Skip To Content
Cambridge University Science Magazine
Disclaimer: This article discusses depression and includes a mention of suicide. Treatment for depression is available and suicide is preventable. If you are struggling, please visit the Samaritans website (or call 116 123) or visit Nightline.

As the coronavirus pandemic sweeps across the globe, it seems as though all scientific forces have been mobilised: labs have worked quickly to identify the virus, sequence its genome, and search for potential targets to fight it, while global health practitioners use intra- and international data to predict the virus’ trajectory. Meanwhile, over the last couple of decades, a more gradual and far less recognised pandemic has been emerging. Depression has become one of the greatest public health problems of our time, with 284 million people affected worldwide according to the World Health Organisation, amounting to nearly 4% of the global population. It is estimated that 800,000 people commit suicide anually, making it the second leading cause of death in 15-29 year olds[1].

Treating depression should therefore be considered an effort to save lives, just as the fight against coronavirus has been. However, in the case of COVID-19, scientists are armed with a very important piece of information: the virus itself. Knowing the specific cause of the infection, they can look at how our bodies respond to it. The fight against depression does not have this advantage. Genome-wide association studies (GWAS) have tested vast numbers of patients with depression, comparing their genes against those of healthy individuals, looking for any genetic variation which could explain what makes some people more vulnerable to depression. Similarly, scientists have also been comparing patterns of brain activity using functional neuroimaging, hoping to find something which sets depression patients apart from their healthy peers. However, findings have been mixed and difficult to replicate, and no clear cause has been identified. This lack of an identifiable, biological culprit may be one of the reasons why depression remains a highly misunderstood and sometimes trivialised disorder. Now, scientists are reflecting back on the approaches historically used to study depression and are beginning to forge a new path, shifting away from a symptom-oriented approach and instead turning to the brain itself for answers.

There is a real urgency to rethink strategies, as treating depression and protecting patients has until now often amounted to shooting in the dark. While a lucky shot can happen (such as with the promise of ketamine-based antidepressants), the development of new treatments has been incredibly slow. Moreover, for the treatments that do exist, it is also not clear why they work for some people but not others. For example, only half of people diagnosed with depression will achieve remission following treatment with the most commonly available anti-depressant drug[2].

So, is there something fundamentally wrong about how we have been studying or thinking about depression? Do we even know what depression is? The key issue might come down to the fact that ‘depression’ is, by definition, a collection of symptoms. The diagnosis is made by a mental health professional if a person experiences at least five symptoms, including either low mood (sadness) and/or anhedonia (the inability to experience pleasure or enjoyment), among other common signs such as weight loss (or gain), trouble sleeping or concentration problems.

In 2015, Dr Eiko Fried and Dr Randolph Nesse outlined the issue with this method of diagnosis using some simple but eye-opening maths: there are 227 possible unique combinations of these symptoms which can give the same diagnosis for depression. Of course, just because it is mathematically possible does not mean that this kind of variation actually exists in patients. However, Fried and Nesse are quick to dispel this point: taking into account a large-scale study of 3703 patients with depression, they found that 83.9% of the possible symptom profiles were only present in 5 or fewer patients[3]. Therefore, what is typically described and diagnosed as depression can amount to a very different experience for different people. “I don’t think depression is a good scientific phenotype to study,” writes Dr Fried on his blog. “It is highly heterogeneous (…) depression symptoms like concentration problems, fatigue, sadness, weight loss are not just interchangeable indicators of a disorder”[4].

Perhaps depression as we know it is therefore an outdated construct, and we need to pay more attention to the specific symptoms, instead. However, this symptom-oriented approach is riddled with its own problems. As Dr Fried writes in another article, describing depression is made more complicated by the existence of multiple questionnaires, or ‘depression scales’, which are currently used to conduct research. These scales determine the symptom profile and severity of depression in patients who are recruited for a scientific study. While the 7 most popular scales are often used interchangeably, with little to no justification for using one over another, Dr Fried has shown that there is strikingly little overlap in what they actually measure[5]. This has brought in another source of variation, making it relatively unsurprising that consistent findings are hard to come by.

This is not the only issue, as factors such as age of onset, sex and co-existing disorders also come into play. Perhaps if the fields of immunology and microbiology were less advanced, we would mistake colds, flus and COVID-19 infections for the same thing. Gradually, we would start differentiating between the individual symptoms and looking for patterns - coughs, sore throats, fevers – but again, we would run into problems. We would discover that older people are more affected than the young, men respond differently to women, and having another medical condition can have a big impact on disease progression. These very same aspects are making the symptom-oriented approach in studies of depression an incredibly bumpy road towards understanding the disease.

The culprits are clearly well hidden, but maybe if we tune into the brain we will find them. A review published this year in Neuropsychopharmacology has presented some new studies which have not only abandoned the idea of depression as a single, uniform disorder, but have also stopped using symptoms as a starting point altogether[6]. A number of studies described in the review use sophisticated statistical methods to look for distinct patterns of brain activity which can then be used to predict a particular feature of depression – rather than looking at symptoms first and the brain second, they are letting the patterns of unusual brain activity lead the way to outlining clinically relevant groups of patients. This can be used to look specifically at people formally diagnosed with depression, as well as a “trans-diagnostic” approach. This means that some studies are not restricted to individuals with depression, and in fact do not specify a diagnosis of any disorder. This is important, as some psychiatric symptoms are shared by multiple disorders. For example, anhedonia, or the loss of pleasure and enjoyment, is a feature of both depression and schizophrenia. It is therefore important to ask the question: is it possible that both are caused by similar changes in the brain?

This new wave of research bears a lot of promise not just to depression, but on other psychiatric disorders as well. Of course, it must be emphasised that there are treatments for depression, both behavioural (such as cognitive behavioural therapy) and pharmacological. However, these new studies could provide a faster and more efficient track towards a better understanding of the disorder, which in itself will hopefully fuel better research. The true test will of course be whether the newly emerging findings can be replicated, and whether they can lead the way towards improved treatment options – options which are not only effective, but targeted towards the very people who will benefit from them the most.

Zuzanna Stawicka is a PhD candidate in Neuroscience at Fitzwilliam College. Artwork also by Zuzanna Stawicka.

  1. [BACK]WHO Depression Factsheet.
  2. [BACK] Nestler et al. Neurobiology of depression. Neuron (2002). 28;34(1):13-25
  3. [BACK] Fried, E. and Nesse, R. Depression is not a consistent syndrome: An investigation of unique symptom patterns in the STAR*D study. J Affect Disord (2015). 1;172:96-102.
  4. [BACK] Fried, E. The replication crisis hits psychiatry: no candidate genes for depression. (2019).
  5. [BACK] Fried, E. The 52 symptoms of major depression: Lack of content overlap among seven common depression scales. J Affect Disord (2017). 208: 191-197.
  6. [BACK] Buch, A., and Liston, C. 2020; Dissecting diagnostic heterogeneity in depression by integrating neuroimaging and genetics. Neuropsychopharmacol. (2020).