Heart and lungs

Air pollution linked to increased hospital admissions for kidney disease and sepsis

"Impact of air pollution on health may be far worse than thought," reports The Guardian.

Air pollution by what is known as fine particulate matter has been linked to increased risks of several conditions, including heart attacks, strokes and breathing difficulties.

Fine particulate matter are tiny particles of solid and liquid pollutants smaller than 2.5 micrograms. It's also known as PM2.5. These tiny particles are thought to be harmful because they can be breathed into the lungs and absorbed into the blood.

Researchers examined more than 95 million hospital admissions in the US for people aged over 65. They found a rise in PM2.5 not only increased the chances of admission for heart and lung conditions, but for other conditions not previously linked to air pollution. These included sepsis, urinary tract infections and kidney failure.

The increase in the chance of hospital admission is very small for any individual. However, researchers estimate a very small increase in PM2.5 in the air could be linked to an additional 5,692 hospital admissions and 634 deaths in the US each year.

We cannot be sure that pollution was the direct cause of the increased hospital admissions, owing to the type of study carried out. Yet the research does contribute to evidence that air pollution has a negative effect on health, and it may inform international guidelines on air pollution that are currently being updated.

Where did the story come from?

The researchers who carried out the study were from Harvard TH Chan School of Public Health in the US, Tsinghua University in China and the Swiss Data Science Centre in Switzerland.

The study was funded by the US National Institutes of Health and the US Environmental Protection Agency. It was published in the British Medical Journal on an open-access basis so it is free to read online.

You can read the study on the BMJ website.

The Times, The Guardian and Mail Online all published reasonably accurate and balanced reports of the study.

What kind of research was this?

This was a case crossover study.

This is where people's exposure to a risk factor (in this case PM2.5) is measured on the day before and the day of the outcome (hospital admission). This is then compared to their exposure on other "control" days (in this case, the same day of the week in the rest of that month and year).

This type of study is useful for spotting links between risk factors, such as air pollution, and outcomes, like hospital admission, but it cannot tell us what is behind any links.

What did the research involve?

Researchers used a large database of everyone admitted to hospital under the Medicare scheme for people aged over 65 in the US, from 2000 to 2012. This was a total of 95,277,169 admissions.

They matched people's postcodes to the nearest estimated PM2.5 measures to their home for the day before, and the day of, their hospital admission. These estimates were within areas of 1km².

The diagnosis given for the admission was categorised into 1 of 214 disease groups.

Researchers then looked at PM2.5 levels at each person's home for the same day of the week, in the same month and year as the hospital admission. For example, if a person had been admitted on a Tuesday in January 2010, the researchers would check PM2.5 levels for that Monday and Tuesday, then for the other Mondays and Tuesdays in January 2010, for comparison.

They then used the data to find out how the risk of being admitted to hospital changed, according to an increase in PM2.5 of 1 microgram per cubic meter.

In addition to hospital admissions, the researchers calculated the number of days in hospital, number of deaths and cost of healthcare associated with the admissions. They then calculated how these were linked to PM2.5 levels.

What were the basic results?

Similar to previous research, the researchers found a link between increased PM2.5 levels and an increase in risk of hospital admissions for heart failure, pneumonia, chronic obstructive pulmonary disease (COPD), heart attack, Parkinson's disease, diabetes complications and several other conditions previously liked to pollution.

However, they also found increased admissions for other relatively common conditions that have not previously been linked to PM2.5 pollution. These included:

  • sepsis
  • imbalances in fluid and electrolytes (such as too much or not enough calcium or salt in the blood)
  • kidney failure
  • urinary tract infections
  • skin and tissue infections

For every 1 microgram per cubic metre increase in PM2.5, the risk of being admitted to hospital for 1 of the conditions identified as linked to pollution increased by 0.02 to 0.68 admissions per 10 million each day, depending on the condition. That means the chance of admission is 0.05% to 0.40% higher for these conditions, compared to a day when PM2.5 pollution is lower.

Because the individual numbers are so small, these figures can be hard to imagine. The researchers calculated that each year in the US, where the population is around 330 million, a short-term rise of PM2.5 levels of 1 microgram per cubic metre corresponds to an additional:

  • 5,692 hospital admissions
  • 32,314 days in hospital
  • 634 deaths
  • $100 million in healthcare costs

The researchers did a separate analysis looking only at data where the PM2.5 levels were below the current World Health Organisation (WHO) guidelines for air pollution of 25 micrograms per cubic meter over a 24-hour period. They said that even at these levels, they found an increased risk of hospital admission linked to each 1 microgram per cubic metre rise.

How did the researchers interpret the results?

The researchers said their results "discovered several new causes of hospital admissions associated with short-term exposure to PM2.5 and confirmed several already known associations".

They added that the findings "provide timely evidence for the revision of WHO guidelines" on air pollution.

Conclusion

It is no surprise that air pollution may be linked to health problems, and we already know of several conditions where that link has been made. This study suggests that PM2.5 pollution may be linked to many more conditions than had not been linked to air pollution previously.

However, it is important to keep this in perspective. The risks to any 1 person from a small rise in pollution, as uncovered in this study, are very small. The biggest increases in hospital admissions were from heart failure, which is already known to be linked to PM2.5 levels. These rose by 0.68 admissions for every 10 million people at risk. Risks of being admitted to hospital for skin infections, for example, rose by only 0.13 admission for every 10 million at risk.

We also need to be aware of limitations to the study.

It does not tell us if the conditions identified are directly linked to pollution, or whether other factors are involved. For example, activities such as smoking, drinking alcohol and physical activity could also trigger hospital admissions for some conditions, and these might vary according to air pollution levels. It is also an observational study, so it has the usual limitations of being unable to prove cause and effect.

The study only looked at people aged 65 or older, so we do not know how PM2.5 affects children or younger adults. It is also possible that not all the admissions data was coded correctly for diagnosis.

The study provides more evidence that PM2.5 pollution – which mainly comes from vehicle emissions and burning fossil fuels – is bad for health. While it may not be something to worry about on an individual level, the study adds to the reasons to reduce air pollution.


NHS Attribution