Adding trazodone showed no clear cognitive advantage over donepezil alone
In a dementia trial, cognitive scores improved in both treatment groups, without a significant difference between them. Tiredness was reported more often with the combination.
Based on the published abstract. The full paper may contain additional methods, results and limitations.
The 30-second takeaway
Adding trazodone did not show an additional cognitive benefit over donepezil alone during this dementia trial. Both groups improved from their starting cognitive scores, but the difference between treatments was not statistically significant. That does not establish equivalence. Overall adverse-reaction comparisons were also inconclusive, while tiredness was reported more often with the combination.
The abstract gives no between-group effect estimates or confidence intervals, so the possible size of an undetected difference is unclear. It calls for larger studies and longer follow-up.
What researchers found
Cognitive test outcomes
Over 6 months, the comparison of donepezil plus trazodone with donepezil alone found no statistically significant efficacy difference; cognition was assessed using the Mini-Mental State Examination and Addenbrook's scale.
Improvement within groups is different
Both groups' cognitive scores rose relative to their starting scores. The question about adding trazodone depends on the comparison between groups, which did not show a statistically significant advantage for the combination during this trial.
Safety findings also need care
The overall comparison of adverse reactions was not statistically significant, although tiredness was more common with combined treatment. The abstract provides no reaction counts, so it cannot establish how much more often tiredness occurred.
The original publication
Efficacy of donepezil versus donepezil and trazodone in combination cognitive dysfunction in patients with dementia: a randomized controlled trial.
Kheirkhah F, Hamzehkolaee ME, Parvizi S et al.
Naunyn Schmiedebergs Arch Pharmacol · 2026
- PubMed ID
- 40586866
- Record checked
AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.
One more question, understood.
Keep track of the research you’ve explored.