No, Your Patients Are Not Wrong: Sometimes Antidepressant Side Effects Do Not Get Better

ksena32/AdobeStock

CLINICAL REFLECTIONS

When a patient starting antidepressants for major depressive disorder voices their concerns about potential side effects, it is common for clinicians to offer patients the same reassurance that many major health agencies have advised: Stick with the medications, and your side effects should improve with time. For example, the National Institutes of Health (NIH)’s public-facing webpage on mental health medications reads: “The side effects [of antidepressants] are generally mild and tend to go away with time.”¹ Likewise, the Centers for Disease Control and Prevention (CDC) publicly states, “Side effects usually do not get in the way of daily life‚ and they often go away as your body adjusts to the medication.”² This perspective that antidepressant side effects will eventually go away is not just exclusive to the United States: The United Kingdom’s National Health Service (NHS) states on their public-facing antidepressants overview page that “the most common side effects of antidepressants are usually mild. Side effects should improve within a few days or weeks of treatment as the body gets used to the medicine.”³

Nevertheless, many psychiatrists and other mental health clinicians have encountered patients who report the opposite experience. Although many patients experience an improvement in side effects with time, not everyone’s side effects improve. In fact, it is not uncommon to encounter patients who report worsening side effects to the point where some decide to quit treatment. Indeed, the No. 1 self-reported reason for why patients prematurely discontinue antidepressant pharmacotherapy is side effects.⁴

One question then arises: Why does such a dichotomy exist between the clinical consensus (as publicly stated by the NIH, CDC, and NHS) that side effects improve with time and the anecdotal experiences of patients who report that their side effects do not go away or, in some cases, even worsen?

We noticed that past research examining antidepressant side effects often failed to account for 1 important confounder: dropout. That is, many studies on antidepressant side effects focused on individuals who completed treatment while neglecting perhaps the most interesting group of patients: those who may have dropped out of antidepressant treatment prematurely due to side effects.

We conducted a secondary analysis of side effects data from patients in the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial, the largest antidepressant trial ever conducted.^5-7 In the first treatment step of the STAR*D trial, all patients received citalopram for an intended 12 weeks per protocol. During these 12 weeks, patients reported their side effect frequency, intensity, and burden on the Frequency, Intensity, and Burden of Side Effects Rating (FIBSER) scale at weeks 2, 4, 6, 9, and 12. Additionally, patients reported which side effects they experienced in 9 organ/function systems on the Patient-Rated Inventory of Side Effects (PRISE) scale.

We wanted to examine how side effect frequency, intensity, and burden on the FIBSER scale changed over the course of citalopram treatment. What we were most interested in was how side effect complaints of patients who dropped out early in treatment differed from those who completed treatment. To answer this question, we used pattern-mixture modeling to model side effect complaints at each time point (weeks 2, 4, 6, 9, and 12) for each potential treatment attrition pattern (dropout at week 2, 4, 6, or 9 or full 12-week treatment completion) while controlling for changes in depressive severity over the course of treatment (Figure).

FIGURE. Side Effect Burden on the FIBSER Scale Over 12 Weeks of Citalopram Therapy

What we found does not disagree with the NIH/CDC/NHS consensus that side effects improve over time: Indeed, when examining only data from those who completed the full 12-week treatment, these patients reported decreases in side effect frequency, intensity, and burden over the course of treatment. Yet our findings also validated the experience of those patients who report that their side effects never improve. Specifically, when examining data from patients who dropped out of the trial early, a different pattern of side effects emerged: Patients who dropped out after weeks 2, 4, and 6 reported significantly more severe initial side effect complaints than those who completed treatment. And perhaps even more importantly, patients who dropped out after weeks 4 and 6 further showed a worsening of side effects over the course of treatment.

Taken together, what we see in the STAR*D data are several distinct patterns in how patients experience antidepressant side effects. On the one hand, there are many patients—namely, those who complete treatment—who are able to tolerate the side effects of antidepressants. These patients not only report lower severity of side effects after antidepressant initiation but also an improvement of side effects over time. It is likely that these are the patients whom the NIH/CDC/NHS consensus guidelines refer to when they offer the reassurance that side effects will decrease over time.

On the other hand, there is a nonnegligible population of patients with a much lower tolerance for side effects. These patients not only report more severe side effects immediately after antidepressant initiation, but many also report experiencing a worsening of side effects over the course of treatment—up to and until the point they drop out. These are the patients whom we and our colleagues see in clinical practice every day: those who attempt to persist with their prescribed antidepressant but ultimately drop out due to the intolerability of their side effect symptoms.

What is especially surprising is that this second group of patients—those with intolerance for side effects and for whom side effects do not improve—have gone previously unnoticed in the research literature. Our analysis was not of novel data: The STAR*D trial is famous for being the largest antidepressant trial ever conducted,⁷ and the data are publicly available from the NIH. However, it seems possible that previous research on antidepressant side effects—from which the major health agencies of the NIH, CDC, and NHS may have derived their guidelines—has focused primarily on treatment completion and has neglected those who drop out of treatment early. The unfortunate part of this oversight is that this second group of patients is perhaps most affected by side effects. After all, side effects are the No. 1 self-reported reason for why patients prematurely discontinue antidepressant treatment.⁴

What do our findings mean for psychiatrists and other mental health clinicians? Past research has shown that patients who prematurely drop out of antidepressant treatment often do not return for any mental health treatment and show a poorer long-term prognosis.⁸ Consequently, it is especially important for clinicians to pay attention to patient reports of severe or worsening antidepressant side effects as potential warning signs of attrition. The next time one of your patients reports experiencing antidepressant side effects, instead of universally offering the same assurance from the NIH/CDC/NHS guidelines that their problems ought to eventually improve, it may be worth considering switching to an alternative treatment, such as a different medication or even a nonpharmacological treatment such as psychotherapy.

There is another question that remains unanswered: Which side effects are more strongly linked to dropout? It is common for patients to self-report being concerned about some side effects more than others, but less is known about which effects cause patients to drop out. We are currently conducting a study to answer this question using data from the PRISE scale in the STAR*D trial and plan on developing a tool for clinicians that will flag patients for dropout risk based on their side effect profile. This tool could help inform psychiatrists in developing their treatment plans, especially for the patients at highest risk for dropout due to antidepressant side effects. We plan on validating our preliminary results in other data from clinical trials or medical records of antidepressant side effects. If you have access to such data and are interested in collaborating, please contact Colin Xu, PhD, at colinxu@uidaho.edu.

Dr Xu is an assistant professor in the Department of Psychology & Communication at the University of Idaho in Moscow. Dr Kim is a fellow of psychology in the Department of Psychiatry at Weill Cornell Medical College in New York, New York.

References

1. Mental health medications. National Institute of Mental Health. Updated December 2023. Accessed February 19, 2025. https://www.nimh.nih.gov/health/topics/mental-health-medications

2. Mental health conditions: depression and anxiety. CDC. Updated October 13, 2023. Accessed February 19, 2025. https://www.cdc.gov/tobacco/campaign/tips/diseases/depression-anxiety.html

3. Overview - antidepressants. National Health Service. Updated November 4, 2024. Accessed February 19, 2025. https://www.nhs.uk/mental-health/talking-therapies-medicine-treatments/medicines-and-psychiatry/antidepressants/overview/

4. Niarchou E, Roberts LH, Naughton BD. What is the impact of antidepressant side effects on medication adherence among adult patients diagnosed with depressive disorder: a systematic review. J Psychopharmacol. 2024;38(2):127-136.

5. Kim TT, Xu C. Not all types of depressed patients who persist with their antidepressant treatment improve in side effect complaints: a comparison of treatment completers and dropouts in the STAR*D trial. Acta Psychiatr Scand. 2025;151(2):152-162.

6. Rush AJ, Trivedi MH, Wisniewski SR, et al. Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR* D report. Am J Psychiatry. 2006;163(11):1905-1917.

7. Gaynes BN, Rush AJ, Trivedi MH, et al. The STAR*D study: treating depression in the real world. Cleve Clin J Med. 2008;75(1):57-66.

8. Burton C, Cochran AJ, Cameron IM. Restarting antidepressant treatment following early discontinuation—a primary care database study. Fam Pract. 2015;32(5):520-524.