Buchanan and St Ambrose Independent Review: final report

Chapter Five: Health Issues

5.1 Specific health concerns known about at the time of the review being commissioned were possible links with attending the school and acquiring cancer, specifically bladder cancer and/or the acquisition and impact of elevated blood levels of arsenic, specifically blindness. The issue of blue water is touched on in this section and dealt with in more detail later in Chapter 6.

5.2 Emails, calls and face-to-face meetings undertaken as part of this review also raised health concerns and will be dealt with in this section.

Investigation of a possible link between attending the school and acquiring cancer, specifically bladder cancer

5.3 On 14 November 2018, NHS Lanarkshire were first notified by a local GP of a patient who was diagnosed with bladder cancer reporting that four of their colleagues had also got this cancer within the previous few years. The patient had noted problems with blue water and wondered if there may be a link.

5.4 The public health department investigated the issue of blue water and discovered from North Lanarkshire Council that the issue was due to corrosion in the pipework leading to a build-up of copper in the water. Scottish Water were aware of the situation. The public health team undertook a review of the potential health effects of copper and found "no carcinogenic properties or indeed any significant health effects".

5.5 Having established there was no link between blue water and the case of bladder cancer, the public health team notified the patient and sought cooperation for further investigation of the possible cluster.

5.6 This process of investigation is in line with standard practice which takes a stepped approach to investigating public concerns about possible health risks.^[34] The public health team began at stage 1a: A belief that ill health exists in the community and that this is linked to exposure to an environmental agent(s) and the potential source of exposure is identified (e.g. from a specific factory or installation).

5.7 In this case, the potential source of exposure was blue water and this was rejected by finding robust evidence that copper is not carcinogenic.

5.8 In the guidance, there follows a stage 1b. This is as in stage 1a except no specific environmental source is under suspicion. The public health team continued in their investigations with an open mind to assess whether there may be still be a link between these cases of cancer and the school where they worked. Consideration was given to involving North Lanarkshire occupational health services, but given the expertise available within the public health team, the work was undertaken in-house. They collaborated with Health Protection Scotland and a University of Glasgow epidemiologist to investigate further, undertaking a literature review, investigations of the health concerns in more detail with the patients themselves and a review of existing epidemiological data.

5.9 The literature review explored possible links between cancer and landfill sites, latency periods for bladder cancer, occupational sources of bladder cancer and more general papers on investigating cancer clusters.

5.10 The findings from this literature review can be summarised as:

• There is no clear link between landfill sites and bladder cancer. The most relevant paper was from 2002 by Jarup et al^[35] in the British Journal of Cancer which demonstrated no excess risk of cancers of the bladder in populations living within 2km of almost 10,000 UK landfill sites. The results were similar if analyses were restricted to landfill sites licensed to carry special (hazardous) waste.
• There is a long latency period for developing bladder cancer in studies which have examined this for high risk occupations. Even when the risk of bladder cancer is known to be high, studies provide mean and median periods of latency from exposure to disease of between 15 and 40 years.
• It is highly unusual for a reported cluster of cancer cases to be directly related to an environmental hazard.

5.11 To investigate the health concerns in more detail, a consultant member of the team contacted the original patient with bladder cancer to seek consent to access their medical records. The consultant also requested that the patient relayed a message to colleagues reported to have cancer to get in touch with public health. Three out of the four colleagues did so and consented to have their medical records accessed. They each completed a questionnaire and were subsequently interviewed by a Consultant in Public Health Medicine.

5.12 Of the four members of staff interviewed, three were confirmed to have bladder cancer. The fourth had a different kind of cancer. The cases of bladder cancer were diagnosed between 2015 and 2018. To protect medical confidentiality, it is not possible to go into more details of these interviews in this report. However, the investigation found nothing from the responses provided by the patients which would point to a common exposure to an environmental hazard on the site of the school as the cause of their cancer.

5.13 In June, after these interviews were completed, the fifth member of staff got in touch with the public health department and was interviewed in the same way as the previous four. This person did not have bladder cancer so the underlying conclusions remain unchanged. There were no other responses in the interview which pointed to an environmental hazard at the school causing the patient's cancer.

5.14 The third part of the investigation was a review of existing epidemiological data. This showed that cancer of the bladder is the ninth most common cancer in Scotland with 80-100 cases a year expected in a population the size of Lanarkshire. The strongest risk factors for this type of cancer are exposure to tobacco smoke and age. A small proportion of bladder cancers are associated with specific exposures within certain industries and occupations. Teaching is not considered one of these industries or occupations.

5.15 At this point in the investigation, the public health team set up a Problem Assessment Group meeting (17 April 2019) to discuss the findings with experts from Health Protection Scotland and Glasgow University. They concluded that the cases identified equated to "what could be deemed the norm in a cross section of the population of a similar demographic to the school teaching population". Having reached this conclusion on the basis of the evidence they reviewed and the systematic approach taken, they concluded no further investigation was warranted (i.e. moving to Stages 2, 3 or 4 in the guidance was not indicated). As such, the Health and Safety Executive was not required to be involved.

5.16 The Consultant in Public Health Medicine who interviewed the patients with cancer sent each of them a letter on 1^st May 2019 to explain these conclusions and invited them to contact the department of Public Health if they had any queries.

Reflections on Investigation into cancer clusters

5.17 The UK and Ireland Cancer Registries have published a useful factsheet^[36] on cancer clusters which was used as one of many sources of guidance by the investigating public health team. It states:

"A cancer cluster occurs when more cases (of the same type or similar types) of cancer than expected are diagnosed in a group of people, geographic area and/or period of time.

"When someone is diagnosed with cancer many people ask "what caused it?" This is especially true if several people they know are affected. Although great progress has been made over recent years into researching cancer risk factors and developing successful treatments, the question "what caused it?" remains very difficult to answer precisely.

"Cancer is a complex disease with many different causes, and the reasons why it affects some people and not others are still poorly understood. Health professionals and researchers do not want to dismiss people's concerns about cancer clusters but at the same time may be genuinely unable to provide the answers.

"Although most cancer clusters occur by chance, it is not uncommon for people to be concerned that cancer clusters are caused by exposure to a cancer-causing agent in the environment. But real clusters that are proven to be associated with an environmental or occupational carcinogen are extremely rare.

"Even if there are more people with one type of cancer in a community than might be expected, this does not necessarily mean that they were all caused by a cancer-causing agent in the environment."

5.18 The factsheet uses the example of a "sharpshooter" who fires bullets into the side of a barn and then draws the bullseye around the cluster of shots which look closest together. In the same way, we tend to notice cases first ("bullet holes" and then the fact they are located in the same place (a single school). It goes onto say:

"It is very important to ask questions such as: "If there is something affecting this street, what other neighbouring areas would it also affect?" and "What is the smallest population that should be studied?""

5.19 A further useful contribution this factsheet provides is considering how a random way of spreading dots on a page can end up generating what look like clusters. The underlying pattern is random but putting gridlines around the dots gives them the appearance of clusters. The factsheet notes:

"Even with a very rare disease, there is always a possibility that, just by coincidence, somewhere, sometime, several cases will arise in people who live near one another. This makes it very difficult to distinguish between clusters of diseases that have a common cause and clusters that are due to chance alone."

Conclusion of the Review about investigation into an apparent cancer cluster

5.20 We conclude that the GP was right to raise the concern when a cluster of cancer cases occurred at Buchanan High and for the public health team to investigate if this was linked to copper in the water supply or an unknown environmental exposure at the school. However, reviewing the evidence set out above, we see no causal link between these three cases of bladder cancer and attending the school. The remaining two members of staff had two different kinds of cancer and no plausible explanation could link all of these together with the school. Bladder cancer is not very rare. There is a real possibility that it can appear as a cluster due to chance alone and once plausible exposures were eliminated, it was reasonable to come to this conclusion.

Attending the school and the acquisition and impact of elevated blood levels of arsenic

5.21 On 8 March 2019, NHS Lanarkshire public health team received a call from a GP asking for advice about a patient who was being investigated for sight loss and had a single positive test for arsenic in his urine (the remit for this review says blood, but the result came from urine). The patient was a first year pupil at Buchanan High School and his mother was concerned that the finding of arsenic in his urine might be due to the school and the cause of her child's blindness.

5.22 On 30 April, the public health team received a second call about a pupil at St Ambrose High School who had been found to have a single positive test for arsenic in their urine.

5.23 For confidentiality reasons, it is not possible to describe the details around the investigations of these two individuals. The public health team took advice from the National Poisons Service and undertook a structured response to these concerns seeking views of senior clinicians and investigating possible sources of exposure. There is nothing to suggest the children tested positive for arsenic as a consequence of attending school or that these positive tests relate to clinical symptoms.

5.24 By way of further explanation, there are many sources of arsenic which can lead to a positive urine arsenic test: seafood, vegetables, rice, including rice milk and water from non-mains supplies. The National Poisons Service recommend repeating the test as dietary sources of arsenic may lead to elevated results. Furthermore, if chronic exposure to arsenic or any other heavy metals is being considered as a possible diagnosis, this should be investigated by a specialist and not undertaken by GPs as interpretation of results is difficult. Testing is recommended only if there are symptoms to suggest exposure or there is an identified source of exposure to inorganic arsenic. Sight loss is a rare complication of prolonged or repeated arsenic exposure over many years and could not explain these symptoms in a first year pupil who has had a few weeks of possible exposure to a possible source at a new school. Furthermore there is an alternative clinical explanation for the blindness.

5.25 Some of the worries expressed by parents, pupils and staff may have been generated through the fact that arsenic is carcinogenic and can cause bladder cancer. Hearing about arsenic and bladder cancer at the school in the same media report sounds alarming. Putting this alongside problems with the "blue water", might imply there is a link. However, arsenic was not detected at elevated levels in the water supply on the school campus. The supply comes from the mains and feeds all the local community around the school campus as well. Later in this report we detail results of soil sampling which show no significant levels of arsenic on the site.

Conclusion of the Review about the investigation into an apparent link between arsenic and blindness

5.26 We conclude that the parent concerned was right to raise a question about the source of arsenic and whether it was the cause of her son's blindness. We can also say, NHS Lanarkshire's investigation was thorough and conclusive. There is no causal link between arsenic and the sight loss or other health conditions reported to them in pupils at the two schools. Nor is there a causal link between arsenic exposure and the bladder cancer cases at Buchanan High school.

5.27 These conclusions are important because the public health department had no grounds to investigate further possible exposures in the wider population. There were no specific health conditions to investigate or environmental toxins to test for. They wrote to GPs with their conclusions and advised them not to test proactively but to continue with normal clinical practice: "If you do receive a request to undertake testing, please treat and investigate the patient as you would for any other patient based on their symptoms and signs."

5.28 A pathway flowchart was provided by the Public Health department for GPs to follow and clinicians were given relevant information and advice to assist clinical decision making. The flowchart enabled further assessment of patients based on clinical presentation. NHS Lanarkshire gave contact details to GPs to allow patients who required additional information to contact Public Health. In the majority of cases the pathway was sufficient to alleviate concerns.

Other health concerns raised through emails to the Review Team

5.29 Out of the emails received by the review team (as at 2 August), there were specific health concerns for 65 pupils and ex-pupils (less than 4% of the total school rolls). There were also concerns from six members of staff and ex-staff which cannot be reported on in detail for confidentiality reasons. The dominant profile for symptoms in pupils and ex-pupils were: headaches, fatigue, stomach cramps, nosebleeds and a small number of other complaints.

5.30 These concerns were very similar to those reported to Scottish Hazards (referenced in paragraph 3.19 above) in the on-line survey to parents that they undertook. Of 220 responses to a question in relation to the symptoms present, most parents responding reported multiple symptoms as outlined in the following table.

Summary of Symptoms from Scottish Hazards Survey

Symptoms	No of responses	percentage of responses
Headaches	192	89
Stomach Cramps	152	50
Nausea	130	48
Abnormal Fatigue	126	46
Joint Pain	81	39
Nose bleeds	57	32
Blurred vision	52	23
Hair loss	19	14

5.31 As with this survey, the emails received from parents to the review group were not drawn from the whole school population and some reported on symptoms in ex-pupils. The review team took the view that the symptoms needed assessing for a possible pattern which was compatible with exposure to an environmental hazard. The symptom profile from the emails received were anonymised and put into table form for NHS Lanarkshire's public health team and Health Protection Scotland to review. The public health team discussed these concerns with a group of experienced local paediatricians who considered the symptom profile to be "neither unusual nor excessive".

5.32 A more detailed analysis comparing this symptom profile with community-based surveys of secondary school aged children was carried out for the Review by Health Protection Scotland.^[37] The table below sets out a summary of their findings.

"Table 2: Ranking of symptoms reported in community based surveys compared to symptoms reported at Coatbridge Schools.

	Symptom	A. Community symptom surveys – symptom prevalence range	B. Community symptom surveys – average symptom ranking	C. Coatbridge Schools – symptom ranking	D. Difference in rankings between community surveys and school reports (B-C)
		Range of prevalence rates (%) quoted in studies.	Frequency of symptoms – average ranking.	Ranking	Rank in (B) less Rank in (C) (- means lower; + means higher).
1	Headaches	8.1 – 41.1	1	1	0
2	Fatigue	4.5 – 36.2	2	2	0
3	Joint Pain	3.9 – 23.5	=3	=9	-6
4	Skin Problems	2.8 – 26.0	=3	=8	-5
5	Sore Throats	2.5 – 35.8	4	=10	-6
6	Impaired Vision	(one survey only) 7.5	=5	6	-1
7	Stomach Cramps	2.5 – 11.6	=5	3	+2
8	Dizziness	0.9 – 6.5	6	7	-1
9	Nausea	0.8 – 17.8	7	5	+2
10	Diarrhoea	1.3 – 9.5	8	=9	-1
11	Vomiting	3.0 – 4.5	9	=8	+1
12	Hair Loss	(one survey only) 1.0	10	=10	0
13	Nosebleeds	(one survey only) 0.6	11	4	+7

5.33 Health Protection Scotland conclude:

"Based on the range of symptoms identified, their lack of specificity and the variation in the frequency of reporting in community based surveys, HPS is of the opinion that it is not possible to identify a sufficiently distinctive pattern of symptoms that could be considered consistent with exposure to any specific chemical.

"The symptoms reported are typical of those commonly reported in the general community and, with the exception of nosebleeds and stomach cramps, are consistent with commonly experienced symptoms in a general population. The symptoms reported do not therefore appear to constitute a distinct clinical syndrome as such."

5.34 Furthermore:

"Other explanations for the symptoms reported cannot be excluded, including that the type of symptoms and frequency are consistent with the variation in range of symptoms likely to be experienced in such an age group normally and that a number of the symptoms could be associated with sub-optimal indoor air quality at the school site rather than due to extrinsic chemical exposure."

5.35 The report also comments on the value of human testing as follows:

"On that basis there is insufficient evidence to justify further investigation of school site users to detect exposure to specific chemicals using biological testing. Testing of blood, urine or other biological samples, for evidence of exposure to one or more specific chemicals, is therefore not recommended by HPS."

Reflections on other health concerns

5.36 Interest is growing in the impact of sub-optimal indoor air quality in secondary schools. Current UK standards permit temperatures up to 32°C, but comfort is greater between 20°C and 25°C. A literature review from 2014 suggests cognitive performance is better at 20°C to 22°C. ^[38]

5.37 Furthermore there is evidence for nosebleeds to be more common in young people experiencing low levels of humidity.

5.38 Another possible impact on indoor air quality is the concentration of carbon dioxide. Good ventilation with outdoor air is needed to keep carbon dioxide levels to a comfortable level. Low ventilation rates and raised levels of carbon dioxide are common in schools.

5.39 The English Education & Skills Funding Agency sets out guidelines,^[39] which are also used as the industry standard in Scotland, on maximum CO₂ levels and minimum ventilation rates to ensure adequate indoor air quality in classrooms.

5.40 The recommended ventilation performance standards for naturally ventilated classrooms can be summarised as follows:

• Average indoor CO₂ levels during a typical teaching day shall not exceed 1500 ppm, and average ventilation rates shall be above 5 L/s-p (Litres per second per person).
• At any occupied time the occupants should be able to reduce the concentration of CO₂ to 1000 ppm, and ventilation rates above 8 L/s-p shall be easily achieved by the occupants.
• Minimum ventilation rates shall not fall below 3 L/s-p. A standard is set for CO₂ levels not to exceed 2000 ppm for more than 20 minutes at a time.

Sickness absence records – pupils

5.41 North Lanarkshire Council provided figures for school absence rates from both Buchanan and St Ambrose High Schools. Pupil absences are recorded twice a day to account for children who are sent home during the morning. Until the first week of June 2019, when the public meeting took place which alarmed a large number of parents, both schools had school attendance for pupils of 90% over the last three years. These levels of attendance are above the average for other high schools and other special needs schools in Lanarkshire (88% for both types of school). From the first week of June (noting that pupils who are leaving school after the summer term finish term early) attendances fell and by the third week were 45% for St Ambrose and 78% for Buchanan. This was a reflection of the level of concern and anxiety that was developing in the school community.

Sickness absence records – staff

5.42 Staff absences from 2015/16 showed no noticeable trend for St Ambrose staff. This averaged 42.3 hours per absence recorded which compares favourably with St Andrews High School in Coatbridge which is of a similar type and size and had an average rate of 47.3 hours per absence recorded. In Buchanan High School, the average hours per absence over this timeframe was 84.5 compared to 60.1 in Glencryan School in Cumbernauld, which is of a similar type and size. This higher level probably reflects the amount of long-term sickness absence in individual members of staff at the school known to have had cancer.

5.43 These findings around sickness absence provide further reassurance about the health concerns which have been raised because both Buchanan and St Ambrose High Schools share the same site.

Possible long-term health effects

5.44 Discussions with a number of people, including MSPs, raised further questions about long-term effects of drinking the blue water and whether there were other possible exposure risks from a school built on a former landfill site. These concerns led the review team to undertake further investigations, commissioning outside agencies and independent consultants to provide an up-to-date risk assessment around the safety of the site. The results of these further investigations are reported later. It is important to say there are no significant long-term health impacts from ingesting copper. It is not carcinogenic and there have not been any reported cases in the research literature of birth defects as a result of exposure.

Conclusions of the Review about further health concerns

5.45 On the basis of these investigations, we conclude additional case finding was not required by NHS Lanarkshire and agree with the advice given to GPs to continue to treat and investigate patients from the schools as they would for any other patient based on their symptoms and signs.

5.46 Further air testing, if taken forward (over and above the air testing which has been carried out in relation to soil testing and gas membrane assessment and is described in chapters 7 and 8), should focus on indoor air quality and require assessment of temperature, humidity and carbon dioxide levels when the school is in use by pupils and staff.