Union Environment Minister Prakash Javadekar said last week that no Indian studies have shown a “direct correlation” between pollution and mortality. International studies estimating that thousands have died from causes linked to air pollution have caused a “fear psychosis among people” when the situation is actually not so bad, Mr. Javadekar said.

Why are his comments untenable?

Last year, the India State-Level Disease Burden Initiative (ISLDBI), which consists of at least 100 health professionals, reported that one in eight deaths in India were attributable to air pollution and that “…the average life expectancy in India would have been 1.7 years higher if the air pollution levels were less than the minimal level causing health loss.” ISLDBI studies are funded by the Union Health Ministry and involve the Public Health Foundation of India, the Indian Council of Medical Research, and the Institute for Health Metrics and Evaluation. These studies were part of the Global Burden of Disease Study 2017 and were published in the peer-reviewed journal, The Lancet Planetary Health.

How is mortality from air pollution calculated?

Researchers overwhelmingly rely on modelling. Epidemiologists and public health professionals routinely rely on correlation to draw countrywide estimates of the health risk posed by a particular pollutant or any risk factor that they’d like to investigate. In principle, this isn’t different from estimating what level of temperature rise in the oceans can lead to increased cyclones.

In the case of pollution studies, first, data on emissions in an area are collected. While there are a range of pollutants, the bulk of contemporary research interest lies in measuring PM 10 or PM 2.5 levels. A study’s ambition depends on how fine-grained the emissions data are. The data can be from a city, industrial regions within a city, a State, residential areas or commercial pockets. It also depends on the sophistication of the sensors used to measure the level of pollutants used. This is influenced by the budget. Invariably, the exercise will involve averaging the emissions in a region to a much larger space.

The next step is to collect data on hospital admissions for respiratory and cardiovascular diseases, cancer and associated mortality. Then exposure-response curves are drawn that show how particulate matter concentrations relate to death and disease prevalence. The other category of investigations involve regularly monitoring the pollution levels people are exposed to over time and recording mortality levels. This is a more time-consuming approach but considered more ideal to single out the effects of pollution on death rates.

What kind of evidence exists to determine mortality from pollution?

The bulk of studies done to gauge exposure response have been conducted in the U.S. and Europe, in cities that have, on average, good air quality. This is because an increase in particulate matter concentrations, above background levels, can be more reliably estimated and correlated to the rise in mortality (gauged from hospital records). Most such studies have found a linear relationship between mortality and PM 10 levels. The caveat is that on average those observed were exposed to less than 10 micrograms per cubic metre. Beyond a certain level, the response “flattens out” and it’s hard to estimate if concentrations, say, 10 times more, would lead to a mortality spike 10 times more. That’s the kind of information relevant for India because background concentrations are much higher than in Europe. The ISLDBI relies on these computed exposures, calculates the pollution levels in various States, finds how many Indians may have been exposed, and computes a death rate. It also adjusts for other causes of mortality.

Are there exposure studies being done in India?

India has embarked on a 20-city plan to calculate exposure levels among Indians, including pregnant women. These are funded by the Union Environment Ministry and are expected to be published next year.

Source: The Hindu

Relevant for GS Mains Paper III; Environment