Research and analysis

Working paper 1. Susceptibility of population groups to air pollution

Published 18 March 2025

Summary

One of the considerations of the Committee on the Medical Effects of Air Pollutants’ (COMEAP) AQIS Sub-group is whether the Air Quality Information System (AQIS) and Daily Air Quality Index (DAQI) can be tailored to an individual’s needs depending on their susceptibility to air pollution. Susceptibility can be defined as being more at risk of experiencing adverse effects of air pollution due to an intrinsic factor, for example, having a health condition or a genetic variation.

This working paper examines evidence for susceptibility of different groups to adverse health effects of air pollution. It principally focuses on four pollutants, Particles less than 2.5 micrometres (PM_2.5), Nitrogen Dioxide (NO₂), Ozone (O₃) and Sulphur Dioxide (SO₂), and 5 susceptibility categories, namely:

• pre-existing respiratory diseases
• pre-existing cardiovascular disease
• diabetes/metabolic syndrome
• children
• older adults

It also touches on obesity and maternal susceptibility during pregnancy. Evidence examining susceptibility to the effects of long- and short-term exposure were reviewed for this working paper.

The DAQI is intended to provide advice to enable individuals to reduce the likelihood of experiencing risks from short-term exposure to pollutants. Examination of the evidence suggests that individuals with respiratory or cardiovascular disease, and older individuals, are likely to be more susceptible to short-term exposure to air pollution than healthy adults. There is also adequate evidence to suggest that children with asthma will be particularly at risk of effects of short-term exposure to air pollution. These conclusions are consistent with the groups regarded as being at risk in the advice accompanying the current DAQI.

The available evidence suggests that individuals with obesity or metabolic syndrome or diabetes may also be at higher risk of the effects of air pollution than non-obese healthy individuals. However, we consider that the evidence is not strong enough to include people with these conditions in ‘at-risk’ groups for DAQI advice at present. There is little evidence relating to risks of short-term exposure during pregnancy.

Further research is required to establish whether more nuanced recommendations can be made for those who are potentially more at risk of the effects of air pollutants. This includes establishing, more specifically, which health conditions are at greater risk (for example, subtypes of asthma, specific cardiovascular conditions) and which specific pollutants are of greater concern. Future studies should also investigate whether diseases other than non-cardiorespiratory diseases engender greater susceptibility to air pollutants and examine whether medical treatment could offer protection from exacerbation of disease during heightened air pollution episodes. Studies should also be designed so that susceptible groups can be compared to the general population, so that differences in exposure-response relationships can be identified between these groups. We also recommend future research on effects of short-term exposure to air pollution on the health of the mother during pregnancy.

Introduction

Air pollution affects everyone, both in terms of the ubiquity of exposure and the health of all humans being affected. However, it is clear that some individuals are more at risk from the effects of air pollution than others. This could be because of an extrinsic factor, for example, they are exposed to much higher levels of air pollution, or they are exposed to specific air pollution as part of their lifestyle or occupation. Or they could be ‘susceptible’ because of an intrinsic factor, for example, they have a biological reason why they are at a greater risk (such as they have a health condition or a genetic variation) (USEPA, 2015; Royal College of Physicians and Royal College of Paediatrics and Child Health, 2016).

Groups that are generally regarded to be more susceptible to air pollution are:

• pregnant mothers, due to the increased stress on the mother’s body and the vulnerability of the unborn baby (the developing fetus)
• children, due to their bodies growing and developing, their immune systems not being fully developed, and their breathing characteristics (for example higher ventilation rate and inhaled dose per lung surface area compared to a healthy adult) - they may also be more exposed because of the greater time, on average, spent outdoors
• older people, due to the potential for a compromised capacity of the immune system and the body’s defences to respond to air pollution, and increased likelihood of undiagnosed disease status (including asymptomatic impairments in physiological function), as well as potential reductions in lung volume
• those with certain health conditions, such as a respiratory or cardiovascular condition

This working paper is part of COMEAP’s input to the on-going review of the Air Quality Information System (AQIS) being undertaken by a steering group convened by Defra and UKHSA; one of the considerations of which is whether the AQIS and DAQI can be tailored to an individual’s needs depending on their level of exposure or susceptibility. This working paper largely focuses on susceptibility, with exposure being addressed elsewhere in the AQIS review. Due to the volume of literature on this topic, and the timescales of the AQIS project, the working paper is based on initial conclusions from 2 published sources:

• COMEAP’s statement ‘Advice on the health evidence relevant to the setting of the PM_2.5 targets (COMEAP, 2021)

• US EPA’s Integrated Science Assessments:

  1. Particulate Matter, 2019 and 2022 (US EPA, 2019a; 2022)
  2. Nitrogen oxides, 2016 (US EPA, 2016)
  3. Ozone and photochemical oxidants, 2020 (US EPA, 2020)
  4. Sulfur Oxides, 2017 (US EPA, 2017)

It also draws on a report by the Environmental Research Group (ERG), Imperial College, London, regarding the susceptibility of different groups to the effects of short-term exposure to air pollutants (Evangelopoulos and co-authors, 2023). This report was commissioned by Defra to inform the AQIS review.

COMEAP’s statement ‘Advice on the health evidence relevant to the setting of the PM_2.5 targets’

COMEAP’s statement ‘Advice on the health evidence relevant to the setting of the PM_2.5 targets’ provides advice to Defra on the setting of new air pollution targets for England. Because it is not focusing on individuals, it does not specifically explore at-risk groups in any detail, however, it refers to this in places. The key relevant comments are quoted below.

Groups who may be particularly sensitive to short-term peaks of air pollution… may include, for example: asthmatic children, those with underlying heart and lung conditions and older people

Some risk factors may be less obvious; for example genetics, lifestyle choices or co-exposures to other pollutants may mean that some individuals have lower antioxidant capacity to combat adverse effects of exposure. Other vulnerable life stages likely include pregnancy and early childhood, when the body is developing.

​In relation to individuals living near roads:

A potential role for particulate pollutants acting as adjuvants to enhance allergic sensitisation”​ was noted in the context of induction of new cases of asthma.

​Concentration-response functions for susceptible groups are not as well established as those for the general population (which do include the susceptible groups, as well as individuals without underlying health conditions). It is also difficult to envisage reduction strategies that would specifically deliver greater reductions for susceptible groups, as they are very dispersed geographically​.

From the REVIHAAP project (2013):

Susceptible population groups and effect mechanisms differ for short-term and long-term exposures …. Even apparently healthy people are susceptible to the effects of long-term exposure to PM, because exposure can potentially accelerate progression of a disease….until it is clinically diagnosed. Most susceptible to the effects of short-term exposures are those with an unstable disease. Different mechanisms may account for the progression of a disease and triggering acute exacerbation of diseases.

Both long-term and short-term exposures affect health but likely in different ways. People with pre-existing disease are likely to be most sensitive to effects of short-term exposure, including effects that might not have occurred without peaks of elevated concentrations. Long-term exposure likely has the potential to affect everyone, by contributing to the initiation or progression of disease.

There is also some commentary on occupations with higher exposure, socioeconomic status, ethnicity, obesity, and so on, but these are not a primary focus of this working paper.

Overall, the COMEAP statement re-iterates the main groups susceptible to air pollution, highlights some of the biological mechanisms that may account for the greater sensitivity to pollutants, and briefly raises the challenges of considering individuals in population level guidelines and interventions.

US EPA’s Integrated Science Assessments

Susceptibility

The US Environmental Protection Agency (US EPA) Integrated Science Assessments (ISAs) are very comprehensive reports on the evidence on health effects of air pollution. The reports cover many different aspects. However, this working paper is based only on the chapters relating to specific groups or life-stages that may be potentially at increased risk. In the ISA for ozone (US EPA, 2020) this is included within the Integrated Synthesis.

It is worth highlighting several aspects of these reports:

• the reports have assessed the epidemiological evidence, occasional controlled human exposure studies and selected toxicological evidence from experimental studies in rodents

• the reports consider differences in exposure and dosimetry (biological factors), but mostly differences in relative risk independently of these (studies directly comparing 2 groups, or stratification by group)

• susceptible groups considered in the ISAs are:​

  • pre-existing disease - respiratory diseases (asthma, chronic obstructive pulmonary disease (COPD)) and cardiovascular disease including hypertension ​
  • risk factors - obesity, genetic factors, (high cholesterol)​
  • lifestage - children, older adults​
  • demographic - socioeconomic status, gender, ethnicity​
  • location - urban vs rural, proximity to traffic​
  • behaviour and other factors - for example, smoking, diet

• the extent and consistency of evidence are used to characterise the evidence for potential at-risk factors​ (US EPA Preamble to the Integrated Science Assessments, 2015) as:

  • evidence of no effect​
  • inadequate evidence​
  • suggestive evidence​
  • adequate evidence​

• the US EPA specifies studies which are short-term (hours up to 1 month) exposure versus long-term (1 month to years) exposure, however, the characterisation of evidence of increased susceptibility is a single rating based on both

• the emphasis is on studies where there is a comparison of a susceptible group with a reference group (such as healthy adults)

A detailed narrative is provided for each of the susceptibility categories. To aid our consideration of this issue, the information in these sections has been tabulated below for the main categories of susceptibility.

Table 1: Summary of factors that may cause a greater susceptibility to PM_2.5, based on the conclusions of the US EPA ISA report

Table 1 notes: COPD = chronic obstructive pulmonary disease, CRP = C-reactive protein, CVD = cardiovascular disease.

Susceptibility Category	Overall characterisation of evidence	Differences in exposure/ dosimetry	Differences in risk of mortality	Differences in risks of specific health categories	General comments
pre-existing respiratory diseases	suggestive	Differences in dosimetry: greater deposition? impaired clearance?	A very limited number of studies suggest respiratory conditions may predispose to some cause-specific mortality associated to long-term exposure to PM2.5.	Asthma predisposes to various subclinical respiratory parameters (for example, symptoms, lung function) linked to PM2.5. COPD does not predispose to risk of CVD.	Notes good evidence for effects of PM2.5 in asthma and COPD, even if studies do not specifically compare risk to a healthy population.
pre-existing cardiovascular disease	suggestive	Not commented on.	Not all CVD imparts a greater risk to health effects – varies with condition and studies are few in number. The evidence is strongest for hypertension increasing risk.	Effects of hypertension (and coronary heart disease, to some extent) are clearer on subclinical endpoints, for example, blood vessel relaxation or blood cytokines.	1 controlled exposure study, but few effects were observed in healthy volunteers or the CVD group. Some evidence from hypertensive rats, although there is a lack of direct comparison with healthy animals.
diabetes/ metabolic syndrome	Inadequate evidence	Not commented on.	Too few studies to determine if diabetes increased the risk of mortality linked to PM2.5.	Some evidence for greater changes in blood pressure, atherosclerosis and biomarkers (for example, CRP), although few in number or inconsistent.	Many studies had a small proportion of diabetics or did not consider pre-diabetes.
children	Adequate evidence	Children may have higher exposure.	Stratified analysis for mortality is not addressed, but suggests similar magnitude of associations for children and adults for endpoints such as hospital admissions.	Evidence for child-specific outcomes, for example lung function growth, lung development, asthma.	A limited number of studies suggest ages 1-5 may be a critical window for experiencing respiratory effects following short-term PM2.5 exposure.
older adults	Inadequate evidence	May have higher dosimetry due to impaired clearance of particles.	Stratified analyses suggest similar magnitude of associations in older and younger adults.	More consistent links between PM2.5 and CVD in older adults, but comparisons with different age groups are inconsistent.	Some controlled exposure and animal data to suggest this group may be at more risk, although direct comparisons between different age groups are limited.

The table summarises evidence from both short- and long-term studies. Pre-existing respiratory disease includes chronic obstructive pulmonary disease (COPD) and asthma. Pre-existing cardiovascular disease includes coronary heart disease, hypertension, coronary artery disease, myocardial infarction, stroke, ischemic heart disease, congestive heart failure.

Table 2: Summary of the characterisation of the evidence for susceptibility to PM_2.5, NO₂, O₃ and SO₂, based on the conclusions of the US EPA ISA report

Table 2 notes: COPD = chronic obstructive pulmonary disease.

Susceptibility Category	PM2.5	NO2	O3	SO2
pre-existing respiratory diseases	suggestive (asthma and COPD)	asthma: adequate COPD: inadequate	asthma: adequate COPD: inadequate	asthma: adequate COPD: inadequate
pre-existing cardiovascular disease	suggestive (asthma and COPD)	inadequate	inadequate	not specifically assessed in the susceptibility section
diabetes/ metabolic syndrome	inadequate	inadequate	inadequate	not specifically assessed in the susceptibility section
children	adequate children-specific effects, for example lung development	adequate (asthma exacerbation)	adequate	suggestive (respiratory, mostly asthma)
older adults	inadequate	adequate (mortality, respiratory)	adequate	suggestive (respiratory)

The table summarises evidence from both short- and long-term studies.

Observations on the tables

Many of the conclusions drawn by the US EPA are based on a very limited number of studies. While there is usually strong evidence that a pollutant (notably PM_2.5) has effects in susceptible individuals, there are few studies that directly compare those who may be susceptible with healthy individuals (or the general population). Therefore, determining whether the magnitude of risk varies between the groups is challenging. While all of the categories examined in the ISAs are considered to be potentially susceptible populations, for many of these groups the evidence is actually deemed inadequate.

In some cases, the basis for the characterisation of the evidence for groups being at risk includes consideration of extrinsic risk factors (such as children spending more time spent outdoors), or biological factors affecting the dose of exposure/clearance of pollutants from the body (for example, impaired lung clearance of particles in older adults). Effects which are only relevant at specific life-stages (such as lung development in children) are also taken into account.

There are ‘sub-categories’ of susceptibility that are important. For example, hypertension may predispose a person to mortality linked to air pollution. However, across a range of other cardiovascular conditions or risk factors there was not consistent evidence for cardiovascular disease predisposing a person to these effects. Examination of increased risks is complicated due to uncertainty in disease diagnoses and variability in disease status.

There are differences in the characterisation of evidence for the various at-risk groups between the different air pollutants. This will be important for the AQIS project, given that the DAQI is based on the pollutant corresponding to the highest index band. The potential differential susceptibility of sub-groups within the population might depend upon which pollutant is responsible for the elevated DAQI band for that day/episode.

Comparisons of evidence for susceptibility to effects of long-term and short-term exposure to air pollutants

The US EPA ISAs report health effects in susceptible groups (people with asthma, cardiovascular diseases, reproductive/developmental, children, older adults, obese (body mass index (BMI) above 30), overweight (BMI 25 to 30)) following exposure to air pollution (PM_2.5, NO₂, O₃ or SO₂). These documents also summarise the evidence for various groups being more susceptible to the adverse effects of air pollutants, compared with the general population. The categorisation of evidence for susceptibility draws on evidence from studies of short-term temporal variations in exposure (such as epidemiological time-series studies) and long-term spatial differences in exposure (such as cohort studies), and therefore does not distinguish between susceptibility to effects of long- and short-term exposures. This distinction becomes important when considering advice associated with the DAQI, which is concerned with effects of short-term exposure.

For children, evidence of adverse health impacts of air pollution exposure comes from both short- and long-term exposure studies. In long-term epidemiological studies looking at children, exposure to PM_2.5 was associated with impaired lung function and asthma development (US EPA, 2019a). Long-term NO₂ exposure has been associated with asthma development, and evidence on O₃ suggests that early life exposure could contribute to airway obstruction and increased airway responsiveness (US EPA, 2016; 2020). Evidence suggests that short-term increases in ambient PM, NO₂ O₃ and SO₂ concentrations are associated with increases in asthma-related hospital admissions in children (US EPA, 2016; 2017; 2019a; 2020). Emerging evidence from two studies in China (one on asthma, one on pneumonia) suggests a critical window for children to experience respiratory health effects following short-term exposure to PM_2.5may be between ages 1 to 5 years (US EPA, 2019a).

Similar to other approaches that attempt to explain the human health response to air pollution exposure, we note that, in studies looking at susceptible groups and long-term exposure to air pollution, there are many factors and confounding variables that would need to be taken into account. This includes errors in exposure assessment, diet, geography, age, health status, genetic susceptibility factors, physical activity (status), employment (status), environment, and cultural habits.

Summary of Environmental Research Group report on susceptibility commissioned for the AQIS Steering Group

As part of the AQIS review, Defra commissioned reviews on several topics, one of which was susceptibility to the effects air pollution, which was undertaken by a team of researchers within the Environmental Research Group (ERG) (Evangelopoulos and co-authors, 2023).

General comments on the ERG report

The report is a very comprehensive assessment of the research literature of the following susceptibility categories:

1. diabetes
2. obesity
3. children
4. older adults
5. sub-types of asthma
6. maternal effects in pregnancy

The literature is large, and the report provides helpful summaries and an overview table at the end of each section. There are useful discussion points in the final conclusion sections, for example, defining susceptible groups, spatial differences in different air pollutants, possible caveats for air pollution messaging and future research directions. It is worth highlighting that the ERG report did not assess pre-existing pulmonary disease (other than looking for the distinction between different subtypes of asthma) or cardiovascular disease as susceptibility categories as these groups were considered to be already identified in the existing DAQI. The report did not consider cognitive diseases, mental health, or emerging work on multi-morbidity patterns. The ERG report also provides summaries of the US EPA ISAs which amalgamate evidence regarding susceptibility from long-term as well as short-term studies, whereas the ERG report focused on short-term studies alone.

ERG report conclusions: Shortened summary sections by individual susceptibility factors

Diabetes

The ERG report comments on the US EPA ISAs: largely suggest that the evidence is inadequate for diabetes conferring susceptibility to pollutants.

Epidemiological data: evidence from 2 studies suggested that the presence of diabetes increased the risk of all-cause mortality following short term exposure to particulate matter but there were no associations found for gaseous pollutants. Diabetes mortality was found to be associated with short-term exposure to ambient air pollution, but it is unclear whether this indiciates a unique susceptibility or is mediated by the already established links with cardiovascular disease. For hospital admissions, limited statistically significant evidence prevents any definitive conclusions about increased vulnerability in those with diabetes, however, in some studies there was evidence for effect modification of diabetes for myocardial infarction, cardiac arrythmia, and respiratory diseases for PM_2.5, and cardiac arrythmia and cardiovascular disease for PM₁₀.

Panel studies: the data is inconsistent for inflammation. There are occasional specific links between various air pollutants and biomarkers (von Willebrand factor (vWF), heart rhythm for various pollutants) that are influenced by diabetes.

Controlled exposure studies: evidence for no increased susceptibility from diabetes.

Animal studies: some specific links suggesting diabetes may confer susceptibility.

Overall, there is limited evidence for PM and a lack of evidence for gases. It does not appear, based on a lack of literature and data, that diabetes should be considered as a susceptibility factor for effects of short-term exposure at this time.

Obesity

The ERG report comments on the US EPA ISAs: state there is suggestive evidence for obesity conferring susceptibility to PM and ozone.

Epidemiological data: no studies.

Panel studies: obesity may influence some specific links for ozone and PM₁₀ and health parameters such as atrial fibrillation, lung function.

Controlled exposure studies: some evidence for heart rate variability (HRV), vasodilation, lung function.

Animal studies: obese animals may show greater susceptibility to air pollution in terms of effects on, for example, airways resistance and inflammation.

Overall, there may be some links between obesity and susceptibility to ozone, especially regarding respiratory effects. In general, though, there does not currently appear to be sufficient evidence of increased susceptibility to air pollutants from obesity for this to be included in the DAQI.

Children

The ERG report comments on the US EPA ISAs: propose that the evidence is adequate that children are more susceptible to air pollution.

There is evidence that PM_2.5 and NO₂ may have a stronger association with exacerbation of asthma in children than in adults.

Overall, children with asthma should continue to be regarded as an at-risk group for the DAQI.

Older adults

The ERG report comments on the US EPA ISAs: propose that there is adequate evidence for greater susceptibility of the elderly to NO₂ and O₃.

The evidence is adequate to suggest that older populations are more susceptible to PM-related health effects, for example, mortality.

There is suggestive evidence for an increased asthma hospitalisation in response to PM_2.5 and NO₂ in older populations.

Controlled exposure and animal studies generally suggest older individuals/animals are more susceptible to air pollutants.

Overall, older adults should continue to be regarded as an at-risk group.

Asthma subtypes

There is extremely limited evidence looking at the effects of air pollution in sub-types of asthma and childhood wheeze.

The document discusses that sub-types of asthma are defined differently in different studies, for example, allergic and non-allergic asthma, mild intermittent and mild persistent asthma, and early and late-onset asthma.

Air pollutants may enhance responses to allergens, which could influence susceptibility in people with certain types of asthma.

Overall, there is not enough research to draw conclusions on differential susceptibility.

Pregnant mothers

The ERG report assessed a review of medium- to long-term (for example, trimester-specific and whole-pregnancy) exposure to air pollution and maternal outcomes such as gestational diabetes and pre-eclampsia, and a study on long-term exposure to NO₂ and maternal outcomes. They conclude that the evidence was “suggestive”. This has also been reviewed by Mazmuder and others (2024).

There was only a single study on short-term exposure (the focus of the ERG report), finding a link between NO₂ and cardiovascular complications during labour.

There is an extensive literature on air pollution exposure in pregnancy and birth outcomes, but it was not reviewed by the ERG (as this literature largely relates to long-term exposure).

Overall, there is suggestive evidence for links between air pollution and pregnancy outcomes, such as hypertensive disorders (based on long-term exposure).

Note: The ERG report does not comment on the discussion of effects during pregnancy in the US EPA ISAs, as the evidence discussed is not directly relevant to the scope of the review: The ISAs’ conclusions on the evidence for pregnancy being considered as a susceptibility factor is mixed (suggestive for PM_2.5, inadequate for NO₂, suggestive for O₃ and inadequate for SO₂), although this categorisation is often based on multiple outcomes (for example, fertility, pregnancy, and birth outcomes) rather than on the pregnant mother’s health itself).

Overall conclusions on susceptibility groups for the DAQI

The report would not recommend changes to the at-risk groups already mentioned in the DAQI, although it may be useful to communicate the state of the evidence regarding other medical conditions to those that could be potentially susceptible to air pollution, but where sufficient evidence is currently lacking.

Further considerations

Baseline differences and control groups

One of the challenges in defining or demonstrating that various groups are susceptible is the need to compare the effects of the air pollutant in corresponding control groups – usually healthy individuals or the general population. This may be easier in non-human studies, where experiments can be designed to have a matched control group. In some cases, results from this type of study may generate data that clearly show animal models of disease are more susceptible to the effects of the air pollutant exposure. For example, in a study looking at the effect of diesel exhaust particles on renal function, the particles had no effect on levels of oxidative stress in healthy animals, whereas they caused oxidative stress in animals that had been treated with an agent that simulates kidney injury (Nemmar and co-authors, 2014). Thus, it would be reasonable to postulate that the healthy animal had robust anti-oxidant defences to protect against the particle exposures, whereas the injury model is susceptible to their action due to impaired defence mechanisms. However, it is important to recognise that adequate non-human models might not be available for some at-risk groups. For example, few animals develop asthma, and animal models of asthma may not exhibit all of the relevant features of asthma.

However, often there are inherent differences in the baseline responsiveness between the groups that could confound this. For example, in disease there may be a difference in their baseline health status that affects the capacity of an air pollutant to induce an effect. If a health condition causes, say, a 30% loss in a health parameter, then there is now only a 70% capacity left for the air pollution to act on. Therefore, the % inhibition of the parameter caused by the pollution may be less in the patient group than the healthy group because of this lower capacity. An example of this is given by the controlled exposure study by Mills and co-authors in patients with ischaemic heart disease, where diesel exhaust exposure had little effect on vasodilator responses (Mills and co-authors, 2007), compared to previous studies showing large effects of diesel exhaust on vasodilator function on healthy volunteers (Mills and co-authors, 2005). Other studies have observed greater effects on certain parameters in healthy volunteers than those in patients with chronic obstructive pulmonary disease (for example, Syed and co-authors, 2021; Gong and co-authors, 2004). It is also worth bearing in mind that patient groups will be likely to take medication for their condition, which may affect the magnitude of response to air pollutants and alter the susceptibility to the effects of air pollution. Different scenarios may be present depending on the size of the response incurred (whether effects are sub-maximal) or if the effects of the disease and the pollutant have additive/synergistic interactions. Therefore, it is often difficult to define if a health condition is affecting susceptibility to air pollution because often it is not known what the differences in the baseline levels of function are between the healthy versus the disease population (in the absence of air pollution exposure).

In some cases, a suitable comparison/control group may not be available. For example, it is often cited that children are more susceptible to the effects of long-term air pollution because their lungs and other organs are developing. A direct comparison with a control group – adults or the general population – is not possible because organ development has already ceased. Nonetheless, there would be justification in categorising children as a susceptible group for this reason even in the absence of a comparative group. A similar argument may be true for events such as heart attacks – air pollution could be a trigger for a heart attack in someone with coronary artery disease, whereas it is very unlikely that someone who does not have underlying vascular disease will be at risk of a heart attack (without another biological reason or stress). While there is not an ideal scientific control, there would still be justification including someone with severe coronary artery disease as a potentially susceptible individual due to the risk of a fatal event.

Further considerations may need to be taken into account in individuals where there are biological factors that may change their effective dose of exposure. For example, clearance mechanisms may be less efficient in older people, or lung volume or breathing characteristics may be different in those with obesity. These biological factors may engender these individuals with different degrees of susceptibility independent of the downstream biological mechanisms that may mediate the response to the pollutant.

The majority of studies are largely confined to whether a pre-existing health condition confers susceptibility for parameters relating to that type of condition, such as the influence of asthma on an air pollutant’s ability to alter lung function. However, given the interplay between organs, it may be the case that having, say, a respiratory condition, may make a person more prone to non-respiratory effects of air pollution. For example, does COPD influence the impact of air pollution on cardiovascular parameters. The evidence base for most inter-organ susceptibility is likely to be limited. There are some instances where there may be more evidence for this, for example whether COPD (Table S12-5) or diabetes (Table S12-2) might make a person more prone to the cardiovascular effects of air pollution (US EPA, 2019b). Or alternatively, there may be indirect evidence from studies investigating effects that are not specific to one organ system, such as blood biomarkers of inflammation.

Respiratory disease and cardiovascular disease

The ERG report works from the basis that, given their inclusion in the existing DAQI, both respiratory disease (in general) and cardiovascular disease are susceptibility factors to air pollution. This statement would concur with the large literature base showing that multiple air pollutants exacerbate cardiorespiratory diseases at different stages in the disease progression. However, evidencing that a person with cardiorespiratory disease is more susceptible than a healthy person is not always straightforward (see above). This is exemplified by the summary tables of the US EPA ISAs shown previously, where the evidence for cardiorespiratory susceptibility is categorised as suggestive or inadequate for some air pollutants (although we do acknowledge that the evidence has grown since the publication of these ISAs).

However, because of the presence of an underlying precipitating disease, these groups are more likely to have severe consequences of the biological effects of air pollution exposure, such as a heart attack or acute exacerbation of asthma. Given the well-recognised actions of air pollution on cardiorespiratory function and the potentially fatal consequences of exacerbation, then there is an argument for including these individuals as a susceptible group from a precautionary principle alone. It is less likely that someone without underlying disease would experience a severe clinical outcome, such as a heart attack, following short-term exposure to air pollution.

Future research on this matter, that could potentially benefit the AQIS review, may wish to address the following questions:

• which specific respiratory and cardiovascular conditions confer susceptibility to different pollutants?
• how severe does the disease need to be before a person is significantly more at risk from exposure to air pollution?
• does treatment of the underlying disease modify the risk from exposure to air pollution?
• should risk factors for these conditions (for example, high blood pressure) also be considered to be a susceptibility category in their own right?

Children

Based on the evidence we have considered, children with asthma should be considered as a group susceptible to the effects of short-term exposure to air pollutants. They should therefore continue to be regarded as an ‘at-risk’ group for the purpose of the DAQI advice.

Both healthy children and those with pre-existing health conditions are regarded as being at risk of specific effects from long-term exposure to air pollutants. These include decreased lung function development and the development of asthma. Although not discussed here, the Sub-group are considering what a long-term air quality information system may look like, to provide information to the public on long-term exposure to air pollution (see working paper 3). Drawing attention to specific risks of long-term exposure might be relevant to provide alongside this long-term air pollution information.

However, there are a number of considerations, firstly, it is not clear if air pollution places non-asthmatic children at greater risk than adults of non-asthmatic respiratory effects such as infection. However, for parameters where there are direct comparisons with adults (for example, hospital admissions) the evidence is inconsistent as to whether children are at greater risk (more likely to present at a hospital) than adults (US EPA, 2016). A new systematic review of the literature may help clarify these two points. Nonetheless, children may be at increased risk due to factors such as increased activity, time outdoors, organs developing, breathing characteristics (US EPA, 2016; 2019a).

Other susceptibility factors that are not considered

There are a number of disease conditions or groups which could be more susceptible to air pollution which may merit more attention.

These include respiratory diseases such as COPD. The US EPA ISAs cover COPD, however, the evidence base is, in general, smaller than for asthma as a susceptibility factor. The ERG report does not assess respiratory diseases other than asthma sub-groups, therefore, the literature after the time of the US EPA ISAs has not been reviewed as part of this AQIS review.

Hypertension is also included. The US EPA ISAs cover hypertension as a factor that may confer susceptibility to air pollutants in terms of cardiovascular disease. A systematic review would be needed to fully assess the recent literature for specific air pollutants and whether this risk factor may affect susceptibility to non-cardiovascular endpoints (a growing literature suggests that hypertension is linked to conditions such as dementia, cancer and other non-cardiovascular conditions).

Conditions or Groups which have not been considered are:

• kidney disease
• neurodegenerative conditions, such as dementia
• mental health
• genetic factors

Individual air pollutants versus combined metrics

Both the US EPA ISAs and the ERG report highlight that, for some groups, the evidence of increased susceptibility is sufficient for one pollutant but not necessarily others. This could have implications for the current DAQI which uses an index for the air pollution mixture rather than specifying individual pollutants that are elevated. For example, there is some evidence that obese individuals may be more susceptible to the respiratory effects of ozone. It would be expected that it would be common that the air quality banding from the DAQI may be based on PM and NO₂, rather than ozone. When the DAQI air quality index is elevated, at risk individuals may respond by taking action to reduce exposure to air pollution, for example, by avoiding commonly perceived sources of air pollution, such as traffic. However, does this then mean that a person may spend more time indoors where some air pollutants may be higher, or a different mixture than outdoors, or further away from urban areas where exposure to pollutants such as ozone or biological allergens may be higher.

It might also be helpful to provide information on individual pollutants, to allow the user to see which pollutant is driving the DAQI alert. Links could then take the reader to further information saying who in particular may be more at risk, and where these pollutants are most likely to be highest.

Concentrations cut-points

At what concentrations of air pollutants should the banding of the DAQI categories be chosen, or specific tailored advice given? For the susceptible groups identified, there is the suggestion that those individuals will suffer effects of air pollution at concentrations lower than those of non-susceptible individuals. The ERG report suggests that controlled human exposure studies could be used to identify thresholds. However, due to the acute nature of exposure, almost all controlled exposure studies use concentrations well above background levels, with even sensitive parameters often not showing effects below 200 µg/m³ of diesel exhaust. Thus, information of relevant concentrations would need to come from large scale epidemiological studies and, in most cases, it is unlikely that this evidence base would be available. The significance (for health) of transient high peaks in air pollutants is also unknown, and would be difficult to assess in epidemiological studies.

Medication

For conditions like asthma, it may be possible for individuals to take ‘quick-acting’ protective or relief medicines when air pollution is high. However, it should be noted that the Sub-group’s working paper on lag structures of symptoms after exposure to air pollution, noted that there is limited evidence on the best clinical strategies to minimise the effects of air pollution in people with asthma (see working paper 2). For other conditions, like COPD and most cardiovascular diseases, individuals will take their medication daily at a dose that is guided by long-term maintenance of their condition. It is unlikely that this medication can be tailored for short-term changes in air pollution. Another consideration is for individuals who belong to a susceptibility category for pre-existing disease (for example, coronary heart disease) but have a stable condition that is well-managed by their ongoing medical treatment. Should these individuals also constrain their activity on high air pollution days? Thus, there may be need to provide more nuanced messaging for individuals who have conditions which are under medical treatment. It would be unadvisable to have very detailed information on specific medical conditions upfront on the DAQI webpages, however, a link could be provided to additional pages of information, or with a more generic message saying that patients with health conditions should discuss these issues with their clinician.

Concluding remarks and recommendations

We have drawn on comprehensive reviews of groups that are potentially susceptible to air pollution carried out by the US EPA. This has been supplemented by an up-to-date review of the literature by the ERG, commissioned to support the AQIS project, who have performed a detailed review of whether biological responses to air pollution are greater (or different) in individuals with obesity, metabolic syndrome, children and older people, as well as considering different subtypes of asthma and the effects of pollutant exposure on the mother when pregnant.

There are differences in the susceptibility of different groups between the biological outcome under investigation and between different pollutants. While there is some inconsistency, and in many cases limited evidence, overall, the evidence suggests that individuals with respiratory or cardiovascular disease, and older individuals, are likely to be more susceptible to air pollution than healthy adults. There is also adequate evidence to suggest that children with asthma will be particularly at risk of the short-term effects of air pollution. These conclusions are reflected in the advice given in the current DAQI.

There is also a suggestion that individuals with obesity or metabolic syndrome/diabetes may also be at higher risk of the effects of air pollution than non-obese healthy individuals. However, the evidence reviewed is not sufficiently strong at present to justify specifically regarding these groups as susceptible to short-term exposures, and providing behavioural recommendations for these groups in the up-front advice accompanying the DAQI. There may, however, be value in providing further information on this subject for individuals who are interested or who wish to gain further insight into the evidence that is available.

At present, there is insufficient evidence to determine if there are particular subtypes of asthma that confer a heightened susceptibility to air pollutants compared to other forms of asthma. A large evidence base is growing on the effects of exposure to air pollutants during pregnancy, on birth outcomes and the child’s health later in life, but there are very few studies looking at the health of the mother in pregnancy, especially in terms of short-term exposure.

It is important that consideration is given to the sort of information that might be most helpful to empower clinicians and health practitioners to advise their patients regarding the potential impacts of air pollutants on their health. This might perhaps be in the form of case studies, example conversations or narrative text, for example. The format of advice and information should be tested with the proposed audiences, to ensure that the most appropriate format is used.

Recommendations for further research

It is recommended that further research is undertaken to examine the extent to which other groups might also have increased susceptibility to the effects of air pollutants. This could include an appraisal of the evidence for respiratory conditions (other than asthma, such as childhood wheeze) and individual cardiovascular conditions, further exploration of obesity and diabetes as susceptibility factors, as well as emerging associations for conditions such as kidney disease, neurological conditions and mental health, and factors such as ethnicity. It is recognised that there is gathering evidence on the effects of air pollution on birth outcomes. We recommend that future research should also focus on the effects of short-term exposure to air pollution on the health of pregnant mothers.

Other considerations for future research include further investigation into the effects of air pollution on disease sub-types (for example asthma), specific responses to different pollutants, whether different exposure-response functions should activate different levels of advice, whether susceptibility may differ in those with different sub-types (for example, asthma) or manifest across different organ systems (for example, does asthma influence the impact of air pollution on cardiovascular parameters?), and to what degree tailoring lifestyle, diet and medication could play a part in moderating an individual’s response to air pollution. Evidence related to pollutant-specific responses, and whether exposure-response functions differ between groups, might also help advice to be tailored in the future.

References

Committee on the Medical Effects of Air Pollutants (COMEAP). 2021 ‘Advice on health evidence relevant to setting PM_2.5 targets’ (viewed 12 February 2025)

Evangelopoulos D and co-authors. 2023 ‘Identifying and defining “At Risk” groups to better target air quality information: Evidence assessment for diabetes, obesity, subtypes of asthma and life-stage’ A report from the Environmental Research Group, Imperial College London, commissioned by Defra to inform the AQIS project.

Gong Jr H and co-authors. ‘Exposures of elderly volunteers with and without chronic obstructive pulmonary disease (COPD) to concentrated ambient fine particulate pollution’ Inhalation Toxicology, 2004: volume 16, pages 731 to 44 (viewed 13 February 2025)

Mazmuder H and co-authors. ‘Maternal health outcomes associated with ambient air pollution: An umbrella review of systematic reviews and meta-analyses’ Science of the Total Environment, 2024: volume 914, article number 169792 (viewed 13 February 2025)

Mills NL and co-authors. ‘Ischemic and Thrombotic Effects of Dilute Diesel-Exhaust Inhalation in Men with Coronary Heart Disease’ The New England Journal of Medicine, 2007: volume 357, pages 175 to 1082 (viewed 13 February 2025)

Mills NL and co-authors. ‘Diesel exhaust inhalation causes vascular dysfunction and impaired endogenous fibrinolysis’ Circulation, 2005: volume 112, pages 3530 to 3536 (viewed 13 February 2025)

Royal College of Physicians and Royal College of Paediatrics and Child Health. 2016 ‘Every breath we take – the lifelong impact of air pollution’  (viewed 24 February 2025)

Syed N and co-authors. ‘Effect of diesel exhaust on exercise endurance and cardiorespiratory responses to exercise in chronic obstructive pulmonary disease and healthy controls-a randomized, placebo controlled, crossover study’ American Journal of Respiratory and Critical Care Medicine, 2021: volume 203, article number A1054 (viewed 13 February 2025)

United States Environmental Protection Agency. 2015 ‘Preamble to the Integrated Science Assessments’ (viewed 24 February 2025)

United States Environmental Protection Agency. 2016 ‘Integrated Science Assessment (ISA) for Oxides of Nitrogen’ (viewed 13 February 2025)

United States Environmental Protection Agency. 2017 ‘Integrated Science Assessment (ISA) for Sulfur Oxides’ (viewed 13 February 2025)

United States Environmental Protection Agency. 2019a ‘Integrated Science Assessment (ISA) for Particulate Matter’ (viewed 13 February 2025)

United States Environmental Protection Agency. 2019b ‘Supplement material: Chapter 12 (Populations At-Risk) Integrated Science Assessment (ISA) for Particulate Matter’ (viewed 13 February 2025)

United States Environmental Protection Agency. 2020 ‘Integrated Science Assessment (ISA) for Ozone and Related Photochemical Oxidants’ (viewed 13 February 2025)

United States Environmental Protection Agency. 2022 ‘Supplement to the 2019 Integrated Science Assessment (ISA) for Particulate Matter’ (viewed 13 February 2025)