Antidepressant side effects may not universally improve as treatment progresses, a new study published in Acta Psychiatrica Scandinavica reveals. Researchers found that patients experiencing severe side effects early in their treatment are more likely to discontinue their medication, which skews the overall perception of side effect reduction. Instead of a uniform improvement, the drop in reported side effects may reflect the absence of those who could not tolerate the medication.
Citalopram (commonly known as Celexa) is a widely used antidepressant that belongs to a class of medications called selective serotonin reuptake inhibitors. Like others in this class, citalopram works by increasing serotonin levels in the brain, which is believed to improve mood, emotional regulation, and overall mental health. Despite its benefits, citalopram, like other antidepressants, is associated with side effects ranging from mild to severe, including nausea, fatigue, sexual dysfunction, and sleep disturbances.
The need for this study arose from conflicting narratives about side effects and their progression during antidepressant treatment. Conventional wisdom and clinical guidance often reassure patients that side effects subside as the body adjusts to the medication. However, in practice, many patients report persisting or worsening side effects, which often lead to early discontinuation of treatment. This mismatch between expected and actual patient experiences prompted the researchers to question whether the perceived improvement in side effects over time is due to physiological adaptation or if it reflects the dropout of patients who cannot tolerate the medication.
“I became really interested in this topic while discussing side effects with friends and collaborators who are psychiatrists,” explained study author Thomas T. Kim, a fellow of psychology in psychiatry at Weill Cornell Medical College.
“They commented that: 1) they wished there was more information about the frequency of side effect symptoms over the course of antidepressant treatment; and 2) while they were taught that the severity of side effect complaints improved over time with medication persistence, they rarely saw this in practice (either they opted to change their patient’s medication for those experiencing severe side effect complaints or patients dropped out of treatment).”
“The ‘traditional belief’ that is taught is the following: ‘[Antidepressant] side effects usually do not get in the way of daily life, and they often go away as your body adjusts to the medication’ (CDC, 2023). When I examined the literature further, I didn’t find that studies controlled for patients who dropped out across time. Generally, studies concluding that side effect complaints improved over time were conducted on treatment completers. So, in short, I found the research to be inconclusive: there was evidence for both sides, and I just wanted to know exactly what the truth was.”
The study utilized data from the Sequenced Treatment Alternatives to Relieve Depression (STARD) trial, a large-scale clinical investigation designed to assess antidepressant treatment in real-world settings. STARD included 2,833 participants aged 18 to 75, all diagnosed with nonpsychotic major depressive disorder. These participants underwent an initial 12-week treatment course with citalopram. Patients who did not achieve satisfactory outcomes during this phase could opt for subsequent treatment steps, but this study focused solely on the first step with citalopram.
Participants attended regular follow-up visits at weeks 2, 4, 6, 9, and 12, during which they completed self-reported assessments of depressive symptoms and side effects. To measure side effects, the researchers used validated tools such as the Patient Rated Inventory of Side Effects (PRISE) and the Frequency, Intensity, and Burden of Side Effects Rating (FIBSER). These tools captured various aspects of side effects, including their frequency, intensity, and overall impact on daily functioning.
To account for differences in when patients dropped out of treatment, participants were categorized into five groups based on their last recorded visit: those who dropped out by weeks 2, 4, 6, or 9, and those who completed the full 12 weeks. The researchers used a statistical approach called pattern-mixture modeling, which allowed them to analyze changes in side effects over time within each group while accounting for the influence of dropout patterns.
The researchers found differences in side effect trajectories depending on whether and when patients discontinued treatment. For those who completed the 12-week course, side effects generally decreased over time in terms of frequency, intensity, and burden. This aligns with the traditional belief that side effects diminish with continued use of the medication.
“Interestingly, if you only look at treatment completers (or patients who drop out later), you will find that side effect complaints are mild and do improve over time,” Kim told PsyPost.
However, the findings for those who dropped out early told a different story. Patients who discontinued treatment within the first six weeks (weeks 2, 4, or 6) reported persistently severe or even worsening side effects before they stopped. For these individuals, side effect burden increased rather than decreased, suggesting that early discontinuation was closely tied to an inability to tolerate the medication.
The most common side effects reported by patients included sleep disturbances, sexual dysfunction, and fatigue. These side effects were often rated as distressing, particularly in the early weeks of treatment. Importantly, the analysis demonstrated that the average decline in side effects across all participants was partly due to the dropout of patients with the most severe side effects, rather than a universal improvement among all patients.
“There is a ‘traditional belief’ in psychiatry that ‘[antidepressant] side effects usually do not get in the way of daily life, and they often go away as your body adjusts to the medication,'” Kim said. “However, we did not find this to be true for all depressed patients. If your side effect complaints are severe or are worsening early on in treatment, you and your doctor might want to consider changing treatments (e.g., psychotherapy or a different antidepressant).”
While the study provides valuable insights, it also has limitations. The analysis focused solely on citalopram, and it remains unclear whether the findings apply to other types of antidepressants. The study also lacked a placebo group, which would have provided a baseline for comparing side effects attributable to medication versus other factors.
Future research could explore several avenues to build on these findings. For instance, collecting pre-treatment data on symptoms could clarify how much of the reported side effects are directly related to the medication. The researchers are also interested in developing predictive tools to identify patients at risk of early dropout due to severe side effects. These tools could help clinicians tailor treatment plans to improve adherence.
“My colleague, Dr. Colin Xu, and I want to use machine-learning algorithms to develop a model that identifies which types of antidepressant side effect symptoms predict early attrition,” Kim explained. “We just submitted a grant to conduct this study: once we have a predictive model, we aim to conduct an implementation trial in which we compare dropout rates between clinicians equipped with this model versus treatment-as-usual.”
The study, “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,” was authored by Thomas T. Kim and Colin Xu.