Increasing Antidepressant Dose Does Not Improve Outcomes

A systematic review of literature and meta-analysis indicates that there is no clinically or statistically significant effect of antidepressant dose increase after non-response to initial treatment.


A team of researchers in Germany investigated the common clinical practice of increasing the antidepressant dose when patients do not respond to the initial treatment. The results of their systematic review of the literature, recently published in the Journal of Clinical Psychiatry, found no support for this practice.

“Despite its frequent use in clinical practice, we found no sufficient evidence for the efficacy of antidepressant dose increase after failure of an antidepressant treatment in patients with major depressive disorder,” the authors write.

Photo Credit: Pixabay

Rates of nonresponse to antidepressant drugs are reported to occur in as many as 30-40% of patients. Current guidelines recommend “antidepressant switch, augmentation with a second-generation antipsychotic or lithium, a combination of antidepressants, and a dose increase of the initially prescribed antidepressant,” when initial treatment fails.

Current studies demonstrate that increasing the dose of the initially prescribed antidepressant is one of the most popular strategies. Yet, clinicians need evidenced-based recommendations. The current research regarding the decision to increase the dose is varied and inconclusive.

In response, Rink and colleagues conducted a systematic review of the literature and meta-analysis on the dose increase of antidepressants strategy when “antidepressant treatment failure” occurs. They included studies that met the following criteria: (1) RCTs with randomization of patients with “antidepressant treatment failure” to either a dose increase regimen or unchanged continuation, (2) patients diagnosed with unipolar depression, (3) standardized diagnosis of “antidepressant treatment failure.”

Their literature search results retrieved 1,780 articles after the removal of duplicates. Nine studies met the inclusion criteria and featured a total of 1,273 patients (635 women). About half of the participants (628) received dose increase when they experienced failure to respond to antidepressants.

In the analyses, the researchers examined the “efficacy of dose increase compared with continuation of initially prescribed dose of antidepressants, expressed as the standardized mean difference (SMD).” Additionally, they looked at the response, remission, and dropout rates, assessed publication bias, and conducted a post-hoc analysis that focused on different definitions of “antidepressant treatment failure.” Their results did not support the dose increase of SSRIs when patients are nonresponsive. The researchers write:

“This systematic literature review and meta-analysis yielded 2 main results, as follows. (1) There is no clinically or statistically significant effect of SSRI dose increase after failure in antidepressant pharmacotherapy. (2) Studies on dose increase of antidepressants other than SSRIs (and of citalopram) are needed.”

This research is particularly relevant in light of studies of adverse effects of antidepressants that are dose-dependent, such as hypertension, QT prolongation, sexual dysfunction, fracture risk, and liver injury. Additionally, high doses of SSRIs increase the risk of withdrawal symptoms. The researchers conclude:

“Clinically, and pending studies in children, with other antidepressants, and with longer initial treatment durations, we recommend not increasing the dose of antidepressants after initial treatment failure in antidepressant pharmacotherapy.”



Rink, L., Braun, C., Bschor, T., Hennsler, J., Franklin, J., & Baethge, C. (2018). Dose Increase Versus Unchanged Continuation of Antidepressants After Initial Antidepressant Treatment Failure in Patients With Major Depressive Disorder: A Systematic Review and Meta-Analysis of Randomized, Double-Blind Trials. Journal of Clinical Psychiatry. (Link)


  1. “Current guidelines recommend ‘antidepressant switch, augmentation with a second-generation antipsychotic or lithium, a combination of antidepressants, and a dose increase of the initially prescribed antidepressant,’ when initial treatment fails.”

    Bad ideas, as always with psychiatric guidelines, especially given these reminders.

    “… we recommend not increasing the dose of antidepressants after initial treatment failure in antidepressant pharmacotherapy.”

    Why not recommend taking a person off a drug that doesn’t work? Oh, that is against the rules of the religion of psychiatry.

    Report comment

    • There was a psychiatrist who helped me a lot. The best I ever had and one of the only two who gave a rip about me as a person.

      Dr. Guilley was his name. He frequently talked about my problems and took my side when Mom begged him to put me on Haldol again.

      He seemed excited at his personal discovery that “less was more” and he could help unhappy people detached from reality by lowering their drugs. Unfortunately I moved with my folks and saw Dr “Rached” in the next town.

      Is Dr. G. still in psychiatry? Something tells me he left the field. He honestly wanted to help folks. Pharma-psychiatry is a bad tool for that.

      Report comment

  2. “antidepressants” not only do not “improve” misery and suffering, they often cause additional misery suffering. Just as a suffering individual who turns to street drugs for escape eventually finds the need to increase their dose, so too do pill pushers/shrinks often need to ramp up the dose of their emotional novocaine pills.

    I suppose knowing that there’s some mainstream criticism of “antidepressants” coming out is somewhat comforting, but…not really. There’s plenty of on-patent “antidepressants,” which means decades more of lies, control, oppression, and…well…psychiatry.

    Report comment

  3. If you hit a screw with a hammer and it doesn’t go into the wood, hit it harder or with a different hammer, or maybe two hammers at once, or maybe a power hammer will work or…

    Does it ever occur to them that if drugs don’t improve the situation, maybe they should try something else besides drugs? Maybe, just maybe, this situation is not resolvable by drugs?

    Report comment

  4. My observations from working with psychiatrists for years is that the typical sequence of “treatment” for depression looks like this:

    1. Start the client on a SSRI, selected based on which company’s attractive young female drug rep has been the most influential lately.
    2. When that doesn’t work, increase the dose. Repeat 2-3 times.
    3. When that doesn’t work, add another drug, most likely an “antipsychotic” or “mood stabiliser.” If the client develops akathisia, add a benzo.
    4. Indefinitely maintain the client on an experimental drug cocktail that can expand but not contract.

    Steve, to answer your last question, I think the answer is usually “no.”

    Report comment

  5. I’ve found that reducing “medication” down to the lowest possible level works.

    There’s lots of effective alternatives to “medication” available that work, and these have been featured on this website.

    Report comment

  6. It’s very hard to resist when they want to max out the dose, you get asked repeatedly if you’ve changed your mind, and that they think its just on the cusp of working. This is wierd. From what I’ve seen the seratonin uptake is 80% clobbered at as little as 25% of a maximum dose and can never get beyond 90%. Where on earth all that antidepressant is actually going at those higher doses is anybody’s guess but it’s not going into reuptake inhibition.

    Report comment