Air pollution alerts without enforceable actions may be inadequate in protecting public health

Toronto has low to moderate air pollution levels, but spikes in air pollution remain common. On these days, authorities issue air quality alerts to warn the public and encourage people to reduce physical activities outdoors. The programme is based on information campaigns (web notifications and media coverage), rather than enforceable public actions. This type of programme is used elsewhere where air pollution levels are also similar, including the UK, many European countries and the USA. However, there is little evidence on whether or not these types of alerts have led to improvements in public health.

The study authors suggest that air pollution problems may be best addressed through collective and enforceable actions. These could include improved urban and transportation planning, and improved fuel standards and emission control (especially targeting power generation and industries that contribute significantly to air pollution), rather than advising individuals and leaving to them to protect themselves from the harmful effects of air pollution.

"In recent decades, ambient air quality has improved in many regions, but occasional spikes in air pollution remain common in high-income countries. These increases can trigger many adverse health effects, and so, in response, government agencies have implemented air quality alert programmes to inform the public of the potential dangers," says lead author Dr Hong Chen, Public Health Ontario, Canada. "These programmes play a central role in the public response to air pollution, but little is known about their effectiveness in reducing health risks from air pollution. Our study is the first to comprehensively evaluate their effectiveness."

The study tracked the health records of around 2.6 million people who lived in Toronto between 2003 and 2012, and monitored air quality and when air quality alerts were issued over the same period. Combining these data, the authors estimated deaths from cardiovascular or respiratory disease, and hospital admissions for asthma, heart attack, heart failure, stroke, or COPD on days with high pollution levels when air quality warnings were and were not issued. This helped them to model the effect of the alert being issued on days of similar pollution levels.

The study did not monitor whether people reduced their outdoor physical activity as recommended by the air quality alerts.

Alerts were issued when air quality was expected to be poor (having a daily air quality index [AQI] of 50 or more), and if there were likely to be persistent and widespread levels of ozone, fine particulate matter, or both. However, unexpected changes to air quality resulted in some forecasts not being accurate, meaning that sometimes alerts were not issued when pollution exceeded trigger level, or that alerts were issued when levels of pollution ended up being below trigger level.

Between 1 January 2003 and 31 December 2012, there were 41 days when the air quality was above the AQI trigger level, and alerts were issued on 62 days.

The authors modelled the available data to estimate the potential effects on days when pollution levels were above trigger levels and alerts were issued. They estimated that alerts were associated with a reduction in emergency department admissions for asthma by around 25% -- equivalent to 4.73 fewer asthma-related admissions per million people per day. However, the study found no effect of air quality alerts on reducing deaths from cardiovascular or respiratory disease, nor on hospital or emergency department admissions for cardiovascular disease.

The authors suggest that this may be because people with these conditions are likely to spend a lot of time indoors already, meaning the recommendation to stay indoors may have less of an effect on them. In which case, enforcing air quality improvements alongside the alerts may be important to benefit everyone at risk from polluted air.

They also say that measures focusing only on days with the highest levels of pollution are unlikely to help address the most harmful effects of air pollution, given the role of long-term air pollution exposure as a risk factor for cancer and for cardiovascular and respiratory diseases.

"Air pollution is a societal problem that can be addressed most effectively through enforceable measures that reduce pollutants in the air we breathe every day, not just on days when air pollution is at its highest. Globally, air quality alert programmes represent one of the most common public responses to protect people from air pollution, but the findings of our study show that air quality alert programmes offer inadequate protection for public health. This highlights the need for implementing enforceable public actions to reduce air pollution." says Dr Chen.

The authors cite Santiago (Chile) where air pollution is much higher than Toronto and enforceable measures to reduce air pollution (such as temporarily restricting driving, closing factories or restricting burning of fossil fuels) are used on alert days. A previous study suggests that these measures may help to prevent 20 deaths per million people per day, and the measures lead to a 20% reduction in air pollution on alert days, compared to similarly polluted days without alerts.

The authors note some limitations, including that the study only includes the most severe outcomes during high pollution, as it does not take into account small changes in symptoms caused by pollution that would not result in people going to a hospital. The authors also flag that they cannot rule out the possibility that the results for asthma hospital admissions are due to chance, given the number of relationships studied in their research.

While it is unclear whether these results from a country with low to moderate air pollution levels can be applied to countries with severe air pollution such as China and India, the authors note that the findings could be relevant for cities in many parts of Europe and North America.

Writing in a linked Comment, Dr Jonathan Samet, Colorado School of Public Health, USA, says: "An air quality alert has potential public health benefits through a chain of possible responses that reduce exposure... In Beijing, for example, the most severe episodes lead to a red alert with mandatory provisions, including suspending primary and middle schools, banning use of specific vehicles, stopping some outdoor construction work, closing predesignated industrial facilities, and banning fireworks. At times of high pollution, individuals might also choose to use personal respiratory protection and air cleaning devices in their residences... If air quality alerts are to work as intended for individuals, awareness of the warnings and adherence to the protective recommendations are essential steps in the chain."

He adds: "The findings of Chen and colleagues should be generalised with caution, in view of the study population from Toronto, which has good access to medical care. Moreover, estimates using the regression discontinuity design have applicability around the threshold assessed but cannot be extended readily to the far more severe air pollution now affecting much of the world... Air pollution alerts remain fundamental to public health protection, particularly in large urban areas where air pollution concentrations routinely spike well beyond air quality standards and guidelines."

This study was funded by Public Health Ontario and the Canadian Institutes for Health Research. It was conducted by researchers from Public Health Ontario, Institute for Clinical Evaluative Sciences, University of Toronto, McGill University, Ontario Ministry of the Environment and Climate Change, University of California.