconvulsive therapy or other neuromodulation, pregnancy or breastfeeding, prior suicide attempt, or comorbid serious somatic disease (narrow and broad definition).
In the Finnish cohort, 33.5% of individuals met at least one criterion, as did 35% of those in the Swedish cohort. If the broader definition of serious somatic disease was used, more than half of each cohort would be ineligible for RCT inclusion.
RCT-ineligible individuals had a significantly higher risk for the primary outcome in the cohorts from Finland and Sweden (hazard ratios [HRs], 2.44 and 2.61, respectively). Those with a history of suicide (HRs, 5.76 and 5.20, respectively) or SUD (HRs, 5.05 and 5.13, respectively) had the greatest risk, Hamina said. “Our findings highlight a major gap between selected trial populations and those seen in routine clinical practice,” Hamina and colleagues wrote in the study abstract. “Individuals routinely excluded from antidepressant RCTs experience substantially worse outcomes. More inclusive trial designs and integration of real-world evidence are needed to inform treatment guidelines that are both evidence-based and representative,” they added.
The researchers said the findings are in press in World Psychiatry.
RCTs Not Representative?
Although the researcher has not yet been peer reviewed, the study “appears to indicate what most of those in the field have been aware of for a number of years: that people taking part in RCTs for MDD may not be representative of the people we look after in clinical services,” Sameer Jauhar, MBChB, PhD, Clinical Associate Professor in Affective Disorders and Psychosis, Imperial College London, London, England, told Medscape Medical News. “Most trials will exclude people with comorbidities, such as substance misuse, and this study appears to show this, and more importantly, it demonstrates that outcomes are worse for this patient group,” he added.
However, ECNP President Martien Kas, PhD, Professor of Behavioral Neuroscience, University of Groningen, Groningen, Netherlands, pointed out that widening inclusion criteria can sometimes dilute results. “A clinical trial is designed in such a way to address a certain question. The study design should be guided by the question,” he told Medscape Medical News. “So you would imagine that if you were more rigid in your exclusion criteria you could really stratify your population to get as homogenous a population as possible,” Kas added. “On the other hand, if you want to see a robust effect of your treatment you would like a more heterogenous population.”
REFERENCE: Medscape; 14 OCT 2025; Kate Johnson