|
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||
|
 |
|
||||||||
|
FOR MEMBERS' USE ONLY COMEAP/2003/MIN2 COMMITTEE ON THE MEDICAL EFFECTS OF AIR POLLUTANTS Minutes of the meeting held on Friday 20th June 2003 in Room 125A, Department of Health, Skipton House, 80 London Road, London SE1 6LH. Present:
ITEM 1. OPENING REMARKS, APOLOGIES FOR ABSENCE AND ANNOUNCEMENTS Apologies for absence were received from Dr John Pritchard, Professor Tony Frew, Professor William MacNee, Professor Dick Derwent, Mr Fintan Hurley, Professor Steve Holgate, Dr Martin Meadows (DEFRA), Dr Martin Williams (DEFRA), Mr Nigel McMahon (Northern Irish Office), Mr David Russell (Welsh Office). The Chairman welcomed the following new members to COMEAP: The Chairman informed the Committee that there had been no interest for either the cardiovascular or epidemiologist positions. In their absence, the Chairman informed Members that Professor Dick Derwent had also been appointed to the Committee as an atmospheric chemist and that Professor Stephen Holgate had been appointed as an ex-officio member. The Chairman thanked, on behalf of the whole Committee, those members of COMEAP who were retiring for all their hard work during their time on the Committee. He acknowledged that by losing Professor Ross Anderson, Professor Roy Richards, Professor Anthony Seaton and Professor Roy Harrison, the Committee was losing not only a wealth of expertise but key players in the field of air pollution. The Chairman also welcomed to the meeting Dr Janet Dixon, who was attending as a representative of DEFRA. The Chairman reminded members to declare any interests they may have
at the beginning of each item. ITEM 2. MINUTES OF THE MEETING HELD ON 21st FEBRUARY 2003 Changes to the minutes were requested in paragraphs: Attendees: change Roy Harrison to Paul Harrison, Subject to these changes, the minutes were considered an accurate representation
of the meeting. Update on Cardiovascular sub-group Professor Poole-Wilson explained to the Committee that the purpose of the report going to consultation would be to provide an element of ownership of the report by the cardiovascular community. 5 Year Review Changes to Remuneration ITEM 4: NEW MEMBER APPOINTMENTS
This was the last of the detailed papers on ozone. The next step would
be to draw the conclusions together. It was felt that the detailed work
had been valuable and needed to be published in report form. It was also
noted that St George's had a contract with DH to produce and publish in
the literature quantitative reviews on several pollutants including ozone.
It was proposed that the COMEAP report should include a layperson's summary.
It was suggested that too many 'hedge' words should be avoided to ensure
clear conclusions. The Committee was informed that the process of publication of the APHEA results on ozone could take as much as a year. The Committee found the qualitative summary (indicating numbers of significant and non-significant studies) helpful but also wished to see information on the direction of the associations. The Secretariat clarified that studies described as "statistically significant" were statistically significant positive associations. Studies described as "non-significant" were not statistically significant but could be either positive or negative. The numerical coefficients for each study were given in the annex to the paper. There were almost no statistically significant negative associations. Regarding the conclusions of the paper summarised in the addendum to
COMEAP/2003/9a, it was felt that it should be made clear that they related
to daily fluctuations in symptoms and daily fluctuations in ozone. It
was also suggested that the conclusions be immediately preceded by a reminder
of the fact that mean levels were rising and the peak levels falling.
The Committee did not agree with the proposed conclusion 2(I) in the addendum paper whilst only a qualitative analysis was available. It was pointed out from Table D that, although only 2 out of 7 studies were statistically significant, if all of the 'non-significant' studies were positive, this could be quite strong evidence. It was possible that a quantitative summary estimate would be statistically significant overall. The Committee did agree that daily fluctuations in ozone appeared to be associated with daily changes in lower respiratory symptoms and breathing problems in asthmatics, and, possibly, in non-asthmatics undergoing vigorous exercise. It was also agreed that ozone appeared to be associated with cough only in asthmatics at higher concentrations. Quantitative analysis The difficulties of performing a quantitative meta-analysis when studies
varied by averaging time, type of symptom, subject group etc was discussed.
It was suggested that forest plots with adults and children estimates
adjacent and with the different averaging times labelled might give more
quantitative information without actually combining the studies. There
was a method for pooling p-values, not currently commonly in use, which
would provide a means of combining the strength of evidence across different
averaging times without having to deal with the expected numerical difference
in the coefficients. On the other hand, it might be possible to combine
certain categories and so do a quantitative meta-analysis for certain
groups of studies. It was proposed that the Secretariat should take this
further, in discussion with Professor Strachan, Professor Anderson and
Mr Armstrong. Thresholds The Committee concluded that an association with lower respiratory symptoms in non-asthmatics was only found above 120 ppb 1 hour average, with the possible exception of those undergoing vigorous exercise. The Committee also concluded that it was prudent to assume that associations with lower respiratory symptoms might extend below 40 ppb 8 hour average in asthmatics. However, the Committee considered that it was not possible to conclude that this was particularly in asthmatics not on anti-inflammatory medication as there were no studies which had examined the same subjects on and off anti-inflammatory medication. In coming to this conclusion, the Committee did not place much weight on the study by Higgins et al (1995) as the statistical analysis had not allowed for serial correlation and a little used analytical method was used to measure ozone. It was considered that the data of Thurston et al (1997) looked as if the dose-response function could be curved even though a linear relationship had been fitted by the authors. This emphasised the fact that a given set of data could be consistent with either a linear or curved relationship. Even if, on balance, the evidence suggested a linear relationship, it was difficult to exclude the possibility of a threshold. The presence of anti-oxidants in the upper respiratory system and lung lining fluid would be expected to provide some protection. However, it was also noted that some people lack uric acid in nasal lavage. It was also speculated that uric acid might need to be in an active form (i.e. soluble rather than crystalline). Personal exposure There was discussion of how much of an asset good personal exposure data would be in assessing the effects of ozone. Unfortunately, it was felt that the measurement of personal exposure by Delfino et al (1996, 1997) had a problem with face velocity at the sampler and lacked sensitivity. Coherence The issue of coherence of the evidence on ozone and respiratory symptoms
with the evidence on other endpoints was deferred to the next meeting. The Committee discussed paper COMEAP/2003/9C which summarised the range
across which ozone levels have been rising in urban areas of the UK. The
25th, 50th and 75th percentiles were rising but it was pointed out that
peaks were falling. It was suggested that 95th percentile data should
be added to the graphs to show this. It was also suggested that graphs
for selected individual long-standing sites should also be shown. This
was to deal with the point that the pattern in the graphs for ozone levels
from pooled sites could be influenced by the increase in the overall number
of sites over time, particularly if these were suburban monitors. There was also discussion of why rural levels did not seem to be going
up despite the evidence that there was a rise in global background levels.
It was pointed out that this increase was small and had been shown at
Mace Head during times when the wind was coming from the Atlantic. ITEM 6. DOES AIR POLLUTION CAUSE ASTHMA? DRAFT REPORT The Chairman introduced this item, reminding members (for the benefit of new members) why the report was being produced. The Chairman asked members to provide general comments on the report before commenting in detail on Chapters 2, 3, 5, 6 and 7. It was explained that the introduction would be tidied and shortened once the main body of the report had been completed. General Comments It was mentioned that the report needed to address a particular audience and it was suggested that in order to make the report useful to a varied audience, a summary for the public and an executive summary (as well as the report itself) should be produced. It was also noticed that the report did not mention indoor air, which would be of interest to members of the public searching for information on asthma. Members agreed that, as with outdoor air, there was no evidence which could suggest that indoor air caused asthma. However, it was agreed that this aspect should be taken into consideration. Members advised that the wording throughout the report should be checked to ensure the use of consistent terms, in particular where the words 'induction', 'initiation' and other words pertaining to the 'cause' of asthma were used. Chapter 1 It was suggested that the paragraph relating to house dust mites leading
to development of asthma on page 5 should be removed. Chapter 2
Pie Charts
Graphs
Other
Whilst discussing the need for air pollution data in Chapter 2, it was
noted that the summary air pollution data on the National Air Quality
Information Archive had become more difficult to access since the website
had been redesigned. The Chairman agreed to write to DEFRA drawing attention
to this point. Chapter 3 The 'diet' and 'hygiene' hypotheses were mentioned for inclusion. The Chairman informed the Committee that this chapter would contain references
which would direct the reader to further information on asthma.
Chapter 4 Chapter 5 Within this chapter it was agreed that the Cytokine information needed to be more specific (explaining which Cytokines are associated with allergic reactions). It was felt that the chapter lacked a certain feeling for which of the inflammatory pathways are important for asthma and how these are different from, for example, inflammatory pathways responding to viral infections. Then the inflammatory pathways which are important for the effects of pollutants, could be described and it could then be ascertained whether these overlap with any of the inflammatory pathways described earlier. One member also suggested that the chapter could outline the type of evidence that would be expected to be seen if air pollution was causing asthma and then describe whether any of this evidence existed. Members were divided on whether there was a need for a discussion of the toxicological evidence. Some members felt that toxicological evidence was only necessary when no human data was available (not true in this case) or in explaining the reason for an effect found in humans. Some members felt that the human evidence strongly suggested that air pollution did not cause asthma, so there was no effect to explain and no need for discussion of the toxicology. It was decided that if the chapter was to: 1) provide a general mechanistic background on what happens in asthma and 2) hypothesise how pollution could initiate asthma, it also needed to consider whether these plausible hypothetical mechanisms for an effect of pollution on causation of asthma actually operate. Consideration of toxicological evidence in animals was needed to address this last point. Members also suggested the counter argument should also be addressed i.e. discussing evidence suggesting that air pollution was not a cause e.g. sometimes large doses of particles have no long lasting effect in animal studies. It was noted that air pollution might even be protective if increased mucus production in the airways could reduce access of asthmagens to the epithelium. Members agreed that this chapter also required some information on dose at which effects occur. It was noted that the epidemiology section had deliberately omitted evidence on exacerbation of asthma by air pollution. However, it was noted that this chapter did include comment on exacerbation of asthma symptoms. Members agreed that it was important to acknowledge that Th1 diseases have increased as well as Th2 diseases such as asthma i.e. a switch from Th1 to Th2 cannot explain the significant increase in asthma. One member suggested that the position could be summarised as 'very little was known about how asthma was caused and even less about how air pollution might interact with this'. It was agreed that paragraph 7 needed updating with more recent references,
although members were informed that this would not alter the content of
the evidence presently provided. Professor Donaldson agreed to provide
some references and it was agreed that the Secretariat would conduct a
literature search. Upon request, Professor Donaldson and Dr Stone agreed to help with the
development of the next draft. Chapter 6 It was noted that there was a need to refer within the text and in table
1 to what the averaging times of the pollution levels were.
It was agreed that the Secretariat and Chairman would discuss this further
and inform the Asthma subgroup. It was mentioned that there was a slight overlap between the small area
studies and the traffic studies which needed to be addressed. Professor Anderson highlighted that he needed to check with Francesco
Forestiere that his unpublished data could be included in this chapter. It was noted that some of the tables and figures were not positioned
as closely to the text which referred to them as members would like. The
Secretariat agreed to amend these accordingly. Chapter 7 It was suggested that trend data be discussed under points 1 and 2, which referred to the prevalence of asthma and the pervasiveness of air pollution. Members were in agreement that the balance of evidence from the general incidence and prevalence studies was strongly against asthma being related to air pollution (point 6). However, at this stage members did not think that the balance of evidence was quite so strongly negative for there being no link between asthma and traffic exposure (point 7). Confounding factors related to living near roads had now been taken into account in traffic studies. However, it was impressed upon members that the effect might not relate to pollutants which were being directly measured but may relate instead to other unmeasured pollutants. Members noted that, although there was little evidence for an association of air pollution and asthma, this did not mean that air pollution was not harmful to respiratory health as there was evidence of air pollution leading to an increase in irritative respiratory symptoms such as cough and phlegm production. It was agreed that this would be discussed further within the report. It was confirmed that cross-sectional evidence showed no relationship to air pollution one way or the other, therefore negating the possibility that air pollution could be considered protective.
The Committee were informed that the Secretariat had received a request for this paper by Bernard et al (2003), which had received much media attention, to be considered by COMEAP. In general, members thought that the study had been well conducted. Members agreed that the results did warrant further research. Members discussed the study in greater detail and agreed that the controls used could have been tighter. One member raised the point that the production of extra surfactant within the airways could be regarded as a protective effect, which may naturally occur during exercise. This could lead to a rise in lung markers in the blood. This could have been tested, for example in the study of acute effects, by including a control group exercising without swimming pool exposure. It was, however, noted that there was an increase in lung markers in the blood in a group resting beside the pool. A control group could have been included involving exposure to a different swimming pool disinfection mechanism. This would also have helped to pinpoint whether it was the chlorination by-products that were harmful. Members remarked at how quickly the level of surfactant protein B in the blood rose and questioned whether the increase could be a result of active secretion rather than permeability changes. On questioning, members were unsure how much difference roof height would make to concentrations of nitrogen trichloride very close to the surface of the water. Members noted that the authors only included confounders in the model if they were statistically significantly associated with the outcome. There was no control for age and sex in the study for this reason. However, it was agreed that it may not be true that there was no confounding even if the association was not significant. Some members also felt that the population attributable risk should be determined, although it was thought unlikely that swimming pools were a major risk factor for development of asthma. As members were not convinced that surfactant proteins were produced
as a result of increased permeability, it was suggested that a letter
should be sent to the journal Occupational and Environmental Medicine
(OEM) and it was suggested that Professor Walters and Professor Richards
could lead on this. Members also queried the ecological analysis. Members queried the number of datapoints which, according to the research carried out, should only have had 15 datapoints (the number of schools) as opposed to 1881 (the number of children). Members queried the p values in Figure 5 which members felt suggested that the authors had not taken account of the clustering of children by school. There was enough data in Table 3 to recalculate the p values appropriately and it was agreed that these needed to be checked to ascertain whether the results lost statistical significance as a result of doing this recalculation. It was suggested that Mr Armstrong and Professor Strachan could write
a letter to OEM on the problems with the statistical analysis Members provided the Secretariat with suggestions for further research and factors which they thought had been overlooked by the study. Members thought that actual measurements of nitrogen trichloride levels would be useful, as would have been a study with a control group swimming at a pool sterilised by another method. The Secretariat would prepare a statement. The Secretariat apologised for the delay in discussing this draft document. The Secretariat provided a short background to the paper and informed the Committee of the problems with the terms guidance, standards and guidelines. The Secretariat informed Members that the 'guidance' (qualitative as well as quantitative advice) had therefore been written, containing numerical guidelines where appropriate for some pollutants. Members were requested to provide comments in writing to the Secretariat. Members were also asked by the Secretariat, whether they would be willing to publish the report as a piece of advice from COMEAP. Members agreed. Members were also informed that the document was specific to the domestic indoor environment only and therefore this document did not include any guidance for places of work (including schools) as this would fall under HSE. The general consensus of the Committee was that this document was a good piece of work. The following comments were made: The section on particles needed some rewriting. Members were keen that the report be promoted beyond the COMEAP website
and that a pamphlet or leaflet should accompany the guidance. The Committee
also agreed that a statement should be drafted. It was agreed that members' comments would be incorporated and that the
Guidance document would be sent to members again for final agreement.
This paper was provided for information to COMEAP members and was not discussed at the meeting. However, the Secretariat wished to draw the Committee's attention to the two attached reports on Nitrogen Dioxide: 1) WHO working group report (2003), Health Aspects of Air Pollution with Particulate Matter, Ozone and Nitrogen Dioxide; 2) IOM (2003) Quantification of the Acute and Long Term Effects of Exposure to Nitrogen Dioxide: A Literature Review - Draft Report (summary only), as Nitrogen Dioxide would be discussed at COMEAP at a later date. ITEM 10. HORIZON SCANNING - COMEAP/2003/14 The Secretariat discussed the main points on the list, noting that the list was longer than could practically be achieved within a short timescale. It was acknowledged that members would be producing a statement on long
term exposure to NO2 for June 2004 as requested by DEFRA. It was acknowledged that the second QUARK report was becoming a high
priority and would be including the wider NO2 data. The Secretariat confirmed
that Mr Hurley had agreed to chair this sub-group. It was agreed that work should be started in 2004 on air pollution and
children. This would involve setting up a sub-group with Professor Walters
chairing. Members were informed that the update on the Air Pollution handbook,
although a lesser priority, would be completed by the Secretariat. Some members had raised the issue of secondary particles. Although members were not convinced that this warranted a special report, this subject would be kept in mind. Metals in particles were also mentioned. Specific problems relating to carbon monoxide, such as long term exposure
to low levels, would be considered. This did not need to be a long report.
It was noted that a member of the Committee had some data on this. One member suggested that there could in the future be a need for setting a standard on particle numbers (although there were too few epidemiological studies for this to be done at present). It was noted that additional members would need to be recruited from
outside of COMEAP for the various sub-groups. The Secretariat requested that members send in any further ideas they
have.
Members were informed that the date of the next meeting was: 19th September 2003, Skipton House. COMEAP Secretariat
|
|
|
|
||||||
You
are here: 
