Health and Economic Effect of Air and Noise Pollution near Airports

Hospital Episode Statistics (HES) data was supplied to King’s College London for processing for the purpose of a study in the Health and Economic Effect of Air and Noise Pollution near Airports.

The importance of London Heathrow Airport to the London and UK economies is well documented (British Chambers of Commerce 2009; Heathrow Airport Limited 2011). This pivotal role of London Heathrow Airport for London and the UK economies is the result of its status as a European hub (or transfer centre) for international passengers travelling around the world. Heathrow is currently at capacity and expanding it would generate between £5.5 and £20 billion of economic benefit over the period 2020-2080 (Department for Transport 2009; British Chambers of Commerce 2009); or even £14 billion a year according to Frontier Economics (2012).

Heathrow Airport published a report to highlight the necessity of a clear action plan for the future (Heathrow Airport Limited 2013), and there is a growing demand from the London population calling for a robust assessment of the actual and future impact (The London Borough of Hillingdon 2011; Policy Exchange 2012). Understanding the full impact of airport expansion is critical to any cost-benefit analysis that will be used in decision-making. A significant gap exists in the knowledge of the health and economic effects of airports on the local population, mainly from air and noise pollution. This project aims to fill the gap that concerns air and noise pollution.

The main purpose for continued retention and processing of NHS Digital data is to be able to conduct robustness checks on the papers resulting from the original PhD project. King’s College London hope and expect that the extension will be sufficient for the write up, submission and answers to the queries raised from referees.

GDPR Article 6 justification of processing the data is that HES data is the only available data allowing to investigate precisely and causally the impact of noise and pollution generated by airports (London Heathrow is particularly interesting due to its close location to densely urbanised areas). This pivotal role of London Heathrow Airport for London and the UK economies is the result of its status as a European hub (or transfer centre) for international passengers travelling around the world.

Public interest (GDPR article 9(2)(j)) is provided by the importance of the airport for the UK as a major (European) Hub. The public is directly concerned by both positive and negative spillovers from airport locations. This study is attempting to get a better understanding and quantitative evaluation of the negative externalities generated by airports and more generally the aviation industry (which currently account for 3% of greenhouse gas emissions, but could account for 21% by 2050 if current maintains, EU (2018).

There is no moral or ethical issues raised by the proposed dissemination. The paper should not lead to adverse effect on the population exposed, it is expected they may benefit by publications showing that flight paths should avoid as much as possible areas with high level of population density.

King’s College London identified the following areas in order to carry out this work:-

B = North West

E = East Midlands

F = West Midlands

G = East of England

H = London

J = South East

K = South West

The above areas, each of which contains a major airport, were investigated simultaneously. King’s College London compared the backgrounds situation at Heathrow airport with other major airports to account for potential confounders in the relation between airports activity and health conditions.

King’s College London investigated how both air pollutants and noise correlate with health outcomes. The pseudonymised extract of HES data at individual level together with environment and airport related data were the necessary elements to undertake such statistical analyses. In particular King’s College London used Stata (software developed by epidemiologists and widely used by applied economists).

The HES data was used only at aggregated level and compared with environment and airport-related data (also at aggregated level). The smallest unit of aggregated analysis would be at postcode district level. This type of data consists of levels of pollution recorded by monitoring stations, noise level contours, taking off and landing aircraft, temperature, humidity level and other weather variables.

The data formed the basis of a dissertation which has since been published - https://kclpure.kcl.ac.uk/portal/en/theses/health-effects-of-noise-and-air-pollution(edfd9ba5-4378-49b3-8cd1-3d8b190c5faa).html. Due to the findings as a result of the work, continued access to the data is required to enable King’s College London to develop academic papers to be written.

This Data Sharing Agreement permits the resulting academic papers to be developed, and if required, re-analyse the data in line with the objectives of the study. Once written, the data will be stored and retained for a short period (until June 2020) in order to answer any queries that arise as a result of the dissertation or the papers.

Following publication of the study findings, it is possible that the findings will be questioned or challenged by third parties through direct contact with King’s College London, contact via the publishing journal or an open letter. In such circumstances, King’s College London may re-run the previously analyses undertaken to verify that the published results were accurate and may write a response to be issued directly to the challenger or published.

If any further data processing is required in addition to the above purposes or if the data needs to be moved to a different location/organisation the applicant must submit an amendment request to NHS Digital before data is accessed.

In order to investigate how morbidity and mortality trends respond to changes in pollution emissions, Kings College London look for shocks to usual emission rates of both air and noise pollution. Regarding Heathrow airport, Kings College London identified one major non-gradual exogenous variation to aircraft noise and air pollution emission rates. For five months, from 5th November 2012 to 31st March 2013, Heathrow airport ran the Early Morning Arrivals Trial in collaboration with the noise pressure group HACAN (Heathrow Association for the Control of Aircraft Noise), British Airways and NATS (formerly National Air Traffic Services). The aim of this trial was to investigate the feasibility of providing a predictable relief period for some communities close to Heathrow's approach paths. The conditions of the five-month trial are set out in a detailed report (Tucker, K., Williams, R., and Leighton, S. 2013. Heathrow Airport - Early Morning Arrival Trial Final report. Helios, London.). These include the identification of two pairs of exclusion zones, which would alternatively be avoided by aircraft between 23:30 and 06:00 for the duration of the trial. Other air traffic variations at Heathrow airport and other airports were expected to be identified.

The goal of this project was to assess the impact of changes in aircraft emissions on health outcomes for those people living underneath flight paths. The clearest way to simultaneously isolate causal effects of the flight changes and control for confounding factors is to explore over time outcome differences between communities that experience the flight change and communities that do not.

The empirical design adopted here is a standard two-period difference-in-differences (DD) approach. It consists of examining differential trends of postcode districts (or LSOAs) affected by the change and non-affected postcode districts over time (before and during the trial).

To actually calculate the health effects King’s College London estimated a regression equation which has hospital admission outcomes as dependent variable and geographical (e.g., postcode district, LSOA), environmental (e.g., noise and air pollution levels at aggregate level, precipitations, humidity, temperature, wind speed and wind direction) and socio-economic dummies as independent variables.

To carry out this work, Hospital Episode Statistics Admitted Patient Care 2008/09 to 2013/14, HES Outpatient 2011/12 to 2013/14 and HES A&E 2011/12 to 2013/14 was requested and disseminated to KCL under a previous iteration of this Data Sharing Agreement.

By using a methodology relying in DD, KCL need data over a sufficient long period before the implementation in order to verify that pre-trial conditions in the chosen treated and control groups were following a similar pattern (a condition named as ‘parallel trend assumption’ in the econometrics jargon).

This DD model furnishes several advantages. Firstly, it produces easily interpretable estimates (i.e. what happens if an average area is exposed to changes in flight path?). Additionally, the difference-in-differences identification is applicable in more complex settings with multiple time periods and manifold treatment intensities.

The control group includes those areas with no change in air traffic before and during the trial. The treatment group can ideally be represented by all the other areas clustered depending on the nature of variation. This range of variations creates the perfect environment for the aforementioned multiple treatment classifications. Both treatment and control regions naturally incorporates both rural and urban regions, both poor and rich socio-economic characteristics. Even though the areas need a robust check for their comparison to be validated, the trial seems to have all the properties to build a natural experiment setting.

The DD regression and all the analysis involved in this project was run (and continues to be) using Stata software (StataCorp. 2013. Stata Statistical Software: Release 13. College Station, TX: StataCorp LP). The DD design was implemented for all the different clusters of health outcomes (respiratory disease, cardiovascular disease, nervous system disease, birth outcomes) and also for the aggregation of all medical conditions. This was to investigate the effect of environmental stressors on specific illness as well as the overall health status depletion.

Data will not be stored, processed or in any other way accessible by a third party organisation and will be stored in only a single location at King's College.

Kings College London confirms that the data received will only be held and analysed in England and no record level data will be shared to any third party. The data will not be used for commercial purposes. The initial study was completed in February 2018 with the dissertation published on King’s College London’s website - https://kclpure.kcl.ac.uk/portal/en/theses/health-effects-of-noise-and-air-pollution(edfd9ba5-4378-49b3-8cd1-3d8b190c5faa).html. A London based newspaper have taken note of this work and published an article (in paper form only). The article highlights that Heathrow airport representatives are also aware.

Through publications in peer reviewed journals, this study will therefore contribute to the broader literature on the health and economic effects of environmental pollution as well as the specific debate on the future of airport development in London and the South East England.

This agreement allows academic papers to be written and published. It is expected that the results in the study will help decide on the correct timings of aircraft landing around Heathrow. But the expectation is that the media coverage (KCL is planning a press release at the time of publication) will allow the results to be well published and can help plan the correct approach for every UK airports situated near urbanised areas.

All outputs consist of aggregated results with small number suppression applied in line with the HES analysis guide.

London Heathrow airport has reached full capacity since 2011. One implication of this situation is that the UK government has to make a decision over its expansion.

Kings College London’s research is motivated by the vivid policy debate on the costs and benefits of the aviation industry. In particular King’s College London is interested in estimating the component of the total health costs imposed on society by air traffic-related pollution.

In terms of benefits to the UK Health and Social Care system, the first contribution this project will make will be to provide an up-to-date insight into the negative effects of ambient noise and air pollution on well-being. In fact, individual well-being and health status are major factors influencing individual productivity (either contemporaneous or future), which in turn affects the aggregate productivity of labour. Airports generate both air and noise pollution, which can each be measured separately and relatively accurately.

Most importantly, with the support of a natural experiment setting, Kings College London’s research explores the causal effect of airport-related emissions on health. Health outcomes include hospital admissions for cardiovascular, respiratory and nervous system-related diseases, low birth weight and length of gestation. The approach used is a difference-in-differences (DID) estimator that exploits changes in flight regimes over time and across areas affected by flights to and from London Heathrow airport.

A subsequent step would be to look at all major airports in England to have a better understanding of health care needs in areas located near to airports.

King's College London will present their results to policy audiences in order to disseminate the results from our research.

The aim this research ‘Health and Economic Effect of Air and Noise Pollution near Airports’ is to rigorously identify a causal relationship between air and noise pollution emissions from air traffic and health outcomes of the population living underneath flight paths.

This project essentially aims to reply to three broad research questions. Firstly, what are the size and the nature of the health effects of airport activities on the local population, mainly from air and noise pollution? In particular, the aim is to univocally define the causal relationship between pollution emissions from aircraft movements and human health. This is done around London Heathrow airport, which is characterised by a highly dense urban area and good availability of data. A further research will be done using a multi-airport approach in order to increase the generalizability of these results. A major contribution of this research will be a clearer understanding of the simultaneous health effects of both air and noise pollution.

Secondly, does airport-related pollution causes new disorders or does it simply induces a worsening of health status to people with previous medical conditions? Having access to HES data is the essential key to unravel the opaque distinction between exacerbation and initiation of ailments.

Finally, what are the costs borne by society? The concluding task is to get some figures of the actual costs of illness induced by airport-related pollution. This aims to take part and fundamentally contribute to the current policy-making discussion over costs and benefits of the aviation market in the UK.

To summarise, the benefits would be both to society in terms of a clearer understanding of the health effects of environmental hazards and to the field of microeconometrics. In fact, the use of a quasi-experimental approach is an econometric technique which is receiving growing interest within the social sciences. In conclusion, information about the impacts of air and noise pollution around airports will help the NHS to better plan its resource allocations, especially with regard to expansion of Heathrow and increased air traffic more generally.

The research work being conducted is key to inform the NHS that Kings College London are doing work that will directly benefit the NHS by providing new and reliable results on the impact of air and noise pollution for population living near major airports.

Kings College London will plan to discuss details and implications of the research to public health departments in the local authorities relating to the data areas covered and by circulating a summary version of their research and offer to meet with them to present the main results and discuss the implications as a result of the research.

The analysis has been subject to an article in a London based newspaper which highlighted that the 5-month flight ban that occurred in 2012 at Heathrow airport recorded a 5.8% decrease in spending on pills including anxiolytics for conditions such as insomnia, anxiety and depression in the no-fly zones. This in turn shows that as the quality of health of people in these areas must have been better, the lower prescribing saved the NHS money.

The data was able to provide evidence that by focusing on the medical cost of air traffic pollution, including “Exclusion zones” would prevent people living in close proximity from being woken with a start and would suffer fewer pollution-related breathing problems.

In Heathrow’s 2012 trial ban of flights between 4:30am and 6am, the study found that fewer drugs were prescribed for respiratory and nervous system conditions in areas with reduced air traffic.

The analysis cross-referenced NHS data with the trial’s findings and found a link between air traffic and health, notably a 5.8% decrease in spending on pills for conditions such as insomnia and depression in the no-fly zones. By calculation, modifying flight schedules could save £5 million in NHS prescription costs.

The study also found “statistically significant increases” in visits for nervous and respiratory conditions at hospitals near seven UK airports.

This study has been made aware to Heathrow airport who stated that the airport has strict limits on flights but will be consulting on the trial’s results to see how flight noise may be reduced overnight as part of expansion plans. It is expected that the academic papers will provide further evidence to enable Heathrow Airport to make that decision.