Do antidepressants work for mild to moderate anxiety?

Imagine that you are a primary care doctor. A young woman comes to see you because she worries constantly about everything, from her health to her studies to the state of the world. At night, she often lies awake for hours, worrying. Because she is having difficulty concentrating, she has failed some of her classes this year. She tells you that she has always been a worrier, but things have gotten worse in the past year. You diagnose her with mild to moderate generalized anxiety disorder and start thinking about your treatment plan. Dutch treatment guidelines state that if medication is preferred by the patient, medications commonly used for depression (antidepressants) are the first choice. Could this patient benefit from these medications?

Antidepressants and severity of symptoms

Antidepressants are often used in the Netherlands. About 6% of the general population takes antidepressants every year [1]. Use has increased a lot over the last few decades, which has been a cause for concern. One worry has been that increasingly, people with relatively mild psychological symptoms also take antidepressants, even though the evidence at some point suggested that these people don’t benefit from them any more than they would benefit from a placebo [2][3]. Yet more recent studies using larger sample sizes and better methods actually did not find a relationship between severity of symptoms and antidepressant effectiveness [4][5][6]. In other words, it looks like patients with mild to moderate symptoms can also benefit from antidepressants.

Importantly, all of this research was done in depressed patients, not in anxious patients. Nowadays, however, about a quarter of all prescriptions are for treating anxiety [7]. But to date, only a few studies have looked at this issue for anxiety disorders, and these usually had a small sample size or suboptimal methods.

Consequently, we don’t really know whether symptom severity is a predictor of antidepressant effectiveness for anxiety. This is an important question because antidepressants can cause side effects that range from merely annoying but common (like nausea) to very rare but potentially dangerous (like suicidal thoughts). These side effects are off-set by the expected therapeutic effects of antidepressants, but if patients with mild to moderate anxiety in fact do not experience much benefit, it is likely that, on average, they are being harmed rather than helped.

Our study

To find out whether pre-treatment severity of anxiety predicts treatment effectiveness, we obtained data for about 9,000 patients [8]. These patients were diagnosed with an anxiety disorder and had participated in a clinical trial of antidepressants. We found that the answer to our question was “Yes” for two anxiety disorders, namely generalized anxiety disorder and panic disorder. For these disorders, the effect of antidepressants was smaller (though not zero) for patients with moderate pre-treatment symptoms than for patients with severe symptoms. Thus, the young woman from the introduction is likely to experience relatively few benefits from taking antidepressants, while being exposed to potential side effects. While this does not mean that antidepressants should never be prescribed to these patients, it does suggest that non-medication treatment options will usually be preferable as the first treatment choice.

For three other disorders, namely social anxiety disorder, obsessive-compulsive disorder, and post-traumatic stress disorder, the answer to our question was “No”. Antidepressants seemed to have a medium-sized effect for both less and more severely anxious patients. This means that patients with moderate pre-treatment symptoms of these other disorders will benefit about as much from taking antidepressants as patients with more severe symptoms, in contrast to what we found for generalized anxiety and panic disorder.

And now?

While this study may help doctors decide when to treat anxious patients with antidepressants, some limitations should be kept in mind. Most importantly, clinical trials generally exclude patients with mild symptoms, so it is not clear whether our results would hold for this group. Moreover, mild symptoms are more likely than severe symptoms to disappear over time even without treatment, so it could make sense to “wait and see” first or to start with “mild” treatments, like self-help. Finally, patient preferences are also important, and because patients often prefer psychotherapy over antidepressants, our results should not be interpreted as treatment advice.

Despite these limitations, our study also shows the potential of doing research with such large sample sizes, which provides much greater confidence that the findings are really true. It used to be almost impossible to get such large sample sizes, because the data from these clinical trials is owned by pharmaceutical companies. However, in the last few years, these companies have started sharing their data with independent researchers [9]. Hopefully, other researchers will also use this opportunity and we will get better insight into who does and who does not benefit from antidepressants in the years to come.

Relevant links and publications

[1] Verweij G, Houben-van Herten M: Bevolkingstrends 2013: Depressiviteit en antidepressiva in Nederland. Den Haag/Heerlen, 2013.

[2]  Kirsch I, Deacon BJ, Huedo-Medina TB, Scoboria A, Moore TJ, Johnson BT: Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration. PLOS Med 2008;5:e45.

[3] Fournier JC, Derubeis RJ, Hollon SD, Dimidjian S, Amsterdam JD, Shelton RC, et al.: Antidepressant drug effects and depression severity. JAMA 2010;303:47–53.

[4] Rabinowitz J, Werbeloff N, Mandel FS, Menard F, Marangell L, Kapur S: Initial depression severity and response to antidepressants v. placebo: patient-level data analysis from 34 randomised controlled trials. Br J Psychiatry 2016;209:427–428.

[5] Gibbons RD, Hur K, Brown CH, Davis JM, Mann JJ: Benefits from antidepressants: synthesis of 6-week patient-level outcomes from double-blind placebo-controlled randomized trials of fluoxetine and venlafaxine. Arch Gen Psychiatry 2012;69:572–9.

[6] Furukawa TA, Maruo K, Noma H, Tanaka S, Imai H, Shinohara K, et al.: Initial severity of major depression and efficacy of new generation antidepressants: individual participant data meta-analysis. Acta Psychiatr Scand 2018;1–9.

[7] Wong J, Motulsky A, Eguale T, Buckeridge DL, Abrahamowicz M, Tamblyn R: Treatment indications for antidepressants prescribed in primary care in Quebec, Canada, 2006 – 2015. JAMA 2016;315:2230–2231.

[8] de Vries YA, Roest AM, Burgerhof JGM, de Jonge P: Initial severity and antidepressant efficacy for anxiety disorders, obsessive-compulsive disorder, and posttraumatic stress disorder: An individual patient data meta-analysis. Depress Anxiety 2018;1–8.

[9] Clinical Study Data Request Available from:


NOTE: Image by afromztoa, licensed under CC BY 2.0

Ymkje Anna de Vries recently defended her PhD thesis on the treatment of depression and anxiety, which she wrote while at the Interdisciplinary Center for Psychopathology and Emotion regulation (ICPE) at the University Medical Center Groningen. Her thesis was focused on examining the effect of reporting bias on the apparent efficacy and safety of antidepressants and psychotherapy, and on investigating clinical characteristics that predict response to antidepressants. She currently works as a postdoctoral researcher at the ICPE and the Department of Developmental Psychology, where she studies the development of comorbidity between mental disorders.

You may also like

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.