Mental health

Study looks at creativity link with mental illness

‘Creativity is often part of a mental illness according to a study of more than a million people’, BBC News has reported.

The image of the tortured artist or the visionary genius beset by personal demons has long been part of our popular culture. But are ‘creatives’ really more prone to mental illness than say, bricklayers or book-keepers?

In an attempt to answer the question, researchers used Swedish health records to identify more than a million people diagnosed with various mental illnesses. They compared the occurrence of creative occupations among these people with that of a matched sample of ‘healthy’ people.

The BBC’s coverage of this research is a little misleading as it actually found that, with the exception of bipolar disorder, overall, people in creative professions were no more likely suffer from a psychiatric condition than anyone else. There was one exception – writers. People who wrote for a living were more likely than the general study population to suffer a range of disorders, including schizophrenia and depression. Writers were also more likely to commit suicide.

This study cannot explain the observed association, nor can it explain whether, how, or why, people with certain creative talents or dispositions may be more likely to suffer from mental health problems. It is worth noting that the researchers had to categorise people by ‘creative’ profession. People who are not in a so-called creative job can still be creative, and the authors’ idea of what stands as ‘creative’ may not be the same as anyone else’s.

Nevertheless, the study highlights how important it is that all people with mental health problems receive the support and treatment that they need.

Where did the story come from?

The study was carried out by researchers from the Karolinska Institutet, Gothenburg University and Uppsaala University, in Sweden.

It was funded by a number of Swedish institutions, including the Swedish Medical Research Council and the Swedish Psychiatry Foundation and was published in the peer-reviewed Journal of Psychiatric Research.

The BBC’s coverage was accurate for the most part, though the headline could have made it clearer that:

  • in general terms, being ‘creative’ was only linked to an increased risk of bipolar disorder
  • an increase in risk of other mental health conditions was only seen in writers

What kind of research was this?

The researchers say the “ancient issue of genius and madness” is of interest to both the public and to doctors.

Although a number of previous studies have shown a link between creativity and mental illness (dubbed by one psychiatrist the ‘Sylvia Plath effect’ after the American writer who killed herself), the quality of the research has often been poor and arguably subject to reporting bias.

That is, artists and writers who kill themselves often become high-profile news items and subject of literary biographies. Artists and writers who live contented and well-adjusted lives probably attract less attention.

The researchers were also interested in what they termed an “inverted-U model” - that is whether increased severity of symptoms of mental illness results in increased creativity up to a certain point, beyond which it starts to diminish. For example, the German philosopher, Friedrich Nietzsche supposedly suffered a ‘mental breakdown’ in 1889, after which he produced no more coherent work.

The researchers argue that any study on the association between creativity and psychiatric problems also needs to address the relatives of those affected by mental illness (presumably as many mental health conditions, such as schizophrenia, are known to be affected by genetics).

The authors presumed that relatives could have lower symptom severity, suggesting that they assume that relatives could be affected by similar mental health illnesses, but be below diagnostic thresholds. However, mental illnesses certainly don’t have to be hereditary, so this assumption is a little confusing.

Previous research by these authors has suggested that people with schizophrenia or bipolar disorder and their relatives were over-represented in creative occupations.

In this study, based on over one million people, they aimed to investigate if creativity is associated with all psychiatric disorders, or restricted to those with psychotic features (psychotic features generally means the presence of disordered thought patterns, delusions or hallucinations). They also aimed to specifically investigate whether writers tend to have more psychiatric illness.

The researchers used a type of study design called a nested case-control study. In this type of study, within the larger cohort study, each “case” (person with psychiatric disorders) is matched for age, sex, and other factors against a group of healthy controls selected from the large cohort population, to measure a particular outcome, which in this study, was creativity.

What did the research involve?

The researchers used a number of Swedish population registries to compare people with psychiatric diagnoses and their (non-diagnosed) relatives, with a matched group of those without psychiatric diagnoses. 

The psychiatric disorders they included were:

  • schizophrenia
  • schizoaffective disorder (a specific mood disorder with elements of schizophrenia)
  • bipolar disorder (a condition characterised by episodes of mania alternating with depression)
  • depression
  • anxiety disorders
  • alcohol abuse
  • drug abuse
  • autism
  • ADHD (attention deficit hyperactivity disorder)
  • anorexia nervosa

The researchers also looked at the number of completed suicides.

To get their sample of people with a mental illness, the researchers drew on a national patient register that provided diagnoses at discharge for all hospital inpatients between 1973 and 2009, and specialist outpatient treatment between 2001 and 2009. To identify those with psychiatric disorders from this register, they used standard disease coding. For each person with a psychiatric disorder and their relatives (the cases), they randomly selected 10 controls, matched on sex and age from the same population registries. The controls had to be alive, residing within Sweden, and without any inpatient episodes of the psychiatric disorders being looked at.

They took occupation data from mandatory national censuses, completed by all adult citizens at regular intervals from 1960 to 1990, which included the classification of people’s self-reported occupations.

They defined as “creative” anyone in a scientific or artistic occupation, including professional writing. Individuals reporting a creative occupation in at least one census were considered creative. However, the researchers provide little further detail of what were considered creative occupations.

They took information on IQ from a compulsory military conscription register, which included IQ results for all 18-19 year old men between 1969 and 2009. As only males are conscripted to the Danish armed forces, the IQ information was only available for the men in the study.

The researchers took their group of ‘cases’ (those with the diagnosis of one of the conditions mentioned above) and their family members, and compared the occurrence of creative occupations in these people to the control group.

What were the basic results?

The researchers identified 1,173,763 patients diagnosed with the specified psychiatric disorders, almost half who had suffered depression. Of these:

  • Aside from bipolar disorder, people in creative professions were no more likely to have a psychiatric disorder than those in the control group.
  • People in creative professions were significantly less likely than controls to be diagnosed with schizophrenia, schizoaffective disorder, depression, anxiety disorders, alcohol abuse, drug abuse, autism, ADHD or to have committed suicide.
  • As a specific group, writers, were twice as likely to suffer from schizophrenia and bipolar disorder as the controls. They were also more likely to suffer from depression, anxiety disorders, substance abuse, and to commit suicide.
  • First-degree relatives of people with schizophrenia, bipolar disorder, anorexia nervosa and siblings of patients with autism, were more likely to be in creative professions.
  • Differences in IQ did not account for any of the associations.

How did the researchers interpret the results?

The researchers point out that with the exception of bipolar disorder, they found no association between being creative and having a psychiatric disorder (although professional writers were more at risk for most of the disorders, and for suicide). The researchers say the findings about first-degree relatives (who share half their genes with the affected ‘cases’), may support the “inverted U-model” between psychiatric conditions and creativity.


Creativity has often been linked to qualities such as “nervous tension”, depression and use of alcohol and drugs.

However, this study seems to put paid to the idea that creativity, in general, is necessarily linked to diagnosed mental disorders.

The only psychiatric condition they found to be associated with a creative occupation was bipolar disorder, and the only specific creative profession they linked to psychiatric problems is writing.

It is difficult to conclude much about cause and effect from this study. Does being a writer, for example, lead to psychiatric problems? Or, do mental health problems result in people trying to express their inner feelings in a creative way?

This was a large and well-designed study, which benefits from the use of large population registries and valid diagnoses of mental health conditions, but it had some limitations.

Definitions of “creativity” are always difficult, and this research relied on peoples’ occupations, including that of academic research, as a proxy for creativity. The researchers considered ‘creative professions’ as scientific and artistic occupations.

Scientific occupations were said to include those conducting research and teaching at university, but beyond authors, no more expansion is given on what was considered to be an artistic occupation (for example, painting, singing, dancing, and acting were not mentioned). Consequently, the authors’ idea of what stands as ‘creative’ may not be the same as everyone elses.

There were also higher rates of missing occupational data among those with mental health illnesses than their controls, which could affect the reliability of the results.

Being a study using data collected over many years, the research is also subject to changing diagnostic systems, which might have made the results less reliable.

While the study is of interest, its implications for the support and treatment of mental health problems are unclear. As the charity Mind has pointed out, one in four people will be diagnosed with a mental health problem, and these individuals will come from a range of different backgrounds and professions. The main focus should be to ensure anyone with a mental health problem gets the information and support they need.

NHS Attribution