Menu Close
Australians are some of the biggest anti-depressant users in the world. Volkan Olmez/Unsplash, FAL

Antidepressants may not be as effective as we thought, and shouldn’t be the only treatment for depression

The treatment of depression too often means treatment with antidepressants. Australia has one of the highest rates of antidepressant use in the world. This continues to increase despite mounting evidence they are not especially effective.

My colleague Andrew Chanen and I have just published an article that describes the apparent falling effectiveness of the medications. We argue that doctors have become too reliant on them. When medications are used to treat depression they should be part of an overall treatment plan and shouldn’t be the treatment plan.

The falling effectiveness of antidepressants

Why are antidepressants becoming less effective? Partly because we haven’t always had all of the data. The clinical sciences have a problem with negative trial results – trials where the experimental treatments don’t appear to work. They are seen as uninteresting, and as undesirable by drug companies, and have often gone unpublished.

Drug trials are, however, regulated and require registration with authorities before they begin. So, over the past decade, researchers have tracked them down. Once they have found the registered-but-unpublished trials, they have included the data in their overall assessment of the medications’ effectiveness. Unsurprisingly, the result has been that the recorded effectiveness of the medications has fallen.

Early drug trials are usually conducted in highly controlled university research environments. The researchers, often working in partnership with the pharmaceutical companies, enrol uncomplicated, motivated, middle-class patients into the trials in an effort to give the trial medications the best chance of success.

Later, researchers are keen to see if the medications work in “real world” patients: the sorts of patients we see in mental health clinics and GP practices, who may not only be depressed but also anxious, drinking too much and distressed about their mounting bills. The medications don’t work as well in these patients.

The increasing effectiveness of placebos

Perhaps the biggest reason for the declining effectiveness of the antidepressant medications is that placebos are becoming more effective. The gap between the medications and placebos is steadily narrowing.

The placebo response is a complicated phenomenon. In part it illustrates the statistical concept of “regression to the mean”, where a measure that is extreme when first measured (depressive symptoms in this case) will tend to be less extreme when remeasured.

Explainer: what is the placebo effect?

The other component of the placebo response is a positive expectation bias. When people expect to improve, this makes it more likely they will improve. This is particularly important for depression, because by providing someone with treatment, if only a placebo pill, we are directly addressing the sense of hopelessness that is one of depression’s core symptoms.

The increasing placebo response rate in depression is likely driven by an increasing expectation that treatment will work. Notwithstanding recent evidence about the declining effectiveness of antidepressant medications, there is a broad cultural belief – one that has been emphasised in recent decades – that taking a pill can help depression.

Combining treatments

Antidepressants might not be as effective as we once believed. But, overall, they are effective.

Other treatments have similar problems with declining effectiveness. In fact, there are no well-studied treatments for depression that have consistently strong effects.

This suggests combining treatments might be the best approach. And the evidence bears this out: combined treatment with medication and psychotherapy is more effective than either alone. We should be moving beyond a simplistic view of alternative treatments as competition for medications and consider whether they might be usefully combined to deliver even more effective treatment.

Treatment recommendations

When medications are used they should be part of a broader treatment plan. When therapy is available – and it isn’t always – there can be few good reasons for not recommending it. Medication should be considered when the depression is reasonably severe, when psychotherapy is refused (not everyone wants to see a therapist), or when psychotherapy hasn’t been effective.

When medication is used, it should be used in a way that maximises its chances of being effective. This means not remaining on the same ineffective low dose for months and months. It means close monitoring by a doctor, so when the medication isn’t effective there is consideration of a dose increase or a change to an alternative medication.

Other treatments can also be added. Improving diet and exercising more are good for depression, and combining antidepressants with nutraceuticals – food-derived nutrients such as fish oil and vitamin D – has been shown to improve the effectiveness of the medications.

Future treatment approaches

It is unlikely we are going to see treatments with significantly greater effectiveness than existing treatments in the near future. Drug companies have reduced their investment in developing new drug treatments for mental illnesses, largely because they have been burnt by so many failures.

And the psychotherapies, while requiring a lot of training and skill to deliver competently, essentially comprise two people in a room talking. It is difficult to see how new therapies could be much more effective than existing ones.

Our task as clinicians is to consider the broad range of treatments that are available, and how they might best be combined in treating the particular patient in front of us. Our task as researchers is to work out the characteristics of the patients who are most likely to respond to particular treatments, so that we provide evidence for delivering the treatments to those patients. There is still much work to do.

Want to write?

Write an article and join a growing community of more than 185,800 academics and researchers from 4,984 institutions.

Register now