in which the authors examined the association between mental illness and testing positive for COVID-19 with data from a national register in South Korea. It was surprising that a diagnosis of a mental illness was not associated with an increased likelihood of testing positive for COVID-19, given that this opposed the results of recent research findings. Yang and colleagues
reported that pre-existing psychiatric disorders are associated with an increased risk of COVID-19, based on data from the UK Biobank. Additionally, Taquet and colleagues
showed that a psychiatric diagnosis within the previous year was associated with a higher incidence of a COVID-19 diagnosis, based on data from the TriNetX Analytics Network in the USA.
We questioned why there were such different results. An interesting finding of Yang and colleagues was that the risk of COVID-19 differed based on pre-existing ICD-10 psychiatric disorders.
The risk of COVID-19 was elevated in cases of depression, anxiety, substance misuse, and psychotic disorders; however, risk was not increased in cases of stress-related disorders.
In Yang and colleagues’ Article, of the 50 809 people with prepandemic psychiatric disorders with COVID-19, 44·0% (n=22 352) had depression, 32·6% (n=16 573) had anxiety, 1·7% (n=872) had stress-related disorders, 46·5% (n=23 620) had substance misuse, and 2·8% (n=1431) had psychotic disorders.
Therefore, according to Yang and colleagues, pre-existing psychiatric disorders were associated with an increased risk of COVID-19, reflecting the results of patients with depression, anxiety, and substance misuse.
Lee and colleagues defined mental illness, on the basis of the ICD-10, as non-affective psychotic disorders, affective psychotic disorders, anxiety-related and stress-related disorders, alcohol or drug misuse, mood disorders without psychotic symptoms, eating disorders, and personality disorders.
However, they did not mention the exact proportions of these diagnoses for the analysis. We considered that the differences in the proportions of psychiatric diagnoses might have affected the differing research findings, and we recommend that the authors mention the number of patients with each psychiatric diagnosis.
We declare no competing interests.