Chicago / Marburg (AP) - Air pollution may influence the incidence of mental illness. This is the conclusion of a study with health and environmental data from the US and Denmark.

The scientists around Atif Khan and Andrey Rzhetsky of the University of Chicago found in regions with particularly bad air quality increased cases for bipolar disorders and other diseases, as reported in the journal "PLOS Biology".

For the US, researchers evaluated data from health insurance for 151 million people. They examined the incidence of four psychiatric disorders - bipolar disorder, severe depression, personality disorder and schizophrenia - as well as the neurological disorders epilepsy and Parkinson's disease.

"These neurological and psychiatric disorders - both financially and socially very costly - seem to be linked to the physical environment, especially air quality," Khan said in a statement from his university. The health data matched the researchers with the air quality of each residential district, they took the information from the US Environmental Protection Agency EPA.

Results: In regions of the worst air quality, six percent more people suffer from severe depression than those from areas with particularly good air quality. In bipolar disorder, the disease risk was even increased by 27 percent.

In the second part of the study, the researchers then analyzed a Danish treatment and environmental register covering more than 1.4 million people born between early 1979 and late 2002 in Denmark. Here the rate of major depression in areas with the highest air pollution was about 50 percent higher than in the most clean areas. Researchers in Denmark also found increased scores for the other mental illnesses: the risk of personality disorders was increased by 162 percent, and that of schizophrenia by 148 percent. For bipolar disorder, the increase was 24 percent, similar to the US data.

The difference in the results is explained by the authors with the difference in the data evaluated: "It is likely that this difference is due to the limited resolution of the pollutant estimates for the US data," they write. But also the composition of pollutants or country-specific genetic variations could play a role.

In a commentary in "PLOS Biology," John Ioannidis of Stanford University, California, criticized the study for "significant shortcomings and a long line of potential bias." For example, in the US part, environmental data were measured in the years 2000 to 2005, while the disease diagnoses were from 2003 to 2013. "These analyzes, as well as subsequent studies in this area, would benefit from rigorous, well-defined protocols that are registered before the data is analyzed," writes Ioannidis.

Despite this criticism, Tilo Kircher from the University Hospital of Psychiatry and Psychotherapy in Marburg sees the study as an important contribution to medical research: "It will hopefully trigger further research in this area." The strength of the study is the huge amount of data.

Kircher considers the results to be plausible, although he wonders that the US data analysis revealed a significant association with air pollution only for bipolar disorder. The expert points to results from animal experiments, according to which particulate matter and pollutants could cause inflammation in the brain.