select search filters
roundups & rapid reactions
factsheets & briefing notes
before the headlines
Fiona fox's blog

expert reaction to systematic review and meta-analysis on air pollution exposure and depression and suicide

Research, published in Environmental Health Perspectives, reports a link between risk of depression and suicide with air pollution levels.

This Roundup was accompanied by a Before the Headlines report.


Dr Katharina Janke, Lecturer in Health Economic, Lancaster University, said:

“The authors repeatedly acknowledge that a causal relationship between air pollution and mental health has not been established. They acknowledge that important confounders were not included in the studies. For example, none of the studies adjusted for green space, which could affect both air quality and mental health. Another potential confounder the authors identified and which most studies did not control for is noise. Noise caused by traffic is clearly associated with air quality and at the same time likely to affect mental health. There are many more potential confounders that the studies did not adjust for. 

“What the studies of long-term effects essentially do is compare two individuals who are similar in terms of age, gender, body mass index, smoking, alcohol drinking, marital status, urbanity, income, education level etc. but different in terms of PM2.5 exposure level. If the general tendency is for individuals with higher PM2.5 exposure levels (who are otherwise similar to individuals with lower PM2.5 exposure levels) to be more likely to experience mental health problems, we will find an association between PM2.5 exposure and mental health. However, association (or correlation) is not causation. Even after controlling for confounding factors such as income (poorer people tend to live in more polluted areas and have worse mental health) there can still be confounding factors that we did not control for. Why might a city with similar demographics have higher levels of PM2.5? Maybe it has a badly functioning public transport system, leading to a stressful commuting experience for residents. Such stressful experience might affect mental health. So we might erroneously attribute the mental health problems caused by congested roads to the higher PM2.5 levels caused by the cars that cause congested roads.

“None of the studies investigating the long-term association between air quality and mental health use experimental or quasi-experimental designs. Arguably, experimental designs – randomly exposing some people to higher air pollution levels than others – are unethical. But quasi-experimental designs that exploit events or policy changes that result in random allocation of air pollution levels are much less likely to be affected by confounding. That said, Isen et al. (2017) used a quasi-experimental design and found negative impacts of early childhood exposure to air pollution on labour market success at age 30! Maybe there is a biologically plausible link between air pollution and mental health, which we simply do not yet fully understand. Some time ago we did not fully understand how air pollution can cause death from cardiovascular diseases. By now this relationship is accepted as being causal, see for example the recent report by COMEAP.


Dr Ioannis Bakolis, Senior Lecturer in Biostatistics and Epidemiology, Institute of Psychiatry Psychology & Neuroscience, King’s College London (IoPPN), said:

“This is a comprehensive review over a 40 year period which augments a similar review on the topic which was published recently (X Gu et al. 2019) which reported almost identical findings for depression and suicide.

“An interesting finding for the reported association between PM2.5  and depression is that although the studies included were from different parts of the world (e.g. China, US, Germany) with variation in the sample size, study designs and measures of depression, the reported associations were very similar – there was not much difference in the size or the direction of the association reported between the different studies. However, these associations might disappear with further statistical adjustments or with the inclusion of additional studies from other countries. 

“These results add to the current literature of a universal positive association between particulate air pollution exposure and depression. However, more studies are needed on the topic to conclude this as these reported results needs to be considered against the limited evidence-base- there were only 5 studies for PM2.5 and 3 studies for PM10 for depression and even fewer studies for suicide and anxiety being included in the review.

“One of the strengths of the review is that it highlights the sparsity of research around the topic as few studies to date have tried to understand the dynamics of air pollution and their links with poor mental health. Perhaps air pollution does really affect depression, or perhaps it doesn’t.  This review summarizes the existing evidence, but in my view we cannot tell with any certainty yet.

“Reducing pollution at its source can have a rapid and substantial impact on health ( Thus, new research is needed on the impact of pollution reduction policies on population mental health, in order to strengthen the findings of the current studies so that public health researchers and policy makers can understand the societal and economic implications in order to intervene.”


Dr Lisa Page of the sustainability committee at the Royal College of Psychiatrists, said:

“This new study builds on increasing evidence that shows a link between air pollution and mental illness.

“The small but strong associations between exposure to air pollution and depression and suicide are concerning. While we can’t be sure that these links are causal, the relationships are similar in importance to those found in physical health outcomes, such as respiratory illnesses.

“The government must do more to address air pollution in the UK as exposure to it is potentially harming people’s physical and mental health.”


‘Air pollution (particulate matter) exposure and associations with depression, anxiety, bipolar, psychosis and suicide risk: a systematic review and meta-analysis’ by Isobel Braithwaite et al. was published in Environmental Health Perspectives at 14:00 UK time on Wednesday 18th December. 


Declared interests

Dr Katharina Janke: I don’t have any declarations. My only source of income is the salary Lancaster University pays me. I am connected to the EPSRC-funded Data Science of the Natural Environment project and the Centre for Health Inequalities Research at Lancaster University.

Dr Ioannis Bakolis: No conflicts of interest.

None others received.

in this section

filter RoundUps by year

search by tag