Antidepressants in primary care: limited value at the first visit.

When patients with a depressive condition first visit a general practitioner, they often get the prescription of an antidepressant1. We think that it is better to prescribe medication at a later stage, if at all. Here we explain why. It is well known that most patients in primary care have mild to moderate depression, while severe depression is an exception. For example, we found that, among primary care patients in waiting rooms, 13% had a score on the Patient Health Questionnaire-9 (PHQ-9) between 9 and 11, which is above the threshold for major depression, but only 5% had a severe depression (PHQ-9 score higher than 14)2. There is also considerable evidence that the effects of antidepressants in mild and moderate depression are small, and may not be clinically relevant. In one individual patient data meta-analysis, the risk difference (percent response to medication minus percent response to placebo) was only 6% in mild depression, which corresponds to a number needed to treat (NNT) of 163. In very severe depression, the risk difference was 25% (NNT=4); in severe depression, it was 9% (NNT=11). These results were recently confirmed in a large individual patient data meta-analysis of 232 trials with more than 73,000 patients4. Furthermore, a recent pragmatic placebo-controlled trial confirmed that antidepressants are not very effective in patients with mild depression seen in primary care: with an average PHQ-9 score of 12, the NNT was only 12.55. It is also well known that many patients in primary care who use antidepressants are not willing to stop their medication, even when it is clearly not working, because they are afraid that they will get worse. Much of the confusion about the effects of medications in depressed patients seen in primary care is due to an earlier Cochrane review6, reporting that the NNT was 8.5 for tricyclic antidepressants and 6.5 for selective serotonin reuptake inhibitors, which would be considered a reasonable clinical effect by most clinicians. However, the problem with that review was that the included trials focused on patients with severe to very severe depression, thus being not representative of the majority of patients with depression seen in primary care. The above-mentioned meta-analyses and pragmatic trial provide a much better evidence of the effects of antidepressants in this population. Even for patients with more severe depression seen in primary care, antidepressants may not be the best treatment at the first visit. Many of the few patients who initially present in primary care with a severe depression get better over time with or without medication7. Indeed, the above-mentioned Cochrane review found a median response rate of 42% with pill placebo. So, what to do at the first visit in primary care with a patient who presents with a depressive condition? Most treatment guidelines, such as those of the National Institute for Health and Care Excellence (NICE), recommend watchful waiting or a psychological intervention before medication for mild to moderate depression, unless it is the person’s preference to receive an antidepressant. Behavioural activation may be the best intervention8, but also other brief therapies specifically developed for this context, such as problem-solving therapies, may be good treatment options. It is less clear what should be done for severe depression at the first visit in primary care. The best strategy may be to reframe some of the negative cognitions of the patient and advice physical activity. In those who do not improve over the subsequent weeks, a psychotherapy or antidepressant medication should be considered. A recent meta-analysis showed that, at one-year follow-up, psychotherapies had better results than antidepressants9. This meta-analysis also found that a combination of psychotherapy and medication was better than either therapy alone. We conclude that most patients in primary care have mild to moderate depression, and that severe depression is an exception. Antidepressants should not be prescribed at the first visit if the patient has mild to moderate depression, because they have a limited efficacy and may have significant side effects. Antidepressant medication should be considered in severe depression, but not at the first visit and as an alternative to or in combination with a psychological intervention.