Frequently Asked Questions

Expand All


What is the Environment and Health Atlas?

The Environment and Health Atlas for England and Wales is an independent publication produced by the UK Small Area Health Statistics Unit (SAHSU), part of the MRC-PHE Centre for Environment and Health which is funded by the Medical Research Council and Public Health England. The atlas was published as a book by Oxford University Press on 24th April 2014, and contains more detailed information to put the results in context. The atlas draws on the knowledge of mapping and the use of health and environmental data that SAHSU has accumulated over 26 years of conducting environmental health research.

The atlas provides maps of the geographical variations for a range of health conditions and environmental agents at a small-area scale (census wards), which is a much higher spatial resolution than available elsewhere. It also provides contextual information such as known risk factors for the health conditions and a summary of what is known about possible health effects associated with the environmental agents mapped in the atlas.

The maps have been developed as a resource for those working in public health and public health policy and for the general public to better understand the geographic distribution of environmental factors and disease and to identify questions for future research. However, it is important to note that simple comparison of the maps of environmental agents and health conditions cannot be used to suggest causal associations. More information about other factors that might affect disease risk and further statistical analysis would be needed, including at individual level.

What are the aims of the atlas?
  • To provide baseline information for policy makers and the public on geographic patterns of environmental pollutants and disease.
  • To help in development of hypotheses to understand and explain variability in disease risk that may relate to the environment, lifestyle factors and/or location.
  • Following on from this, to help in development of research to investigate potential causal relationships between environment and health factors – where either evidence or lack of evidence for an effect provides important information to inform public health and policy.
An area with high environmental exposures has higher risks of a health condition. Is this likely to be causal?

Similar geographic patterns of environmental agents and health conditions does not mean the association is causal – i.e. where [ward X] has both high concentrations of [Y environmental agent] and higher risk of [Z disease], it does not mean that Y caused Z. The health condition maps show average long-term risks for an area. Each individual living in that area will have different risk factors such as genetic factors, age, whether they smoke or not, diet and exercise patterns, other medical conditions etc. that may have important influences on their individual risk. Similarly, the concentrations of an environmental pollutant in an area may not reflect how much of the agent the person has individually been exposed to, especially if they have recently moved into that area.

A simple comparison of maps can sometimes be useful to develop hypotheses about whether there might be a causal link. However, further study would be needed, including at individual level, to investigate whether associations might be causal or not.

Who is the audience for the atlas?

The atlas has been developed to be accessible to a range of audiences, including researchers and health professionals, policy makers and the public.

What maps are available in the online atlas?

We present maps for women and men for 14 health conditions of public health importance and four selected types of environmental agents in England and Wales, listed below. The print version of the atlas also includes radon. Interactive radon maps are available from Public Health England here.

Health conditions:

  • Lung cancer
  • Breast cancer
  • Prostate cancer mortality
  • Malignant melanoma
  • Bladder cancer
  • Leukaemia
  • Brain cancer
  • Liver cancer
  • Mesothelioma
  • Coronary heart disease mortality
  • Chronic Obstructive Pulmonary Disease (COPD) mortality
  • Kidney disease mortality
  • Stillbirths
  • Low birth weight

Environmental agents:

  • Agricultural pesticides
    • Fungicides - 2000
    • Herbicides - 2000
  • Air pollution
    • Nitrogen dioxide (NO2) - 2001
    • Particulate matter (PM10) - 2001
  • Chlorination disinfection by products (DBPs) in drinking water – trihalomethanes (THMs)
    • DBPs Summer - 2000
    • DBPs Winter - 2000
  • Sunlight duration - 1980-2005

Why were the health conditions chosen?

The health conditions selected are of public health importance (for example, cancers and some common causes of death) where environmental risk factors may potentially contribute to risk of developing that condition. We have included stillbirth and low birth weight, as the developing fetus is potentially highly sensitive to environmental agents.

Other important health conditions such as asthma or diabetes could not be included in this atlas due to difficulties obtaining the most relevant data. For example, a map of asthma would be difficult as there is no national register of asthma. Asthma mortality is very rare so would not provide accurate maps. The best measure of local variability in asthma incidence would be from GP records, but these data are not currently nationally available.

Why were the environmental agents chosen?

The following environmental agents were chosen for the maps:

  • Agricultural pesticides (including fungicides, herbicides and insecticides)- 2000
  • Air pollution – nitrogen dioxide and particulate matter (PM10) - 2001
  • Chlorination disinfection by products in drinking water – trihalomethanes (THMs)- 2000
  • Sunlight duration - 1985- 2000

And additionally in the print version of the atlas:

  • Potential for radon levels in homes to be at or above the Action Level of 200 Bq/m3 - 2007

We call them agents rather than hazards because sunlight can be beneficial as well as harmful to health and agricultural pesticides and THMs in drinking water are potential but not proven hazards.

The selected environmental agents were chosen because of the availability of good quality data at very high spatial resolution on a national level and their potential to be associated with health conditions. We were restricted in our choice of environmental agent as data were not always available, particularly at national level or at fine spatial resolution.

What can the maps be used for?

The health maps can be used to provide a baseline picture of the longer term geographical variability at a small-area scale (census ward level). This is a much finer spatial resolution than in most previous atlases. Health conditions are generally averaged over a 25 year period (1985-2009). Health maps are adjusted for deprivation and age distribution in each area, so these should not explain the geographical patterns. The environmental maps describe geographic patterns of some important environmental agents, again at fine spatial resolution.

The maps by themselves cannot answer why geographical patterns exist, but they can be used to help identify important questions for further research and potential policy interventions.

What can't the atlas show? (clusters of disease, individual risk of disease, individual exposure)

The atlas cannot be used to provide evidence of clusters of disease. The health maps shows averaged area-level risks over a long time frame (usually from 1985-2009, i.e. 25 years) that have been adjusted to allow for small numbers of events (smoothed). Additional information and different analysis techniques would be needed to demonstrate clusters of disease.

The maps do not show the current risk an individual has of developing a disease, which will depend on other factors including genetic susceptibility, lifestyle and whether they have other medical conditions.

The environmental maps show averaged area-level exposures related to specific years, which may not correspond to, an individual’s exposure to an environmental agent. The environmental agents mapped are:

  • Agricultural pesticides (including fungicides, herbicides and insecticides) - 2000
  • Air pollution – nitrogen dioxide and particulate matter (PM10) - 2001
  • Chlorination disinfection by products in drinking water – trihalomethanes (THMs) - 2000
  • Sunlight duration - 1985-2000

And additionally in the print version of the atlas:

  • Potential for radon levels in homes to be at or above the Action Level of 200 Bq/m3

How is this atlas different from other atlases?

The atlas provides health maps at census ward level (average population 6,000), giving a finer spatial resolution than available previously to better show and understand geographical variability. Previous UK atlases have used coarser spatial resolution e.g. at district, NHS board or primary care trust level.

Health maps usually adjust for age and sex, but the atlas maps are additionally adjusted for deprivation and for small numbers, to help identify variations in disease risk that may potentially be due to environmental risk factors.

The atlas is fully interactive down to ward or postcode level and contextual information is given. For health outcomes, a graph of risks in the selected ward compared with all other wards in the same county can be displayed and we also provide the number of people living in the ward. Information about national rates and major risk factors are provided.

Who provided the data? Can I download the data?

SAHSU does not own the data used to create these maps therefore we cannot make the underlying data available to download.

The health data came from official national data sources in England and Wales: mortality and birth statistics came from the Office for National Statistics (ONS) , while cancer registrations came from ONS and from the Welsh Cancer Intelligence and Surveillance Unit (WCISU) Population and contextual data came from ONS. Health and population data at a coarser spatial resolution (i.e. regional level) are readily available to download from the ONS website; the more detailed health data used to produce the maps in the atlas can be obtained from ONS on request provided the correct permissions and data security are in place.

The environmental maps came from a variety of sources but the majority of the environmental agent maps required at least some geospatial modelling by SAHSU. SAHSU has permission to present these data as maps but ownership of the underlying data is still held by the original providers.

I live in England and had a cancer during the time period of these maps. Is there any chance I could be identified?

The health maps use the relative risks for each ward not actual numbers of cases or rates. While these risks are based on actual numbers of cases they have been adjusted for age and deprivation as well as “smoothed” to account for small numbers. It is therefore not mathematically possible to work back from a smoothed relative risk to the actual number of cases in a ward over a twenty five year period. In addition, the underlying data for these maps is not online and is not available to download.

SAHSU and does not hold names. Each health record is encrypted and held on a secure stand-alone private network (i.e. with no links to the internet or university networks) following strict information governance procedures under permission from the National Research Ethics Service, the Health Research Authority, Public Health England and the data providers.

What population does the atlas study? Why are Scotland and Northern Ireland not covered?

The availability of national environmental data at small area level is limited and in the majority of cases was only available for England and Wales (and for some environmental maps only England). SAHSU holds health data from ONS going back 25 years for England and Wales but it does not currently hold Scottish or Northern Irish data as these are held by separate bodies.

Health maps:

What type of data do the health maps show?

The health conditions are mapped at small-area (census ward) level and generally show relative risks for a 25 year period (1985-2009), presented as separate maps for males and females. The relative risks represent the risk of an area (ward) relative to average risk in England and Wales. These are adjusted for age, deprivation and chance fluctuations due to small numbers. Maps include two birth outcomes (incidence of stillbirth and low birth weight), nine cancers and mortality from heart disease, kidney disease and chronic obstructive pulmonary disease (COPD).

What is the exact definition and years used for the health maps?

The underlying causes of mortality and cancer incidence that form the basis of the maps presented were coded according to the International Classification of Disease (ICD) system.

This is an internationally recognised standard system for classifying the causes of morbidity and mortality. It also provides rules to establish the underlying cause of death from a medical certificate, allowing for a uniform derivation of information, aiding comparability between places and times.

Health condition ICD 9 ICD 10 Years
Lung cancer incidence (Trachea, bronchus and lung cancer incidence) 162 C34 1985–2009
Breast cancer incidence (female) 174 C50 1985–2009
Prostate cancer mortality (male) 185 C61 1985–2009
Malignant melanoma (skin cancer) incidence 172 C43 1985–2009
Bladder cancer incidence 188 C67 1985–2009
Leukaemia incidence 204–208 C91–95 1985–2009
Brain cancer incidence 191 C71 1985–2009
Liver cancer (liver and intrahepatic bile ductcancer) incidence 155 C22 1996–2009
Mesothelioma incidence 163 C45 1985–2009
Coronary heart disease (CHD) mortality 410–414 I20–I25 1985–2009
Chronic obstructive pulmonary disease (COPD) mortality 490–492, 494, 496 J40–J44, J47 1985–2009
Kidney disease (KD) mortality 580-599 N00-N39 1985–2009
Reproductive outcomes    
Low birth weight Live births <2500 g (denominator – live births) 1986–2009
Stillbirth Stillbirths >24 weeks (denominator – live and still births) 1992–2009
Why have we focused on wards?

A census ward has an average population of approximately 6,000 people; this is small enough to show local spatial variations whilst giving enough cases over a twenty-five year period to provide some statistical stability in the data.

Why use a 25 year period? Why not show health maps for the most recent year?

Many of the health conditions being mapped are very rare. A 25 year period contains enough cases of the condition in question to be able to map health risks as well as provide information on the long-term patterns of disease. The atlas does not aim to identify and investigate clusters of disease (for which different statistical techniques will be more appropriate).

What is the relative risk?

The relative risk is the risk of an area (ward) relative to average risk in England and Wales.

The maps use a two-colour scale to present nine categories of relative risk. The middle category (in white) represents a risk similar to the average for England and Wales and the different colours are used to show areas with risks higher than average (in orange) or lower than average (in purple). The national rates are provided for each health condition in the contextual information to the right of the health maps.

Why show risks, rather than rates?

We have not presented the rates for each ward because for many of the health conditions the numbers of cases are very small and rates based on a small number of cases are not reliable. By using risks instead of the rates we were able to use a technique called “smoothing” which accounts for chance fluctuation and makes the relative risk estimates more reliable.

What is deprivation and how is it measured?

Deprivation is a measure of socio-economic status and there are several different ways to represent this. In this atlas we have used the Carstairs index, which is a standard measure of deprivation, derived from information from the census. It describes deprivation in terms of access to goods and services customary in society. The Carstairs index combines four variables from ward level census data for households: 1) unemployment of males over 16 years of age, 2) car ownership, 3) numbers of people per room, 4) social class of economic active head of household.

Why are the maps adjusted for age and deprivation?

Deprivation is strongly associated with both the risk of a disease and also with important disease risk factors such as smoking. Therefore maps not adjusted for deprivation may reflect differences in lifestyle factors rather than possible environmental factors. Adjustment for deprivation is commonly used in epidemiological studies looking at environmental risk factors. While it is very important to adjust for deprivation, some pollutants tend to be higher in more deprived areas, so adjustment for deprivation may in some cases over adjust for (i.e. partially mask) risks associated with environmental factors.

Adjusting for age (also called standardising for age) adjusts for differences in disease risk that might result from different age structures in wards (e.g. more young people or more old people). This will particularly affect diseases that are strongly associated with age. For example, without adjustment, inner city wards that had much lower proportion of older residents might appear to have a lower risk of a cardiovascular disease or cancer when compared to rural wards that had higher proportions of older residents.Some or possibly all of the difference would be an artefact due to the differences in age structure of the population.

To explore how much age difference there is in different areas of the UK and how this has changed over time, we suggest you look at the ONS interactive tool "Ageing in the UK"

What is "smoothing"?

“Smoothing” was used when calculating the disease risks presented in the health maps. Smoothing is a statistical method used to adjust for chance fluctuations in disease risk that can occur when risks are calculated using small numbers of cases or small populations. The relative risk of each ward has been smoothed towards a combination of the national average and the averages of neighbouring wards. Smoothing will have relatively little effect on risks calculated for wards with larger numbers of cases and populations. A more detailed description of the Bayesian hierarchical model (BHM) used is provided at: [1] Richardson, S. and Best, N. Bayesian hierarchical models in ecological studies of health-environment effects. Environmetrics 14[2], 129-147. 2003 More information on the impact of smoothing on the distributions of risks is given in the print version of the atlas.

Why are the maps produced separately for men and women?

Health maps for England and Wales are shown separately for men and women as both disease patterns and disease risk factors can vary by sex. For example mesothelioma is a rare cancer that is very closely related to asbestos exposure. It is extremely rare in women because most asbestos exposure occurs in an occupational setting, such as construction work and ship building, where most jobs are done by men.

How common are the health conditions in the atlas?

The graph below illustrates how common the selected health condtions were in Great Britain. The data are from 2010 from Cancer Research UK (for all cancers) and the British Heart Foundation (CHD). The COPD and kidney disease data are from 2008 and the World Health Organisation and the Global Burden of Disease respectively. Rates for the first eight diseases are incidence rates (based on numbers of new cases). Rates for the next five diseases (which are starred and below the line) are mortality rates rather than incidence rates.

Figure 1: Age standardised rates in Great Britain of health conditions in the atlas

 Age standardised rates in Great Britain of health outcomes in the atlas
Why do the maps show prostate cancer mortality while incidence data are shown for the other cancers?

Most of the cancer maps are based on data on the number of newly diagnosed cases in a certain time period (incidence data). However, data on numbers of new diagnoses of prostate cancer (incidence) have been strongly influenced by the introduction of prostate-specific antigen (PSA) testing. We chose to show mortality for prostate cancer because geographical patterns in mortality rates will be less influenced by geographical variations in PSA testing than incidence rates, so are a better estimate of underlying prostate cancer risk.

How are the maps for coronary heart disease (CHD)/chronic obstructive pulmonary disease (COPD)/kidney disease different from the cancer maps?

The maps show mortality (deaths) for CHD, COPD and kidney disease but incidence (new cases) for cancers (except for prostate cancer). This is because there is a national register that records new cases of cancers, but no national register to record new cases of CHD, COPD or kidney disease. Mortality data (available from ONS) has been used in the atlas where incidence data are not available, except for prostate cancer.

My ward has a higher risk than the neighbouring ward! Should I be worried?

The atlas shows average risks for an area (ward) compared with the national average, this is not the same as the risk to an individual who will have a collection of individual risk factors (genetic, lifestyle, previous disease and life events). Also, the health maps show relative risk not the absolute risk.

The relative risk tells us how much more or less likely a health condition is in one group compared to another group. For example the risk of bladder cancer in smokers has been estimated to be three times higher than in non-smokers; this is not telling us anything about the actual risk of developing bladder cancer. This difference between the types of risk is explained well on the Cancer Research UK website:

The understanding uncertainty website also has a useful animation to help understand the difference between the two types of risk:

The health maps only go to 2009 – why?

The data originally available to SAHSU at the time of analysis was until 2009. The data were checked and cleaned before being analysed and the maps took time to design and produce. The book was extensively peer reviewed and several workshops with Sense About Science have been held to try to present the information in a way to make it accessible and understandable to a wide audience. Following this changes were made and there were a further nine months between submission to publisher and publication of the book.

My aunt lives in a ward with higher than average relative risk. Should we encourage her to move?

Risks presented are long-term averages for the area for the period 1985-2009. There are many different individual risk factors for disease, including age, sex, genetic factors, whether or not a person smokes and their diet and exercise levels as well as pre-existing conditions such as high blood pressure. . So the reasons for higher (or lower) risks in an area are complex and do not necessarily represent risks for an individual.

Environmental maps:

Why not show the environmental maps from a more recent year?

Lack of availability of data at a high spatial resolution for most or all of England and Wale affected which years of data we were able to provide for the environmental maps. Several of the maps relate to 2001 and this falls within the time period of the health mapS.

The maps represent specific time periods and may not be representative of current levels.

Why can I not see census wards for some of the environmental maps?

The environmental maps are presented at the highest resolution possible depending on the data available. For example, air pollution estimates are presented at 100 m grids, disinfection by-products at water supply level and pesticides at census ward level.

When I select a postcode a value for the environmental agent is presented, what is this based on?

When a postcode search is carried out, the atlas looks for the corresponding geographic point representing the postcode centroid (a postcode centroid is the geometric centre of the actual postcode area). This centroid has a specific pair of x-y coordinates which is then used to locate and extract the estimated measure from the underlying modelled environmental agent data.

My area has high levels of an environmental agent. Does that mean I have a high level of exposure?

The maps show average concentrations for each area but concentrations may vary within that area. Concentration in an area is not the same as the exposure for an individual living in that area. This may be affected by factors such as their job, house, pre-existing state of health, age and lifestyle choices. The important metric for human exposure is the dose – the amount of a pollutant that actually enters the human body. Dose may further be considered as the dose received by the organ or organs affected, or the biologically effective dose.

Note that the environmental maps at area level are not the same as dose to an individual.

Does comparing the environmental maps with the health maps provide information about environmental causes of disease?

No. Simple comparison of the environmental and health maps cannot be used to show causal associations. More information about other factors that might affect disease risk and further study would be needed, including at individual level, before any conclusions about possible causal links could be made.

My area shows high sunshine duration – are you saying greater sunshine duration is bad?

Sunshine is a major source of ultraviolet (UV) exposure. Skin exposure to sunshine increases vitamin D levels, which is good for general and bone health, but is also associated with increased risk of skin cancer. The amount of solar radiation individuals are exposed to is influenced by the amount of radiation that reaches the earth’s surface as well as individual factors such as skin colour, sunbathing, holidays in sunny places and use of sun protection. Therefore, the sunshine map does not directly indicate dose of ultraviolet (UV) exposure for individuals living in those areas.

Why is my area missing from the maps of agricultural pesticides or chlorination disinfection by-products?

The maps of agricultural pesticides were originally produced for a separate study and the data came from The Pesticides Usage Survey (PUS), conducted by The Food and Environment Research Agency. Areas that were not surveyed in the PUS will be missing from the maps and there were no data available to be able to provide equivalent maps for Wales.

The chlorination disinfection by-products data were obtained for a specific study and was supplied by various water companies. Not all water companies in England and Wales participated in the study, which is why there are areas missing from the maps.

I am concerned about the high levels of pesticides/THMs/air pollution in my area and believe they may be linked to my health, what should I do?

The published version of the atlas provides more detailed information on what is known about risk factors for each of the health conditions, and potential health risks associated with the environmental agents in the maps. Public Health England is also a useful resource to find out about potential health risks from environmental agents. The NHS choices website is a useful resource for learning more about health conditions and their risk factors.

We suggest that you contact your GP, if you have any specific health concerns.

My area has a higher risk of leukaemia and there is an industrial installation here – is this the cause of the higher risk?

There are many possible risk factors for leukaemia, including exposure to ionising radiation, occupational exposures and smoking, and these may differ by area. Before making any assumptions about a possible link with e.g. an industrial installation, more information about other factors that might affect disease risk and further study would be needed, including at individual level, and in other areas.

About us:

What is SAHSU?

The UK Small Area Health Statistics Unit (SAHSU) at Imperial College London is a national centre established in 1987 to investigate potential environmental risks to population health, with particular emphasis on the use and interpretation of routine health statistics (such as births, deaths and hospital admissions) and development of the methodology for small-area studies. It is part of the MRC-PHE Centre for Environment and Health, a national centre for research and training in environmental health.

Who are you funded by?

SAHSU is part of the MRC-PHE Centre for Environment and Health funded by the Medical Research Council and Public Health England. SAHSU is based at Imperial College London. SAHSU’s core funding comes from Public Health England although specific research studies may have additional funding from research councils, charities and government departments.

How to use the application:

How do I switch between the health outcome and environmental agents?

At the top right hand side of the screen are two buttons, the 1st will switch between the health outcome and environmental agent maps. The second button labelled “Home” will take you back to the main website.

How do I select a different health condition or environmental agent?

The drop down menu in the top left of the screen will enable you to change between different health conditions or environmental agents.

How do I switch between maps for each sex?

At the top right hand side of the maps there are two buttons with the standard gender symbols next to them. Clicking on the button next to the relevant symbol will switch between the male and female maps.

How do I perform a postcode search?

Type the postcode into the box called ‘Postcode search’ (top right of map under ward name) and then click search.

How does the clear selection function work?

The clear button allows you to "unlock" your current selection. When a ward is selected all components (map, chart, legend and information box) will not update when hovering over different areas. The clear button can be used to unselect the ward and let the application update.

How can I zoom out to see the entire country?

Click on the plus or minus buttons to zoom in or out as required.

Can I use the maps or text from the atlas?

This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.

You are free to: Share — copy and redistribute the material in any medium or format for any purpose, even commercially. The licensor cannot revoke these freedoms as long as you follow the license terms. Under the following terms: Attribution — You must give appropriate credit, provide a link to the license, andindicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use. NoDerivatives — If you remix, transform, or build upon the material, you may not distribute the modified material. No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits. Notices: You do not have to comply with the license for elements of the material in the public domain or where your use is permitted by an applicable exception or limitation. No warranties are given. The license may not give you all of the permissions necessary for your intended use. For example, other rights such as publicity, privacy, or moral rights may limit how you use the material.