Written evidence submission from Professor Anna Hansell (NIT0043)
There are many well-documented health effects of ambient air pollution, with a comprehensive recent summary in the Chief Medical Officer’s Annual Report in 2022 on Air Pollution (Exley et al, 2022; Dimitroulopoulou et al, 2022) and detailed considerations in reports from the Committee on the Medical Effects of Air Pollutants (COMEAP), which I draw on in this written response, providing details below.
Nitrogen oxides
Nitric oxide and NO2 are emitted by cars and combustion process, but nitric oxide is rapidly oxidised to NO2. Health studies generally consider NO2 (as nitric oxide readily reacts with oxygen or ozone in the air to form NO2). There has been debate over several decades over whether associations are causal or whether NO2 is a marker for other pollutants. The majority view now is that both views are correct – there are both causal impacts and NO2 is a partial marker of air pollution mix.
Health effects of short-term exposure to NO2
Health effects of long-term exposure to NO2
Ammonia
This is not considered as a major pollutant by itself in terms of health effects, but reacts with acids (NOx, SO2) in the environment to form secondary particle matter (PM) – ammonium sulphate and ammonium nitrate aerosols. There are multiple sources for PM and these vary by time and place. However, ammonium compounds form potentially 50% or more of the fine particulate load.
Particulate matter
The risks to health from inhaled particulate matter come from the small size fractions that can enter the body – usually considered as PM10, PM2.5 and ultrafines, relating to small particles with an aerodynamic diameter <10µm (about 1/7 the diameter of a single human hair) for PM10, <2.5 µm for PM2.5 and < 0.1 µm for ultrafines. The health evidence is strongest for the PM2.5 fraction, which is detailed below. There is also some literature, chiefly occupational, on PM4 (aerodynamic diameter <4µm). There is much less evidence available on different the different types of particles comprising the PM mix, considered either by source and or by chemical composition (such as black carbon).
Health effects of short-term exposure to PM2.5
Health effects of long-term exposure PM2.5:
Health effects of ultrafine particulates
The evidence is more mixed. Short-term exposure has been linked with inflammation and cardiovascular changes and ultrafine particles have been detected in the in brain and placenta. There is less consistent evidence of associations between ultrafine particulate exposures and mortality. The 2021 WHO Air Quality Guidelines concluded there was insufficient data to recommend a guideline for ultrafines, but issued a good practice statement on monitoring, mitigation and future research.
Are some groups more vulnerable?
Evidence is more limited on vulnerability factors. There are two main considerations:
Estimated mortality from NO2 and PM2.5 in the UK
UK Health Security Agency (UKHSA) mortality calculations for NO2 and PM2.5 calculate an effect from NO2 of 18,735 deaths and PM2.5 of 39,745 deaths in 2019 (Mitsakou et al, 2022). Given close associations between NO2 and PM2.5 concentrations (one could be a partial marker for the other, as often emitted from similar sources such as transport), the adjusted mortality from both NO2 and PM2.5 combined is calculated as 29,000-43,000 deaths. ‘Attributable deaths’ relates to the number of deaths that could be avoided by modifying the environment. Air pollution is a risk factor (for example, as cholesterol is) – while contributing to mortality and ill-health, it is extremely rarely the only or main cause of death, hence the calculated number of deaths attributable to air pollutants, should be considered an effect equivalent to the number of attributable deaths at a population level.
Differential toxicity of particles from different sources
There are multiple sources for PM and composition varies by time and place. Nitrates and ammonium sulphate potentially make up 50% and more of the fine particulate load. A paper in 2023 estimated that 38% of annual average of the PM2.5 particles in Leicester arose from agriculture and 25% in London (Kelly et al, 2023). The differential toxicity of particles of different chemical composition and sources is an area of uncertainty, with limitations in available evidence.
The published COMEAP statement on differential toxicity of particles from 2022, which took on board a range of evidence including comprehensive reviews from national bodies in France (ANSES 2019) and the US (USEPA 2019) concludes: “while the substantial body of recent evidence adds weight to the conclusion that different constituents are likely to have different toxicological actions, it does not consistently indicate specific components of PM that are more toxic than others”.
The COMEAP 2022 statement further comments “The literature reviewed highlighted that a number of different sources and constituents of PM are associated with adverse health outcomes across various exposure periods (‘short-term’ or ‘long-term’). PM (and PM constituents) associated with combustion and road traffic received the most attention, however, there was insufficient coherence in the epidemiological evidence to conclude that these sources of PM played a significantly greater role than other sources or constituents.”
There is a recognised lack of available evidence on the health effects of outdoor ambient air pollution from anthropogenic sources such as agriculture, maritime transport and airport traffic (ANSES 2019).
Toxicological (experimental) evidence generally shows low toxicity for nitrates, sulphates and ammonia compounds; however, associations with adverse effects are often seen in epidemiological studies. There are several reasons why the laboratory studies may differ from findings in epidemiological studies. The ANSES 2019 report notes that nitrates and sulphates may modulate the toxicity of the PM mix rather than having an intrinsic toxicity themselves. It has been suggested that the acidity of sulphate particles increases solubility trace metals (which may be components of some airborne particles) and hence bioavailability, increasing their toxicity. Other factors may include the combination of low toxicity but higher dose and wider population exposures leading to effects detectable at population levels.
The International Agency for Research on Cancer (IARC), has classified particulate matter in outdoor air pollution as carcinogenic to humans (a Group 1 carcinogen), citing lung cancer and bladder cancer and primarily based on evidence from diesel engine exhaust particles (IARC Monograph, 2016). It also found sufficient evidence for carcinogenicity of outdoor air. The Monograph commented on a role of air pollution in both initiation of cancer through genetic damage and promotion of cancer progression via oxidative stress, responses to oxidative stress, and sustained inflammation. Recent influential work from the Crick Institute (Hill et al, 2023) has provided further support for a role of fine air pollution in promotion of cancer progression in lung cancers not associated with smoking. However, the work did not explore differential toxicity of different types of particulates.
More research into the health effects of nitrates and ammonium compounds, which form such a substantive component of the particulate load, should be encouraged. Anecdotal discussions suggest it is currently easier to get funding and high impact journal publications for research into particulates such as those from brake and tyre wear or from more recently considered exposures such as microplastics. Complementary levels of research on lower toxicity particulates is needed to answer the broader question of whether different particles have different impacts on health at a population level.
Ozone (O3) is a secondary pollutant – created by photochemical reactions between NOx and volatile organic compounds (VOCs) – it is mainly a summer pollutant in UK. Its levels are lower in cities as NO2 (from traffic and other sources) scavenges O3. It is a highly reactive irritant gas and an oxidant (leading to oxidative stress).
Health effects of short-term exposures (8 hour and 24 hour) to ozone
There is evidence for a causal association with all-cause mortality, and some evidence for associations with CVD and respiratory mortality.
The strongest evidence is for respiratory effects on symptoms, including asthma symptoms, and respiratory admissions.
There are likely causal metabolic effects relating to impaired glucose tolerance and increases in Triglycerides and fatty acids
There is insufficient evidence related to cardiovascular disease and nervous system impacts.
Health effects of long- term exposure (annual) to ozone
There is suggestive evidence supporting an association with all cause and cause-specific mortality (US EPA Integrated Science Assessment 2020). A recent authoritative systematic literature review and meta-analysis by Kasdagli 2024, did not show associations with all-cause mortality, but did with respiratory mortality with a high level of heterogeneity (a lot of variability across studies).
The evidence for impacts on respiratory disease is likely causal with limited but consistent epidemiological studies on asthma development, hospital admissions for asthma, and allergic responses. Evidence is not as strong as for short-term exposures.
Possible associations have been documented with metabolic disease, cardiovascular disease, fertility, birth outcomes
Health risks for ozone therefore show a not dissimilar spectrum to those described for NO2 and particulates (see above), but the impact of ozone can be complicated to interpret in the UK due to lower levels than in some other countries where health studies have been conducted and also because of temporal and spatial correlations with other pollutants. For example, there can be inverse findings of ozone with NO2 in some studies, which is likely related to atmospheric chemistry (NO2 scavenges ozone, NO2 is involved in the formation of ozone).
One relatively recent approach to estimate health impacts of these two linked pollutants has been to calculate the combined oxidative capacity of NO2 and of ozone (oxidative stress being a key mechanism of the adverse effects of air pollution). Oxidative capacity is calculated by adding gaseous concentrations, weighted by the oxidative potential of each gas (ozone is about twice as oxidising). Oxidative capacity gives higher associations with mortality than for one pollutant individually or that for adding mutually adjusted associated, but these are not as high as adding estimates for the sum of each individual association (e.g. Faustini et al, 2019). There is some overlap and possibly interaction with oxidative potential of particulates; so the use of oxidative capacity as a measure to estimate combined health effects is still in development.
(Health impacts addressed in evidence in section 1a)
(Not a health question, so outwith my expertise)
This is a complicated question. To my knowledge, no-one has directly compared health impacts of NOx from gas stoves vs. those from particulate matter from wood burning. While there is evidence of health impacts from both, it is difficult to definitively quantify health impacts. However, reduction of both NOx and particulates in indoor air will be beneficial to health. Some general comments first:
Health impacts of NOx from gas cooking
Gas cooking is associated with impacts on asthma.
An often-quoted systematic review and meta-analysis, Lin et al 2013 provided quantitative estimates for impacts of gas cooking on asthma and wheeze in children, which included findings from 41 studies. A more recent systematic review published in 2023 by Li et al including 66 studies (including those 41 studies reviewed by Lin) cautioned that the literature is very heterogeneous and some was of limited quality (e.g. self-report rather than objective assessment of exposures, inadequate control for confounders) and therefore does not provide sufficient evidence regarding a causal relationship between gas cooking and childhood asthma.
There are some good quality studies in adults suggesting that gas cooking increases risk of wheeze and asthma attacks, particularly in those with atopy. For example, a UK study by Jarvis et al, 1996. Also, a large European study, the European Community Respiratory Health Study by Pan et al, 2024, finding gas cooking increased risk of wheeze and breathlessness at rest. The authors’ conclusions were that “using gas cookers in the home was more strongly associated than electric cookers with certain respiratory symptoms in adults”. Both the Jarvis and Pan studies showed higher effects in females.
A systematic review by Puzzolo 2024 considered 116 studies comparing use of use of electricity, gas and solid fuel with risk of respiratory illnesses. Meta-analysis showed a statistically significant increased risk of asthma of ~10% for gas vs. electricity combining studies of children and adults. The review mainly used cross-sectional studies (therefore determining causality is less clear) and included studies from low and middle income countries, where exposures may differ from the UK, so results may not be directly applicable to the UK.
Putting this evidence together with that for outdoor air, where ambient NO2 is considered to have a causal role for asthma development, it seems probable that associations between gas cooking and asthma are partially or fully determined by NO2 dose – which will be affected by factors such as who does the cooking, ventilation and relationship of cooking areas with other areas of the building.
Gas cooking and respiratory disease
The systematic review by Puzzolo 2024 mentioned above, found significantly increased risk of pneumonia and chronic obstructive pulmonary disease comparing cooking with gas (which could be any of methane/propane/butane) with electricity.
Health burden due to gas cooking
A report funded by the European Climate Foundation published 2024 suggested large numbers of deaths and asthma cases due to gas cooking and gained wide media coverage e.g. https://www.theguardian.com/environment/2024/oct/28/pollutants-from-gas-stoves-kill-40000-europeans-each-year-report-finds
The report estimated
However, there are a number of important uncertainties in methods used for calculation:
Given these uncertainties, I would recommend against using the numbers provided in the European Climate Foundation report as a basis for policy, but they do illustrate that the number of individuals affected by NO2 from gas cooking could be substantial.
Relative importance of indoor and outdoor NO2 pollution with respect to health
Both appear to be important. A recently published study by Evangelopoulos et al, 2024 is one of very few to look at both indoor and outdoor sources using detailed personal exposure monitoring. The study followed 76 patients with chronic obstructive pulmonary disease (COPD) for 134 days – a susceptible group. The authors examined the relationship between indoor and outdoor NO2 and COPD exacerbations and respiratory symptoms. They observed an increase of 33% in the odds of exacerbation for an inter-quartile range increase in total personal NO2 exposure, of which 19% related to indoor-generated exposure and 12% to outdoor-generated exposures respectively.
Health effects of particulate matter from wood burning stoves
Wood burning stoves are a source of airborne particulates and various chemicals including polycylic aromatic hydrocarbons. The Defra 2025 report - Emissions of air pollutants in the UK – Particulate matter (PM10 and PM2.5) comments “Domestic combustion refers to households burning a variety of fuels including wood, solid smokeless fuels, coal and fuels derived from waste such as coffee logs. This was a major source of PM emissions in 2023, contributing 20 per cent of total PM2.5 emissions and 10 per cent of total PM10 emissions”.
The US National Academy of Sciences 2024 conducted a comprehensive report on PM2.5 indoors titled ‘Health Risks of Indoor Exposure to Fine Particulate Matter and Practical Mitigation Solutions’. The report considered there was enough evidence on PM2.5 to, as a priority, reduce population exposure to PM2.5 indoors in susceptible populations (elderly, young children, those in institutional settings, those with pre-existing conditions).
(Addressed in oral evidence)
Yes (addressed in oral evidence).
References
25/03/2025
14
[1] TNO = Dutch Organization for Applied Scientific Research
Nederlandse Organisatie voor Toegepast Natuurwetenschappelijk Onderzoek