Air pollution disease and deaths: Don’t fight over numbers – fight the problem

The operative word in the minister’s response was “direct correlation” as she went on to elaborate that these diseases (cardiac conditions, cancers etc.) have multiple causal factors including tobacco, alcohol, diet, hereditary, and occupation and therefore one cannot lay the blame only on air-pollution. This assertion would be true for estimates for any of the above-mentioned risk factors including tobacco use. But nobody questions the tobacco-related disease burden estimates. Why are those estimates treated as acceptable and one on air pollution questioned?

Both tobacco smoke and air pollution consist of multiple chemicals and pollutants which can lead to multiple diseases. One understands that intuitively, exposure to tobacco is perceived to be at the personal level while exposure to air pollution is at the population level. This does introduce additional epidemiological and statistical challenges. Also, while the strength of association is much stronger for tobacco than air pollution, the exposure is at a much larger level for air pollution. There is no acceptable scientific range for exposure to tobacco, while for air pollution we have some degree of acceptance, given our current state of development.

Epidemiologists routinely account for other risk factors when they look at association with a disease for a specific risk factor. There are statistically robust methods for estimation of the fraction of a disease or death burden “attributable” to a particular risk factor. Thus, while individual deaths cannot be attributed to air pollution, or for that matter tobacco, as a “direct” causative agent, morbidity and mortality at population level can be ascertained. Air pollution as a population-level cause of cancers and cardiovascular disease satisfies the epidemiological criteria of causation in terms of dose-response (higher the pollution level, more the health effect), biological plausibility (the cellular mechanism action of the pollutant chemicals known), consistency (multiple studies done in different populations show similar findings), temporal relationship (exposure precedes the disease). Thus, from an epidemiological point of view, there is no doubt that deaths attributable to air pollution can be estimated.

This does not mean that I am endorsing the above estimates. A critical appraisal of that estimate should occur in the academic domain. All estimates can be criticised as they invariably involve some assumptions and generalisations. As in the case of other public health problems, there will be many researchers who will come up with different estimates using different methods. The differences are because of inherent statistical treatment as well as data challenges.

It is the government’s responsibility to make available a “true” estimate of any public health problem or at least provide the data that is required for such estimates to be generated. This is to ensure that policy development is based on good science and not to please the global academic world. This is also important for advocacy and for evaluating interventions in terms of prevention of disease and deaths. The government could generate these estimates itself and put them in public domain. Experts would then be free to look at it and comment, suggest improvements which can be handled in the next iteration of estimation. There should be a standing mechanism for data collection, collation and estimation for major public health problems as this is a continuous cycle. This is one investment the state and central governments must make.

The minister listed many steps taken by the government to show that it is cognisant of the problem and is addressing it. Then why question the estimate? It appears that the government is ready to acknowledge that air pollution is a problem that it is addressing, yet not ready to put a number to it. Why not look at an estimate as the first step in our understanding and addressing a problem rather than as an admission of failure to address it? There is no doubt that air pollution, like malnutrition, is a complex public health challenge that requires an all-of-government, all-of-society response. Having an estimate of disease burden is the starting point to get everyone on board (so that they collude rather than collide). The availability of periodic estimates enables us to understand how we are doing, where we have failed and modify our strategies accordingly.

One understands that public perception is very important in politics but so is performance. Instead of fighting over numbers, let us fight the public health challenges together.

The writer is a professor at the Centre for Community Medicine at the All-India Institute of Medical Sciences, New Delhi. Views expressed are personal