2. APHEIS Air Pollution and Health: A European Information System Bertil Forsberg, Sylvia Medina, Antoni Plasència, Ross Anderson, Lucia Artazcoz, Klea Katsouyanni, Michal Krzyzanowski, Hans-Guido Mücke, Emile de Saeger and Joel Schwartz On behalf of the APHEIS group APHEIS is co-funded by the Pollution Related Diseases Programme of DG SANCO of the European Commission (contract No. SI2.131174 [99CVF2-604] -SI2.297300[2000CVG2- 607]-SI2.326507[2001CVG2-602]) and by the participating APHEIS institutions

4. l An epidemiological surveillance system l Aims to provide up-to-date and easy-to-use information on health impact assessment (HIA) of air pollution l For decision makers, environmental and health professionals, media and general public l Enables better-informed decisions about political, professional, personal issues What is Apheis

5. How Apheis works APHEIS coordination centre Paris and Barcelona Advisory groups Exposure assessment Epidemiology Statistics Public health Health impact assessment Participating APHEIS Cities Technical committee Exposure assessment Epidemiology Statistics Public Health Health Impact Assessment City committee NEHAPs Local/national authorities Medical/environmental sciences Citizens Local/regional coordinator

6. l Created five advisory groups: public health; health-impact assessment; epidemiology; exposure assessment; statistics l Drafted guidelines for designing and implementing the surveillance system, and for developing a standardised protocol for data collection and analysis for HIA l Review of capacities for HIA in institutions of participating cities Actions, steps and results during the first year

8. l Implement or adapt organisational models designed during first year l Collect and analyse data for health-impact assessment l Prepare different health-impact scenarios l Prepare HIA report in standardised format (HIA in 26 cities) Actions, steps and results during the second year

9. 1. Specify exposure * PM10, BS * Urban background stations Five main steps in HIA

10. 2. Define the appropriate health outcomes * Acute effects - Premature mortality excluding accidents and violent deaths - Hospital admissions for respiratory diseases 65+ age group - Hospital admissions for cardiac diseases all ages * Chronic effects - Mortality (all natural causes) Five main steps in HIA

11. 3. Specify the exposure-response functions * Short-term exposure: APHEA2 Five main steps in HIA Health indicator RR for 10 µg/m 3 95%CI Total mortality All ages ICD9 <800 (Katsouyanni et al, 2001) 1.006 1.003-1.008 Respiratory hospital admissions 65 years + ICD9 460-519 (Atkinson et al, 2001) 1. 001 1.00-1.009 Cardiac hospital admissions all ages ICD9 410- 414.427.428 (Le Tertre et al, 2002) 1.011 1.004-1.018

12. 3. Specify the exposure-response functions * Long-term exposure: HIA in Austria, France and Switzerland based on two American cohort studies (Künzli et al, 2000). Five main steps in HIA Health indicator RR for 10 µg/m 3 95%CI Total mortality 30 years + ICD9 <800 1.043 1.026-1.061

13. 4. Derive population baseline frequency measures for health outcomes 5. Calculate number of attributable cases in target population (based on attributable proportion) Five main steps in HIA

14. Demographic characteristics * Nearly 39 million inhabitants in Western and Eastern Europe * Proportion of people over 65 years: 15%, with highest proportion in Barcelona and lowest in London Descriptive findings

15. Air pollution levels * Black smoke measurements provided by 15 cities: Athens, Barcelona, Bilbao, Bordeaux, Celje, Cracow, Dublin, Le Havre, Lille, Ljubljana, London, Marseille, Paris, Rouen and Valencia * PM 10 measurements provided by 19 cities: Bordeaux, Bucharest, Budapest, Celje, Cracow, Gothenburg, Lille, Ljubljana, London, Lyon, Madrid, Marseille, Paris, Rome, Seville, Stockholm, Strasbourg, Tel Aviv and Toulouse Descriptive findings

16. Annual mean levels and 10th and 90th percentiles of the distribution of PM10 Descriptive findings

17. Annual mean levels and 10th and 90th percentiles of the distribution of black smoke Descriptive findings

18. Acute effects scenarios * Reduction of PM10/BS levels to a 24-hour value of 50 µg/m 3 (2005 and 2010 limit values for PM10) on all days exceeding this value * Reduction of PM10/BS levels to a 24-hour value of 20 µg/m 3 (to allow for cities with low levels of PM10/BS) on all days exceeding this value * Reduction by 5 µg/m 3 of all the 24-hour daily values of PM10/BS (to allow for cities with low levels of PM10/BS) Health impact assessment findings

19. Chronic effects scenarios * Reduction of the annual mean value of PM10 to a level of 40 µg/m 3 (2005 limit values for PM10) * Reduction of the annual mean value of PM10 to a level of 20 µg/m 3 (2010 limit values for PM10) * Reduction of the annual mean value of PM10 to a level of 10 µg/m 3 (to allow for cities with low levels of PM10) * Reduction by 5 µg/m 3 of the annual mean value of PM10 (to allow for cities with low levels of PM10) Health impact assessment findings

20. Potential benefits of reducing daily PM10 levels by 5 µg/m 3 - Number of deaths per 100 000 inhabitants (95% confidence limits) attributable to the acute effects of PM10 HIA findings: PM10 acute-effects scenarios

21. Potential benefits of reducing annual mean values of PM10 by 5 µg/m 3 - Number of deaths per 100 000 inhabitants (95% confidence limits) attributable to the chronic effects of PM10 HIA findings: PM10 chronic-effects scenarios

22. Standardised protocol for data collection and analysis Conservative approach * Did not consider newborn or infant mortality separately * Did not consider many other health outcomes listed by WHO * Did not consider independent effect of ozone * Used range of reference levels in different scenarios Interpretation of findings

23. Transferability of Exposure-Response (E-R) functions * Short-term exposure: Question avoided by using E-R functions developed by APHEA 2 * Long-term exposure: Open question - used U.S. E-R functions Interpretation of findings

24. Conclusions * Our HIA provides a conservative but accurate and detailed picture of the impact of air pollution on health in 26 European cities, and shows that air pollution continues to threaten public health in Europe. * Even very small and achievable reductions in air pollution levels have an impact on public health * This impact justifies taking preventive measures even in cities with low levels of air pollution Interpretation of findings

25. The Apheis second-year report also provides an individual picture for each of the 26 cities that describes: * Local characteristics * Air pollution sources * Exposure data * Health outcomes * HIA * Specific comments City by city reports

27. To keep our HIA as accurate and up-to-date as possible: * produce new exposure-response functions on short-term effects of AP * calculate years of life lost or reduction in life expectancy, in addition to the attributable number of deaths based on long- term effects Actions, steps and results during the third year

28. To fulfill our mission of making our learnings available to the broadest possible audiences, and to evaluate the usefulness of our work on HIA among those who need to know: * Explore and understand how best to meet the information needs of government decision and policy makers concerned with the impact of air pollution on public health and * Understand how to meet those needs in terms of content and form Actions, steps and results during the third year

29. l Calculate costs to society of health effects of AP l Involve Apheis more closely in municipal, regional, national and European programmes (NEHAPS, AIRNET,CAFE, EUROHEIS). Actions and steps to be undertaken in future years

30. The Apheis programme * Multiyear, multiphase, proactive * Answers key questions on air pollution and public health in Europe for a broad range of audiences * Translates epidemiological findings into decision-making tool * Bridges the gap between data and action The broad view

31. The Apheis programme * First broad-based European HIA of air pollution * Consistent with other HIAs on air pollution worldwide * Provides information on both local and European levels simultaneously with two main benefits: - Local data can be used for local decision making (e.g. urban and transport planning; devising of steps to reduce air pollution) - European authorities gain a global view on air pollution and public health The broad view

32. The Apheis programme * One more brick in the wall of evidence that air pollution continues to threaten public health * Each phase builds on learnings of previous phase * Needs ongoing commitment and funding by European Commission and EC member states * For further information visit www.apheis.org The broad view