Website of the Department of Health

Introduction

1. Volcanic activity on Montserrat has led to the people being exposed to volcanic ash. The ash has a significant, though variable, content of cristobalite: a toxicologically active polymorph of silica. The particles of ash are small in size and a significant proportion are small enough to be inhaled deep into the lung, ie they are respirable particles. It is known that exposure to active forms of silica, including cristobalite, can cause lung disease. Studies of occupational exposures, where the concentrations of silica to which people may be exposed are sometimes high, are available and provide a background against which the possible effects of exposure to ash can be set. However, the composition of the ash is not identical with that of dusts encountered in occupational settings, for example in coal mines and quarries. Because of this it is difficult to use the mathematical relationships of exposure and risk of disease that have been derived from occupational studies to predict the risks that may be associated with exposure to ash on Montserrat.

2. It was realised early in work on the possible effects on health of exposure to ash on Montserrat that new studies, epidemiological and toxicological, would be needed. Such studies would allow a better comparison of the possible effects on health and of the toxicity of the ash with those of dusts encountered in occupational settings. This increased knowledge would then allow more appropriate use of relationships between exposure and risks of adverse effects to predict the possible effects on health of exposure to ash on the island. It was appreciated that an increased confidence in these predictions would be of value in developing guidance for limiting exposure to ash on Montserrat. This guidance will be valuable in developing procedures for protecting those at risk of exposure to ash during ash-clearing operations and also in developing policies regarding re-occupation and, if necessary, temporary evacuation of parts of the island that have been subjected to significant ash falls. The sustainability of life on Montserrat depends, in part, on occupation of a significant part of the island being recognised as being associated with only a small and acceptable level of risk to health as a result of exposure to volcanic ash.

The Studies

3. Four studies have been conducted. These are:

(i) a questionnaire study designed to investigate the respiratory health of Montserratians who chose to relocate to the UK during the period of regular volcanic activity on the island;

(ii) a study of individuals on the island that used the same questionnaire as used in the UK study and added chest X-rays and lung function tests (spirometry);

(iii) a study in which rats were exposed to ash by inhalation;

(iv) a study in which rats were exposed to ash by instillation of ash directly into the airways.

4. The studies have been completed and the final reports provided by the research groups involved are available. These are listed in the bibliography at the end of this report. A short summary of the findings of each study is provided below.

The Study of Montserratians who Relocated from the Island to the UK

5. This study was undertaken by the Institute of Occupational Medicine (IOM) at Edinburgh and involved sending questionnaires to all Montserratians living in the UK and whose names and addresses were held by the Montserrat Community Support Trust (MCST). Responses were received from 465 individuals: around 26% of the eligible study population. This is a low response rate but was in fact only achieved by several mailings and after two initial pilot studies. The low response rate means that the findings should be interpreted with caution.

6. Particular emphasis was placed on questions that allowed an estimate of exposure to ash to be made. The questionnaire drew on a widely used Medical Research Council respiratory symptom questionnaire and on the International Union against Tuberculosis and Lung Diseases questionnaire on bronchial symptoms. Questions were adapted for self-administration and emphasis was placed on wording questions for ease of understanding. A full occupational history was sought for each adult respondent. Responses for children were provided by parents.

7. The findings of the study are summarised below. This summary is taken from the IOM report.

Responses were received from 465 individuals (39% men, 60% women, 1% did not given information on gender) from 255 households, a response rate of around 26%. The majority (91%) were non-smokers, with men and older subjects most likely to smoke. Of the 465 responders, 30% reported at least one respiratory symptom. Prevalence of symptoms ranged from 7% with asthma attacks, through 14% with asthma and 14% with chronic bronchitis to 18% with breathlessness. Symptoms were more prevalent in current smokers and among older subjects, with the exception of asthma attacks which were most common in children.

No association was found between risk of reporting symptoms and exposure due to place of residence on the island or due to domestic cleaning. Risks of reporting asthma, chronic bronchitis and asthma attacks were statistically significantly related to exposure from heavy ash clearing tasks. Exposure-related risks of reporting these symptoms were increased, but not statistically significant, for symptoms which first occurred on Montserrat. Total exposure to volcanic ash was available for 52% of the study group. In this subgroup there was evidence that the risk of reporting each of the four symptoms increased with increasing exposure to symptoms overall and, more strongly, for symptoms which first occurred on Montserrat.

8. The findings have been discussed with advisers to Sir Liam Donaldson (Chief Medical Officer, England). The findings:

(i) an association between symptoms and heavy exposure to ash during clean-up operations;

(ii) no association between symptoms and place of residence on the island or with domestic ash clearing;

(iii) the rather greater prevalence of symptoms than that found in the UK population generally need to be considered in the light of the possible sources of bias listed below.

9. (i) The limited response rate. It may be that those suffering from respiratory symptoms were more likely to respond to the questionnaire than others.

(ii) The fact that individuals, particularly women and children, suffering from respiratory ailments were actively encouraged to leave the island.

(iii) The possibility that those occupationally involved in dealing with the consequences of volcanic activity were more likely to remain on the island.

(iv) The observation that populations which move from one country, and particularly from one climate, to another often suffer from an increased prevalence of respiratory disease.

10. Discussion of the findings and the above possible sources of bias led to the following conclusions.

(i) A proportion of those involved in ash clearing work and thus likely to have been exposed to high concentrations of ash, appeared to suffer from prolonged respiratory symptoms indicative of the ash having had an irritative effect on their airways.

(ii) Those not involved in heavy ash clearing did not appear to have sustained significant ill-effects from their exposure to ash.

(iii) It was not possible to reach any firm conclusions regarding the likelihood of those exposed to ash developing the chronic lung disease silicosis. On comparative grounds, considering the likely level and duration of exposure to ash it was felt that those not involved in occupational ash clearing were not at significant risk of developing silicosis. The case of those who are so exposed was agreed to be more difficult and it was accepted that a questionnaire study would be able to shed only very limited light on this.

The On-Island Study of Montserratians

11. This study was also conducted by the Institute of Occupational Medicine. In all, 421 Montserratians participated who were selected as the most heavily exposed workers in occupations that were partly or mainly outdoors and involved in exposure to ash. Emphasis was, again, placed on estimating individual exposure to ash and detailed occupational histories were recorded.

12. The key findings are as follows:

(i) no association between any respiratory symptoms and residential, domestic or occupational clearing of ash was found;

(ii) no association between indices of lung function (FEV1 and FVC) and place of residence, domestic ash clearing or heavy ash clearing was found;

(iii) no association between indices of lung function and estimated occupational ash exposure was found;

(iv) subjects working as gardeners and road workers had on average, and after adjustment for other variables, lower lung function than other subjects;

(v) six radiographs showed abnormalities but these were insignificant and not related to silicosis in four of the persons. In two individuals, the lungs showed interstitial changes and they were referred for investigation at the Aberdeen Royal Infirmary, where they were diagnosed as suffering from sarcoidosis.

(vi) a comparison of symptom prevalence indicated more symptoms amongst Montserratians who had relocated to the UK than amongst those studied on the island.

13. These findings have been discussed with the staff at IOM and with expert advisers and the following conclusions reached:

(i) no evidence of adverse effects on health due to exposure to ash has been found amongst those exposed by reason of their place of residence or as result of domestic ash clearing;

(ii) though it is not possible to define an association between lung function and level of exposure to ash, those exposed occupationally appear to have a reduced level of lung function compared with other subjects. It was noted that a small number of people (occupationally exposed) showed low-level X-ray changes indicative of exposure to ash. It is impossible to be certain that these changes will not become more significant, though the absence of symptoms and the minor nature of the X-ray findings makes it likely that the risk of individuals developing silicosis is low. It certainly seems to be the case that despite up to five or so years of occupational exposure to ash no signs of established silicosis have yet appeared;

(iii) it was concluded that the risk of silicosis amongst those not occupationally exposed to ash was negligible. For those exposed occupationally the risk is probably small but cannot be regarded as negligible. This point will be returned to later.

The Toxicological Studies

14. Concern about the possible effects on health of exposure to ash has been increased by the finding that ash can contain up to 20% (by weight) of the form of crystalline silica described as cristobalite. Different forms of silica vary widely in their toxicity but cristobalite is accepted to be one of the more active forms and standards for occupational exposure to silica include special consideration of this species. The cristobalite content of ash varies: the proportion of the silica which is present as cristobalite increasing with the period for which it has been exposed to high temperatures. Thus, the cristobalite content of material comprising the dome of the volcano increases with time and ash generated as pyroclastic flows resulting from dome collapses contains a higher proportion of cristobalite than "younger" material expelled during explosive events.

15. A range of factors affect the toxicity of the ash generated by the volcano. The cristobalite content of the ash is clearly likely to be important but so is the distribution of particle size in the ash as this will control the likelihood of the ash particles being inhaled deeply into and being deposited, in the lungs. It is also known that samples of material including cristobalite and other forms of silica can vary in their toxicity depending on the exact physico-chemical nature of the surface of the particles. The adsorption of some materials, including aluminium and clays, is know to reduce the toxicological activity of the particles. It is also known that the toxicological activity of quartz (a common form of crystalline silica that is converted to cristobalite at high temperatures) is greatest when it is freshly broken and declines with age. It might, therefore, be expected that the toxicological activity of ash would vary depending on its history: the exact nature of the change is difficult to predict. Ageing could reduce toxicity but exposure to acid rain from the volcano's plume could remove materials from the surface of ash particles which could increase the toxicity of the ash. It will be appreciated that though a great deal is known of the toxicology of the various forms of crystalline silica, comparatively few studies have looked at volcanic ash. The picture is further complicated by different volcanoes producing ash with differing cristobalite contents. These considerations form a background to the toxicological studies of ash produced by the Montserratian volcano.

The Inhalation Study: Undertaken by the Institute of Occupational Medicine

16. Rats were exposed to ash, daily, for periods of two to eight weeks. All rats were killed after eight weeks. By this time about 20 mg of ash had been deposited in the lungs. A similar study was undertaken using a toxicologically inert dust: titanium dioxide of an appropriate particle size. An assessment of lung inflammation was made by washing out the lungs before they were processed for histological study. It was found that the average particle size of the ash used in the study was larger than that of the titanium dioxide: this means that a slightly smaller amount of ash than titanium dioxide will have been deposited in the animals lungs. The ash used in this study came from a dome collapse event and contained c 20% cristobalite.

17. It was found that animals exposed to ash exhibited a more marked inflammatory reaction than those exposed to titanium dioxide. It was also found that histological changes (concerning thickening of the tissue separating the air spaces, and the reaction of macrophages - the dust ingesting cells) were more marked in the case of ash exposure than in the case of exposure to titanium dioxide.

18. Comparison with data obtained from earlier experiments revealed that though the ash was toxicologically rather more active than titanium dioxide it was, on a lung burden basis, significantly less toxic than pure quartz and a little less toxic than coal mine dust which also contains silica. It was concluded that though the ash could not be regarded as toxicologically inert it was not of high toxicity.

The Instillation Study: Carried out at Cardiff University

19. Rats were exposed, by intra-tracheal instillation on a single occasion, to ash with a high (c 20%) cristobalite content. Three dose levels were used.(1, 2.5 and 5 mg) and animals were sacrificed at 6, 13, 25 and 49 weeks after dosing. Other groups of rats received identical doses of anorthite (a major feldspar component of Montserrat ash) or a pure preparation of cristobalite. The cristobalite was intended to act as a positive control as the titanium dioxide had acted as a negative control in the inhalation study described above.

20. As in the inhalation study lungs were washed out (lavaged) prior to processing for histological study and cell counts in the lavage fluid and biochemical indicators of inflammation, epithelial change and fibrosis, measured. In addition to looking at histological changes in the lungs a histological examination of bronchial lymph nodes was undertaken. The volume of lymph nodes was also recorded: an increase in size being an indicator of nodal inflammation.

21. The key findings of the study were:

(i) anorthite produced no inflammatory effects nor were any histological changes observed;

(ii) cristobalite produced a brisk inflammatory response and in the highest (5 mg) dose group, significant histological changes indicative of fibrosis. The inflammatory response produced by cristobalite was evident 6 weeks after dosing;

(iii) ash produced a much less marked response than cristobalite though a greater response than anorthite. The authors stressed changes in lymph nodes that in the highest (5 mg) dose group were detectable at 13 weeks post-dosing and in the intermediate dose group (2.5 mg) at 25 weeks post-dosing. It was noted that no significant changes occurred at any time point in the lymph nodes of rats exposed to 1 mg of ash. Some small changes occurred in nodes of rats treated with 5 mg of anorthite, none occurred at lower doses, but those failed to follow the same pattern of increasing with time seen in the higher dose ash groups and in the majority of the cristobalite groups;

(iv) as regards indices of an inflammatory response in the lung the ash produced only minor changes with the exception of a significant effect in the highest (5 mg) dose group at 49 weeks post exposure;

(v) no evidence of fibrosis was found in the lungs of rats exposed to ash.

22. From this study it was concluded that the Montserratian ash was significantly less toxic than cristobalite though more toxic than anorthite. It was noted that cristobalite comprised about 20% (by mass) of the ash and so an interesting comparison could be made between the effects of 1 mg of cristobalite and 5 mg of ash: these doses both providing about the same amount of cristobalite. It was noted that, with regard to an inflammatory response, 1 mg of cristobalite produced an earlier response than 5 mg of ash though by 49 weeks the response to 1 mg of cristobalite had declined and the response to 5 mg of ash had become significant.

23. The appearance of a late inflammatory response to the ash was unexpected and is not easy to interpret. It may be suggested that in its initial state the cristobalite in the ash is in some way rendered inactive but that its activity becomes manifest after a long period in the lung. This could be due to the removal of some blocking or occluding material from the surface of cristobalite during its period in the lung. It is stressed that such interpretations are purely speculative.

Drawing Together the Findings from the Studies

24. Four studies have been completed. The epidemiological studies provide no evidence for a significant effect on health as a result of exposure to ash by reason of location of residence on the island nor as a result of undertaking domestic ash-clearing work. Occupational exposure, however, does seem to be associated with some effects though the data are not entirely consistent. Amongst those individuals in the UK responding to the questionnaire there was an association between respiratory symptoms and a history of heavy exposure to ash: this was not clear in the on-island study, Average lung function appeared to be reduced amongst those involved in dusty work but a clear association between indices of lung function and level of exposure to ash could not be demonstrated in the on-island study. Minor X-ray changes found in four individuals were not associated with ash exposure.

25. From these findings it is relatively straightforward to conclude that such exposure to ash as has occurred to date has not produced a significant effect on health. It will be appreciated that it is much more difficult to say whether:

(i) those heavily exposed in the past will in time, if this exposure continues, develop lung disease;

(ii) those only lightly exposed will in time, if such exposure continues, develop lung disease.

26. Current opinion holds that the risk of lung disease appearing in category (i) above is too large to be accepted. Steps should thus be taken to reduce exposure in those likely to be exposed to high concentrations of dust. It may be argued that this group is already showing minor effects of exposure and thus that further exposure should be reduced in accordance with good occupational health practice. It is clear that such good practice should be applied to all likely to be exposed to high concentrations of ash irrespective of their past history of exposure. Such action is in accordance with application of the Precautionary Principle.

27. In considering category (ii) the toxicological evidence needs to be taken into account and an attempt made to predict future effects. It is clear that the ash cannot be regarded as a toxicologically inert dust and thus does not fall into the category of dusts described as nuisance dusts. Equally it is clear that the ash lacks the marked toxicity of active forms of silica such as quartz and cristobalite. It is less clear, though possible, that the toxicity of the ash fails to reflect its cristobalite content, ie, the ash is less toxic than would be expected given that at least some samples contain 20% (by mass) cristobalite.

28. In an effort to predict the long-term effects of ash attempts have been made to draw parallels with exposure to other mixed dusts that have been studied in occupational settings. Coal-mine dust provides such a comparison though it too can vary significantly both in its quartz content and in its biological activity. This variation has led to a family of curves representing associations between exposure and risk of those exposed developing silicosis. Comparatively toxic coal-mine dust (with a high quartz content) produces steep exposure-effect curves, whereas comparatively inactive coal-mine dust produces rather flat exposure-effect curves. The toxicological studies described above suggest that Montserratian ash might best be represented by the less toxic examples of coal-mine dust.

29. A comparison might also be made with the exposure-effect curve derived from studies of workers exposed to diatomaceous earth. Naturally occurring diatomaceous earth contains about 2% quartz. Calcination of the raw material at high temperatures increases the cristobalite content which rises to 10-60%. The comparison may not be exact - especially if the cristobalite in the Montserratian ash is present in a toxicologically rather inactive form.

Why it is Important to Try to Predict the Likelihood of Lung Disease Developing in the General Population on Montserrat

30. If it can be established that the general population of Montserrat is at only a very low or negligible risk of developing lung disease as a result of domestic and casual exposure to ash, policy development regarding the sustainability of current arrangements on the island and plans for the future will be greatly aided. If, on the other hand, the long-term risks are not negligible and are judged to be unacceptably high then plans to reduce exposure by evacuation of some currently occupied areas during periods of volcanic activity and major expenditure on road cleaning may well be needed. It is stressed that current expert opinion suggests that the general population is at very low risk of developing lung disease as a result of domestic and casual exposure to ash. However, it is not currently possible to estimate this risk in quantitative terms. That this should be done is urged.

Quantitative Assessment of the Risk of Lung Disease in the General Population of Montserrat as a Result of Domestic and Casual Exposure to Ash

31. This risk assessment involve the combination of two separate sub-assessments:

(i) assessment of the likelihood of lung disease following prolonged exposure to a range of ash concentrations. The toxicological studies suggest comparison with the studies of coal-mine dust and of exposure to diatomaceous earth. This comparison will lead to a prediction of risk which will vary with exposure level;

(ii) assessment of the likelihood of exposures to the range of concentrations considered in (i) above, actually occurring. This can only be based on prediction of volcanic activity and on studies undertaken by IOM to determine exposure to ash on the island.

Combination of these two sub-assessments will yield a probabilistic and quantitative estimate of the risk of lung disease as a result of domestic and casual exposure to ash on the island over the next 20 or so years.

32. In undertaking this risk assessment one very important group of the population should not be forgotten: the children. Nothing is known of the risks of lung disease that may occur as a result of exposure to ash in early life. In general toxicological practice an uncertainty factor is often included in work designed to set standards that will protect children as well as adults. It is known that children spend more time outdoors than adults, play in dusty areas and take in more air per unit body weight than adults. Also, the lung continues to develop in complexity, in addition to simply growing, for up to 12 years after birth. A recent study has shown that there is a high prevalence of asthma amongst children on the island. There is currently no evidence to suggest that exposure to ash plays a part in causing children to develop asthma but the acidic and irritative nature of the ash may play a part in worsening the symptoms of those already suffering from the disease. The amount of effort that will need to be put into developing an objective means of allowing for the likely increased sensitivity of children will depend in part on the outcome of the risk assessment for adults. If the risk to adults is predicted to be negligible then the risk for children is also likely to be very small: the problem will be difficult if the risk for adults is judged to be small but not negligible.

33. Peter Baxter is currently discussing these approaches to a risk assessment with a staff at the Institute of Occupational Medicine and with leading volcanologists.

Conclusions

34. Epidemiological and toxicological studies have been undertaken to discover the effects on health of exposure to ash on Montserrat. The following conclusions have been reached.

(i) Those occupationally exposed to high concentrations of ash are at some small but not negligible and not acceptable risk of developing lung disease if their exposure continues. Measures should be taken to ensure that exposure is reduced and that regular medical examinations, including chest X-radiography, the next survey should be done 3 years after the first - thereafter surveys every 5 years should be adequate, depending on the risk assessment. The purpose of these examinations is to detect early signs of disease so that subjects can be advised to transfer to occupations associated with a lower level of exposure to dust. It is understood that the number of people, all men, occupationally exposed to high levels of ash is comparatively small.

(ii) The general population who are exposed to ash domestically and casually (ie, not as a result of their occupation being ash-clearing, road work or gardening) are probably not at any significant risk of developing lung disease if their exposure continues at levels that have occurred in the past.

(iii) It is suggested that a formal risk assessment be undertaken to estimate the risk of lung disease due to exposure to ash in the general population, taking into account the best available predictions of future volcanic activity and of exposure to ash. This will allow a firm and transparent basis for policy development regarding the need for planned evacuation of some parts of the island and the need for extensive road cleaning work.

(iv) The special case of children exposed to ash should be considered carefully during the risk assessment described above.

Bibliography

Cardiff University. Final Report of Surveys of Ash Toxicity Montserrat. Cardiff: Cardiff University, 2002.

Cowie HA, Searl A, Ritchie PJ, Graham MK, Hutchison PA, Pilkington A. A Health Survey of Montserratians Relocated to the UK. Edinburgh: IOM, 2001.

Cowie HA, Graham MK, Searl A, Miller BG, Hutchison PA, Swales C, Dempsey S, Russell M. A Health Survey of Workers on the Island of Montserrat. IOM Research Report TM/02/02. Edinburgh: IOM, 2002.

Cullen RT, Jones AD, Miller BG, Tran CL, Davis JMG, Wilson M, Stone V, Donaldson K, Morgan A. Toxicity of Volcanic Ash from Montserrat. IOM Report on DfiD Contract: CNTR 99 8764A. Edinburgh: IOM, 2002.

(Updated and expanded 4 September 2002)