They point out some limitations in their study:
> This study has several limitations. First, we collected the claims data based on ICD-19 codes, which led to potential errors regarding mismatching or misclassification could have occurred. Second, previous studies have suggested the impact of poverty rates on mental disorders but there are differences in the baseline characteristics between the two groups in this study [45, 46].
...
> Third, we measured psychiatric AEs up to three months following COVID-19 vaccination. This study did not contain the long-term follow-up COVID-19 vaccine-associated AEs. Last, despite being a population-based study targeting the population in Seoul, South Korea, it cannot be generalized to the entire population, globally since psychiatric disorders could be related to ethnic and genetic backgrounds.
And summarize their conclusion as such:
> In conclusion, this population-based cohort study revealed that COVID-19 vaccination differentially affects occurrences of psychiatric disorders. It increased the risks of depression, anxiety, dissociative, stress-related, and somatoform disorders, and sleep disorders while reducing the incidence and risk of schizophrenia and bipolar disorder. Our findings suggested that the relationship between COVID-19 vaccination and mental illness may be underestimated along with the complexity of its impact on mental health.