|
About the Department
At a glance
About NDPBs
NDPBs
Advisory Bodies
COMEAP |
|
Home
Public Health
Air Pollution
COMEAP Meetings |
|
COMMITTEE ON THE MEDICAL EFFECTS OF AIR POLLUTANTS DRAFT Minutes of the meeting held on Friday 3 November 2000 in Conference Rooms GL/GM Eland House, Blessenden Place, London SW1 Present:
ITEM 1. APOLOGIES FOR ABSENCE AND ANNOUNCEMENTS 1. Apologies for absence had been received from Professor Jon Ayres, Professor Peter Blain, Dr Sarah Walters, Dr Tim Carter (DETR), Dr Angela Cooper (MRC) and Mr George Morris (SCIEH). 2. The Chairman announced that Mr John Crook had left DH in order to take up a post with Kensington and Chelsea Health Authority. 3. The Chairman welcomed Dr Brian Miller (IOM) who was present for Item 4, and noted that Dr Kate Leverton (Imperial College) and Ms Emma Jenkins (DH) would be attending later for Items 6 and 7, respectively. ITEM 2. MINUTES OF THE MEETING HELD ON 16 JUNE 2000. COMEAP/MIN/2000/2 4. The following amendments to the minutes were noted: (i) Professor Roy Harrison's name was added to the list of Members present on page 1. (ii) Item 8: It was noted that the paragraph headed Public availability should read: "The review paper could be made publicly available given the agreement of the DH Toxicology Unit at Imperial College." (iii) Paragraph 27 was amended to read: "…review had been commissioned from the DH Toxicology Unit at Imperial College." 5. Subject to the above amendments, the minutes were agreed as an accurate record of the meeting. 6. The Secretariat noted that subject to Dr Andrea Venn's agreement with the statements made in paragraph 29, the minutes could be made publicly available without further amendments. Shortly after the meeting, Dr Venn wrote to Secretariat saying that she had no objections to the text in paragraph 29 being available to the public. 7. The Chairman raised the matter of the need to place information on the COMEAP website as soon as was practicably possible, noting that this was in line with the Government's policy of openness and transparency. It was pointed that as a result, most of the COMEAP papers being considered at the present meeting would either have been placed on the website already, or would be made available over the forthcoming weekend, under, in the main, a new section called "Items Currently under Consideration." It was noted that this section of the website made clear that the papers might not reflect the final view of the Committee, and the Secretariat further pointed out that the section did not include draft papers of research work which was either incomplete, in the process of being completed, or which was being prepared with a view for future publication in a peer-reviewed journal. 8. The Chairman also commented that the Secretariat had written to 5 peer-reviewed journals for a view on how the procedure noted above might affect future publication of papers. The Chairman pointed out that the editor of the New England Journal of Medicine had made it clear that from an editorial policy view point, it was not prepared to publish any work that had previously been made available publicly in any form. The Secretariat undertook to keep Members informed of developments with regard to this issue. [ACTION: Secretariat] ITEM 3. MATTERS ARISING 9. Members were informed that statements on MTBE, the application of time-series coefficients to areas around industrial processes and communicating with the public had been finalised and placed on the COMEAP website since the last meeting. 10. With regard to the COMEAP statement on MTBE, the Secretariat noted that the Committee had commented that: "In the UK, petrol typically contains up to 5% MTBE". The UK Petrol Industries Association had since informed the Secretariat that whilst a limited quantity of the petrol sold in the UK can contain up to 5% MTBE, the average concentration of MTBE is below 1%. 11. Members were told that the COMEAP comments on the DETR Air Pollution Service had been place on the COMEAP website. This issue had been dealt with by correspondence. 12. It was noted that the Good Practice Guidelines for Investigating the Health Impact of Local Industrial Emissions were now on the COMEAP website. The Secretariat commented that a hardcopy version, available as a free DH publication, should be available before Christmas. ITEM 4. LONG-TERM EFFECTS OF PARTICLES ON HEALTH. COMEAP/2000/17 HEI REANALYSIS OF THE HARVARD SIX-CITIES STUDY AND THE AMERICAN CANCER SOCIETY STUDY OF AIR POLLUTION AND MORTALITY. COMEAP/2000/17a PAPER BY KÜNZLI ET AL: PUBLIC HEALTH IMPACT OF OUTDOOR AND TRAFFIC RELATED AIR POLLUTION: A EUROPEAN ASSESSMENT. COMEAP/2000/17b Public availability: These papers are available on the COMEAP website. 13. The key points included in COMEAP/2000/17 were introduced. Members were reminded that in the COMEAP Quantification Report published in 1998 (QUARK Report) it had been decided not to quantify the long-term effects but that it had been noted that such effects might be substantial. At the time of writing the QUARK Report, a number of uncertainties, including the limited number of studies available, the possible insufficient control for socio-economic factors and the lack of UK or European studies had led to this view. Since 1998 a number of developments had occurred: (i) The HEI Reanalysis: This examined a further range of socio-economic factors and had not found evidence of major confounding by such. It was thought that this might lessen Member's concerns. (ii) Limited progress had been made on further studies including one in 7th Day Adventists,1 another of the effects of early life exposure to air pollutants and studies in the Netherlands and France, which had begun but had not yet been completed. (iii) The report collated by the Institute of Occupational Medicine (IOM Report) which had developed a methodology for modelling the effects of long-term exposure to air pollutants on the UK population.2 This study had used a coefficient derived from the American Cancer Society study.3 14. The Committee's advice was requested on whether these developments were sufficient to lead it to change its opinion from that expressed in 1998. It was also asked whether, if the Committee still had reservations about quantification there remained the option of providing illustrative examples of the possible size of the effects if the risks described by the US coefficients applied in the UK. It was noted that the sub-group set up the last meeting of COMEAP had developed some illustrative examples and that these were presented in the paper COMEAP/2000/17. 15. The Secretariat was congratulated on the papers (COMEAP/2000/17, 17a, 17b). It was agreed that these were a lucid summary of a difficult area. Members felt that the HEI analysis confirmed the interpretation of the original authors of the American cohort studies and had not revealed any major new confounding factors. However, spatial variation of effects across the US had been described and it was felt that this increased the uncertainty bounds about the central estimate of likely effects. Members pointed out that formal confidence intervals reflected only uncertainty due to sampling. It was felt that the uncertainty bounds should include zero: ie, no effect. Members commented that it remained uncertain as to whether the associations reported in the cohort studies reflected a causal, ie, real effect. It was agreed that even if it was accepted that the association was likely to be causal the possibility that the effect was very small, approximating to zero, remained. The Committee expressed the view that though uncertainties regarding causality remained it was more likely than not that the associations were causal and that it would be sensible to move beyond the position taken in 1998 and to provide some quantification of effects. 16. Having come to this view, Members moved on to discuss the most appropriate method for calculation and how the results of that calculation should be presented. In considering methods for estimating the size of the effect in the UK, a recent paper by Nino Künzli (COMEAP/2000/17b) was considered. Members noted that the authors ascribed some 6% of overall mortality to the effects of air pollutants (based on long-term effects). Some doubts were raised regarding the method of calculation used to arrive at this figure. Members felt that good work on the acute effects of air pollutants and on baseline rates had been undertaken in the study but that, for long-term effects, the method of analysis used by the IOM Report was more appropriate than the method used by the authors of this paper. Specifically, it was felt to be better to express the effects of long-term exposure to air pollutants in terms of a loss of life expectancy rather than in terms of number of deaths. 17. Members agreed that the effects of long-term exposure to air pollutants should be presented in terms of the effects likely to be experienced by people alive now as compared with the effects that might be experienced by a new birth cohort. It was also agreed that it should be made clear that any presentation of effects in terms of an average loss of life should be explained: there would be a distribution of effects across the population, ie the loss of life for some would be greater than that suggested by the average figure; for others the loss of life would be less. It was further agreed that it might be helpful to put the estimates of the effects of long-term exposure to air pollutants, especially particles, in context with risks from other hazards, eg, smoking. However, Members agreed that the effect on mortality reflected in the time-series studies would remain and might be used to define a lower estimate of the likely effects of long-term exposure to air pollutants. 18. Members noted that the uncertainty reflected in the effects of spatial variation in the US studies meant that a lower estimate of zero/long-term effects should be included. Thus, it was agreed that the range of possible effects should include: (a) zero/long-term effects, ie, a low estimate made from the time-series studies (further work is needed to produce this figure); (b) a low estimate made from the cohort studies; (c) a mid estimate made from the cohort studies; and, (d) a high estimate made from the cohort studies. (b) and (d) would be made by applying the confidence intervals produced by the authors of the American Cancer Society study. It was agreed that further work by Members would be needed to provide advice on the relative weight which the Committee placed on each of these estimates. [ACTION: Members] 19. It was discussed whether it was acceptable to present the results in terms of the effects of 10 µg/m3 reduction in PM10. It was concluded that such a reduction was not implausible over the longer term but some justification would need to be provided in the statement. It was also agreed that no threshold assumption should be made though concern was expressed regarding the reliability of predictions of effects as concentrations of particles fell and that the impact of competing causes of death might be increased at low concentrations. Some concern was expressed over the assumptions of linearity, in the relationship between concentration and effect, though it was agreed that the impact of errors from this assumption was likely to be small. With regard to this last point, Dr Pritchard agreed to prepare a paper looking theoretically at this assumption. A number of Members pointed out that they were uncomfortable in talking purely about changes in mass concentration and stressed that if the chemical composition, or size distribution, of the ambient aerosol changed then predictions of effects might well become unreliable. [ACTION: Dr Pritchard, DETR] 20. It was agreed that the Secretariat would draft a statement incorporating the above points and that this statement would be discussed by the COMEAP sub-group dealing with long-term effects of particles on health. Once agreed by the sub-group, the statement would be circulated to the full Committee who would be invited to submit further comments. Once an agreed position had been reached, the Secretariat said that this statement would be published on the COMEAP website. [ACTION: Secretariat, COMEAP sub-group, Members] References 1. Abbey DE, Nishino N, McDonnell WF, Burchette RJ, Knutsen SF, Beeson WL, Yang JX. Long-term inhalable particles and other air pollutants related to mortality in nonsmokers. Am J Respir Crit Care Med 1999; 159:373-382. 2. Hurley JF, Holland MR, Markandya A, Miller BG, Anderson HR, Ayres JG, Donnan PT, Harrison RM, King K, Stedman JR, Stevenson KJ. Towards assessing and costing the health impacts of ambient particulate air pollution in the UK. Edinburgh: Institute of Occupational Medicine, 2000. 3. Pope CA, Thun MJ, Namboodiri MM, Dockery DW, Evans JS, Speizer FE, Heath CW. Particulate air pollution as a predictor of mortality in a prospective study of US adults. Am J Respir Crit Care Med 1995; 151:669-674. ITEM 5. UPDATING THE COEFFICIENTS USED BY COMEAP IN THE QUARK REPORT TO ESTIMATE THE EFFECTS OF PARTICLE ON HEALTH IN THE UK. COMEAP/2000/18, 18a, 18b, 18c Public availability: These papers are not publicly available since they are based on preliminary results from the APHEA II study kindly provided, in confidence, by the authors. 21. Introducing this item, the Secretariat reminded Members that in the 1998 QUARK Report, coefficients derived from time-series studies had been used to quantify the effects of ozone, sulphur dioxide and particles on all-cause mortality and respiratory admissions to hospitals in the UK. The effects of nitrogen dioxide and carbon monoxide had not been quantified. Also, the effects of air pollutants on admissions to hospital for treatment of cardiovascular diseases had not been quantified. Current examination of the objectives for particles in the UK National Air Quality Strategy required that Members' opinions on the effects of short-term variations in concentrations of particles should be revisited. 22. Points made in detail in COMEAP/2000/18 were summarised. It was pointed out that papers summarising results from the APHEA II study were being prepared for submission to peer-reviewed journals. Some preliminary data had been made available to the Committee, in confidence, by Professor Ross Anderson. It was pointed out that as this information had been provided both in confidence and in an unfinished form it would not be placed on the COMEAP website. It was noted that the APHEA data provided information on the relationship between concentrations of particles and all-cause mortality. Information on all age respiratory admissions was not currently available though this could be provided if necessary. As regards the effects of particles on admissions to hospital for treatment of cardiovascular diseases, the QUARK report of 1998 had noted that studies were very limited in number and it had been decided not to use these for quantification of effects. Since then more papers dealing with cardiovascular effects had appeared and a number of potential mechanisms had been put forward. Members were also told that the Department of Health had funded a study at St George's Hospital which involved the setting up of a large database of time-series studies. This work was not yet complete, but some preliminary results had been provided. Finally, completion of the NMMAPS study was noted. This had also generated coefficients relating daily levels of particles with effects on health. Members' attention was drawn to a table which compared the findings of these different studies. 23. It was pointed out that in the APHEA II work all the statistical analyses were being conducted by one centre. This contrasted with the APHEA I work in which each collaborating centre had undertaken their own statistical analysis. Members considered that it would make sense to retain the 1998 coefficients for all-cause mortality and respiratory hospital admissions for the moment since it would be better to review the coefficients in another 6 months or so once the APHEA II studies were published and further results from the NMMAPS study were available. Members pointed out that though different studies yielded different coefficients these were not dissimilar to the coefficients selected by QUARK in 1998. The overlap of confidence intervals about the coefficients (those used by QUARK and those reported more recently) was noted and it was agreed that the coefficients could not be distinguished from one another with confidence. 24. It was agreed that information on the association between day-to-day variations in levels of particles and in admissions for cardiovascular diseases had now developed to a point where it was worth considering again whether quantification could be attempted. Professor Ross Anderson offered to draw together data on this; these data would then be included in a short paper to be prepared by the Secretariat and circulated to Members and it was agreed that a short COMEAP statement would be prepared as soon as these data were available. [ACTION: Professor Anderson, Secretariat]
ITEM 6. EFFECTS ON HEALTH OF ALDEHYDES IN AMBIENT AIR. COMEAP/2000/19 FORMALDEHYDE AND ALLERGIC SENSITISATION. COMEAP/2000/19a GLUTARALDEHYDE AND ADDENDUM. COMEAP/2000/19b and 19b/Addendum Public availability: These papers are available on the COMEAP website 25. Members were reminded that the effects of ambient levels of aldehydes had been discussed at the COMEAP meeting in June (COMEAP/2000/12). At that meeting the review prepared by the DH Medical Toxicology Unit at Imperial College had been considered. The text of the review had now been updated and it was intended that this would be placed on the COMEAP website as soon as permission had been given for the inclusion of certain tables in the report. Permission would be sought from the original authors of these tables. A number of points raised by Members in June were considered. Details were provided in COMEAP/2000/19a and COMEAP/2000/19b. It was noted that the issue of respiratory sensitisation to formaldehyde was a complex one but there was no doubt that formaldehyde could act as a respiratory irritant at high concentrations. Although most papers did not demonstrate a link, two papers suggesting that exposure to formaldehyde in homes could lead to allergic sensitisation in children had been identified. These were attached to COMEAP/2000/19a. Members had also requested information on glutaraldehyde. It was noted that this was the only aldehyde currently listed in the UK as a cause of occupational asthma. A paper recently published in the journal Occupational and Environmental Medicine was circulated as was a letter from Dr Dick Derwent (Meteorological Office) giving information about likely levels of glutaraldehyde in the UK. A number of amendments to the review prepared by the DH Toxicology Unit were considered and agreed. These are appended to the minutes at Appendix 1. Members had also asked about the genotoxicity of formaldehyde and a note had been added to the review clarifying this point. It appeared that though formaldehyde was agreed to be genotoxic, its action depended on levels being high enough to produce cytotoxicity. At levels of exposure that did not cause cytotoxic effects, the development of cancer was regarded as very unlikely to occur. It was agreed that this point should be submitted to Professor David Phillips (DH Committee on Carcinogenicity) before being finalised. [ACTION: Secretariat] 26. It was agreed that ambient levels of formaldehyde in the UK were below those likely to be associated with an increased risk of cancer or respiratory sensitisation. Information on levels of formaldehyde measures in the Marylebone Road, London, and at an industrial site in Hull were presented. The levels were as follows:
It was noted that these concentrations were well below the threshold for irritant effects (0.1 mg/m3). Attention was also drawn to a paper published in 1990 by Dr Michal Krzyzanowski, which had shown that indoor exposure to formaldehyde had been associated with increased complaints of symptoms amongst asthmatic subjects. 27. It was agreed that the draft COMEAP statement on the effects of health of aldehydes in ambient air should be revised to take account of the following points: (i) aldehydes in ambient air might cause irritant effects but it was not known for certain whether this was the case; (ii) some reference to background levels should be included; (iii) some reference to indoor air levels where effects had been reported should be included; (iv) the use of the term occupational asthma should be clarified. 28. The Secretariat undertook to produced a revised statement for circulation to Members and agreement by correspondence. [ACTION: Secretariat]
ITEM 7. OFFICE OF SCIENCE AND TECHNOLOGY CONSULTATION OF A PROPOSED CODE OF PRACTICE FOR SCIENTIFIC ADVISORY COMMITTEES. COMEAP/2000/20 29. The Chairman welcomed Ms Emma Jenkins who had prepared this paper for the Committee. The Secretariat commented that the Proposed Code of Practice issued by the Office of Science and Technology suggested reasonable measures and procedures to be taken by Committees. It was noted that the majority of the suggested procedures were already being followed by COMEAP. However, the Secretariat drew attention to some concerns regarding the timescale for publication of the full Committee meeting minutes. The Code suggested that minutes should be published in full (ie, on the website) within three weeks of the meeting. The Secretariat would have to consider the administrative implications of such a tight timescale, and Members would need to be aware that such a timescale would dramatically reduce their time to comment and amend the full minutes. Whilst this would not be a problem for the brief version of the minutes published shortly after each meeting, it would have serious implications for the way in which Members commented and signed up to the full version of the minutes. It was noted that comments on this particular issue or any other items in the Code could either be sent into the Secretariat or passed directly on to the Office of Science and Technology by Friday 1 December 2000 at the latest. 30. Members raised a point of concern regarding openness and transparency about unpublished results of research work and how this might affect future publication of the work. The Secretariat undertook to look into this matter further. A further point was raised with relation to how individual Members should respond to journalists/the media if approached for a personal opinion on a particular statement when, whilst this might have been agreed overall in a collective manner, an individual might have had some minor dissension to some of the points made. Again the Secretariat undertook to make some response to this. [ACTION: Secretariat] 31. With regard to the Minutes, Members felt that the suggested proposals would be difficult to meet, regardless of how quickly the Secretariat could produce the minutes, and felt that there needed to be some guidance on what constituted a final response from the Committee as a whole, ie, did every single Member need to contacted? (Members could easily be away over a 1 or 2 week period.) It was agreed that any further comments should be sent either to the Secretariat or to OSC directly [ACTION: Secretariat, Members] ITEM 8. REAPPOINTMENTS TO THE COMMITTEE 32. The Secretariat explained the need to replace Members who would be leaving the Committee in 2001. Members could, in future, serve for up to 9 years though reapplication for Membership would be required at 3 year intervals. The attached chart (Appendix 2) shows the succession planning for the Committee until 2010. Steps to replace Professor Holgate, Professor Ayres, Professor Burney and Dr Sarah Walters were currently being taken. The positions held by Mr Hurley, Professor Strachan and Lord Harris would be advertised although they were eligible to reapply. Professor Blain was a member because he was Chairman of the COC so his position would not be advertised.
ITEM 9. DAILY DEATHS ARE ASSOCIATED WITH COMBUSTION PARTICLES RATHER THAN SO2 IN PHILADELPHIA (DISCUSSION OF PAPER BY SCHWARTZ). COMEAP/2000/21 33. The Secretariat introduced this item noting that the paper (which had been circulated primarily for information purposes) was important in that it cast doubt on the importance of sulphur dioxide as a causal explanatory variable for the day-to-day variations seen in health outcomes, including mortality, in time-series studies. It was noted that, if the author's conclusions were correct, they would have significant implications for estimates made by COMEAP of the overall effects of air pollutants on health. It was also commented that the paper described the use of sophisticated analytical techniques and advice on these was requested. 34. It was agreed that the methodology used by the author (Schwartz) was likely to be robust. Members wondered, however, about the statistical power of the technique to exclude a small effect of sulphur dioxide. Two important points were raised: (i) It was noted that the data analysed by Dr Schwartz extended over 15 years and over summer and winter seasons. The observation that there had been little variation in concentrations of pollutants monitored during warm seasons over later years was made, and it was suggested that there might have been a change in the pollutant mixture over the period analysed. It seemed possible, at least, that coal burning played a larger role in the early years and pollutants emitted by motor vehicles a larger role in the later years. (ii) It was suggested that the inclusion of SO2 as an explanatory variable attenuated the size of the effects attributable to particles. Thus the calculation undertaken in the QUARK report of 1998 for the effects of particles might have underestimated the size of the effects. It was felt that it was possible that adding estimates of the effects of particles and sulphur dioxide might yield a figure not too dissimilar from that which might have been attained had an unattenuated coefficient for particles been applied. 35. It was pointed out that a caveat regarding double counting had been included in the 1998 QUARK Report. It was also noted that the statistically significant association between sulphur dioxide and daily hospital admissions for most respiratory diseases was puzzling, given that the levels of sulphur dioxide were very low, and it was felt by some that the effects were biologically implausible at these levels. The suggestion was made that different measures of particle concentrations (PM10 versus Black Smoke) might be used in an attempt to dissect the problem, and it was argued that sulphur dioxide produced outside the city might contribute, via this formation of sulphate particles to levels of PM10 within the city. Black Smoke measurements, on the other hand, would not reflect the presence of sulphate particles. It was agreed that DETR would provide information on the sources of sulphur dioxide in urban areas as this would help to judge whether sulphur dioxide was acting as a surrogate for something else. [ACTION: DETR] 36. The point was made that, when reviewing the effects of sulphur dioxide, nitrogen dioxide or carbon monoxide upon health, it should be asked whether these pollutants had: (i) a modifying effect on the effect of particles; (ii) an independent effect, with regard to the effect of particles; and (iii) any additional effect, with regard to the effect of particles. The Committee concluded that Dr Schwartz's paper was a most important contribution to the literature and that further work of this nature should be monitored closely.
ITEM 10. PAPERS FOR INFORMATION 10.1 Guidelines 2000 Scientific Advice and Policy Making. COMEAP/2000/22 37. An earlier version of these Guidelines was issued to Members at the June meeting. They have since been amended and an up to date version had been circulated to Members by post.
ITEM 11. ANY OTHER BUSINESS 11.1 Air Pollution Research Initiative 38. The Secretariat informed the Committee that it had been unsuccessful in its bid for funding for a further round of research on the effects of air pollutants on health. Members were disappointed by this. It was agreed that a very useful research community had been set up in the UK which had made a very important contribution to the area on a worldwide basis. Not only did this need to be maintained but it would be considerably weakened by a lack of funding and would have to be built up again in the future. It was also mentioned that the number of people the Committee and the Department could turn to for expert advice would become fewer as research interests moved away from air pollution. The Chairman undertook to write a letter on behalf of the Committee to Sir John Pattison expressing these concerns. [ACTION: The Chairman] 39. It was noted that a letter had been published by The Times recently stressing the lack of work on the effects of indoor air pollutants on health. It was agreed that this should be circulated to the Committee. [ACTION: Secretariat]
ITEM 12. DATES OF FUTURE MEETINGS 40. It was agreed that the dates for the first two meetings in 2001 should be changed. The proposed dates are: Friday 9 February 2001 Friday 22 June 2001 The November 2 date remains unchanged. Members are asked to note these dates and to let the Secretariat know as soon as possible if there are difficulties. ACTION SHEET Para Reference Action Responsibility
APPENDIX 1 Changes to the document "Report on the Health Effects of Aldehydes in Ambient Air" are as follows: 1. A reference to the AEA technology report as the data source for Table 1 on page 7 has been included. 2. Misprints have been amended and units have been checked. 3. A reference to a review on the toxicology of glutaraldehyde (1992) has been included within the initial summary of the report (see page 1). 4. Section 2.8 para 1 (role of cytotoxicity/genotoxocity in the carcinogenicity of formaldehyde) has been amended to say "….suggesting that cytotoxicity, not genotoxicity, may be the rate-limiting step in this process". 5. A comment on GSH levels in the nose/lower airways has been included within the initial summary of the report. 6. The conclusions from the COM statement on the mutagenicity of acetaldehyde (1995) and a reference to the minutes of the most recent meeting of the COM re this issue has been included in Section 4.7. |
| Information | Help | Feedback | Search |