Information Services Division

Analysis of cancers associated with radiation in the area

around Dalgety Bay, 2008-2017 An update for the Committee on Medical Aspects of Radiation in the Environment (COMARE)

10 December 2019

An Official Statistics publication for Information Services Division

This is an Official Statistics Publication

The Official Statistics (Scotland) Order 2008 authorises NHS National Services Scotland (the legal name being the Common Services Agency for the Scottish Health Service) to produce official statistics. All official statistics should comply with the UK Statistics Authority’s Code of Practice which promotes the production and dissemination of official statistics that inform decision making. They can be formally assessed by the UK Statistics Authority’s regulatory arm for National Statistics status.

Find out more about the Code of Practice at: https://www.statisticsauthority.gov.uk/osr/code-of-practice/

Find out more about official statistics at: https://www.statisticsauthority.gov.uk/national-statistician/producers-of-official-statistics/

1

Information Services Division

Contents

Summary ...... 3 Introduction ...... 4 Methods ...... 5 Results and Commentary ...... 8 Discussion ...... 12 References ...... 13 Glossary ...... 14 Contact ...... 15 Further Information ...... 15 Rate this publication ...... 15 Appendices ...... 16 Appendix 1 – List of radiation-associated with cancers ...... 16 Appendix 2 – Publication Metadata ...... 17 Appendix 3 – Early access details ...... 19 Appendix 4 – ISD and Official Statistics ...... 20

2

Information Services Division

Summary

• Increased radioactivity levels have been found around Dalgety Bay, , since at least 1990. In 2011, more radioactivity was found in particular locations. It is the result of scrapping luminous aircraft instruments that contained Radium-226, after the Second World War.

• The Committee on Medical Aspects of Radiation in the Environment (COMARE) has led investigations of this radioactivity and ISD provided information on cancer occurrence. While there was no overall increase in cancer risk found between 1975 and 2009, COMARE recommended in 2014 that ISD should repeat its analysis of cancer incidence in the area in the future. This report comprises the update of our previous analysis of cancer incidence in the area around Dalgety Bay.

• No significant increase in overall risk of cancer in Dalgety Bay was found, compared with the rest of Scotland between 2008 and 2017. No significant increased risk of any single group of cancers was found, including those for which there is stronger research evidence that radiation is a causal factor.

• ISD will repeat its analysis of cancer incidence in and around the Dalgety Bay area in the future on the advice of COMARE.

3

Information Services Division

Introduction Radioactive particles have been found in the vicinity of the town of Dalgety Bay, Fife, since at least 1990.1 These have been shown to be radium-226 and were deposited in the area as a result of military activity during and shortly after World War II. At this time, an air base was present which was involved in the servicing and disposal of aircraft fitted with components luminised with radium paint.

COMARE (the Committee on Medical Aspects of Radiation in the Environment) has led investigations into radiation at Dalgety Bay since 1991. ISD has provided information on cancer occurrence to COMARE and the results have been publically available since at least 1994.1

In 2011, following the discovery of particles of significantly higher radioactivity than had previously been found in Dalgety Bay, restrictions were put in place, including fenced off areas and a ban on fishing. This also led to the 15th report of COMARE,2 which provided a comprehensive history of the site, the nature of the contamination, an assessment of the risk to human health and an analysis of the actual incidence of certain cancers in the area. Although COMARE’s conclusion was that, “the sources present at Dalgety Bay pose a potential risk to public health,” no evidence for a general increased risk of cancer was found in the examination of cancer incidence. The incidences of two specific types of cancer, primary liver and non-Hodgkin lymphoma (NHL), were found to be higher than would be expected between 2000 and 2009, but after a clinical review of liver cancer cases, COMARE concluded that, “It is thus very unlikely that the excess cases of liver cancer or of NHL during 2000–2009, or of [non-melanoma skin cancer] during 1975–2002, are due to the presence of radium-226 in the area.”2 Radiation monitoring and the disposal of contaminated material continue under the supervision of the Scottish Environmental Protection Agency (SEPA) and there is a long term remediation plan in place.3

Following a request from the Chair of COMARE in 2018, ISD has produced this publication. It updates the previous analyses of cancers in the Dalgety Bay area, using similar methodology used to describe cancers in the area between 2000 and 2009 for COMARE. This report covers the period 2008 to 2017 (the latest date for which cancer registrations are published).

4

Information Services Division

Methods Cancer is not uncommon and about 2 in 5 people in Scotland will be diagnosed with it at some point in their lives.4 In order to understand whether the numbers of cancers in the Dalgety Bay area were significantly different to anywhere else in Scotland, we compared the incidence in Dalgety Bay to that expected in a population of the same size, in the same calendar years, with the same age, sex and socio-economic circumstances using indirect standardisation. We produced ratios of observed/expected (O/E) numbers, (also known as Standardised Incidence Ratios (SIRs)). We took into account the likelihood of chance findings by giving 99% confidence limits and also reported exact one-sided p-values. The methodology in this report is very similar to that used in previous analyses by ISD but our adjustment for socio-economic deprivation is more precise. Where previously results in Dalgety Bay were compared with those in the least deprived quintile of the Scottish population, this time the expected numbers were obtained from matching each socio- economic decile. We reported on cancers up to the end of December 2017, which was the latest complete year that had been published at the time of writing.

There is evidence for associations between some types of cancer and exposure to radiation. In 2006, Boice5 (Appendix 1) categorised these into four groups, from 'frequently' to 'never or sporadically' associated with ionising radiation. Boice's groups were used to structure the results of this report. Additionally, childhood leukaemias are also looked at in a fifth group due to the known association with ionising radiation.6 Records of incident cases of the relevant cancers in the period 2008-2017 were extracted from the Scottish Cancer Registry.

The geographical area examined in this report is the same as that used for previous investigations. Dalgety Bay has been defined as the 13 datazones between the and A921 on the south and north, by Letham Hill to the west and the fields above Braefoot Point to the east – Figure 1. Datazones are small area geographies used by Scottish Neighbourhood Statistics, and are based on groups of 2001 Census output areas, with populations of between 500 and 1,000 residents. (NOTE: where possible, the datazones are drawn to respect physical boundaries and natural communities. They have a regular shape and, as far as possible, contain households with similar social characteristics.)

5

Information Services Division

Figure 1: Area in and around Dalgety Bay. Area included in the analysis is highlighted in orange.

Dalgety Bay is a relatively affluent area in comparison to the rest of Scotland. All 13 datazones included in the analysis are in the least deprived 40% of the Scottish population, according to the Scottish Index of Multiple Deprivation 2012. Twelve of the 13 datazones are in the least deprived 20% of the Scottish population and eight of the 13 datazones are in the least deprived 10%. Incident cases of cancer in the population of this area during 2008-2017 were enumerated by identifying cancer registration records with postcodes which map to one of the 13 datazones described above.

Ratios of the observed numbers to expected numbers were then calculated for the various types of cancer in Boice’s groups, plus childhood leukaemias, to ensure continuity with the previous analyses. Person-years at risk were calculated using annual population estimates for the selected datazones up to 2014, based on the 2001 Census datazones. These have not been updated since 2014, and as such, population estimates for 2014 were used as substitutes for the populations in the 13 datazones for 2015 to 2017.

For each of the cancer types, exact confidence intervals for the observed to expected ratios were calculated using the Poisson distribution directly.7,8 In recognition of the large number of comparisons being made and the risk of false positive results, the confidence intervals were calculated at the 99% level.

6

Information Services Division

One-sided p-values are included where the number of observed cases is greater than expected. One-sided rather than 2-sided significance tests were performed because our priority was to identify increases, rather than decreases, in cancer risk, under the assumption that higher radiation levels would only confer a risk to cause cancer, rather than protect against it. These p-values are included, in addition to the confidence intervals, as a measure of the strength of evidence for the result. The p-value gives the probability of obtaining an observed/expected ratio which is greater than or equal to the reported value due to chance when the null hypothesis is actually true, i.e. the observed number is no different to the expected number. For example, a p-value of 0.20 suggests that there is a 20% probability that an apparently raised risk occurred by chance alone. To reject the null hypothesis and hence accept the alternative hypothesis (in this case, that the observed results truly are greater than expected) we used the 1% threshold to ensure chance was much less likely to explain the result were a significant result to be found.

Where either the number of observed cases or the number of expected cases is less than 5, we have suppressed the data (replacing it with *) to reduce the risk of potentially identifying individuals, in keeping with ISD’s disclosure control protocol.

7

Information Services Division

Results and Commentary The overall incidence of cancers in the Dalgety Bay area between 2008 and 2017 was similar to the rest of Scotland (SIR 0.97, 99% CI 0.87-1.09). The results are shown below broken down into four groups: cancers that are frequently associated with radiation and would therefore be most likely to show an increase if radiation exposure had occurred (Table1); those that are occasionally associated with radiation and have other risk factors (Table 2); those with an uncertain relationship with radiation where any increased cancer risk would be difficult to ascribe to radiation (Table 3); and those that have never, or only sporadically, been associated with radiation (Table 4). We also report childhood leukaemia incidence (Table 5).

None of the cancer sites showed evidence of a significantly higher than expected incidence rate in Dalgety Bay area.

A previous analysis (covering 2000-2009) found that liver and non-Hodgkin lymphoma rates were higher than expected in the Dalgety Bay area (O/E ratios of 2.47 and 1.71, respectively);2 although neither was raised before this between 1975 and 2002. In this analysis between 2008 and 2017, the O/E ratio for liver cancer was 1.19 (99% CI 0.41-2.64, p=0.35) and for non-Hodgkin lymphoma was 1.09 (99% CI 0.58-1.85, p=0.37). That is, neither was significantly higher than the rest of the Scottish population.

The overall O/E ratio for cancers that are frequently associated with radiation was 1.09 (99% CI 0.84-1.38, p = 0.20) – Table 1.

Table 1. Incidence of cancers frequently associated with radiation with authoritative risk estimates, Dalgety Bay, 2008-2017.

Cancer Observed Expected O/E 99% CL 99% CL 1-sided p Scotland (ICD-10) Cases (O) Cases (E) Lower 2 Upper 2 value 2 Cases 1 Myeloid 6 5.0 1.20 0.31 3.13 0.39 2,969 Leukaemia (C92)

Thyroid 7 4.9 1.44 0.42 3.52 0.22 2,534 (C73)

Female 100 94.2 1.06 0.81 1.37 0.29 45,777 breast (C50) CL – Confidence Limit 1 - Based on comparison with age, sex and deprivation matched population of Scotland 2 - Calculated using exact Poisson distribution where O/E≥1

8

Information Services Division

The overall O/E ratio for cancers that are occasionally associated with radiation and have other risk factors was 0.99 (99% CI 0.80-1.20) – Table 2.

Table 2. Incidence of cancers occasionally associated with radiation with robust risk estimates, Dalgety Bay, 2008-2017

Cancer Observed Expected O/E 99% CL 99% CL 1-sided p Scotland (ICD-10) Cases (O) Cases (E) 1 Lower 2 Upper 2 value 2 Cases Lung (C34) 65 55.0 1.18 0.84 1.62 0.10 52,278

Stomach 8 10.3 0.77 0.25 1.80 6,988 (C16)

Colon (C18) 41 48.9 0.84 0.54 1.24 26,672

Oesophagus 14 13.2 1.06 0.47 2.04 0.44 8,909 (C15)

Bladder 13 14.2 0.91 0.39 1.79 8,216 (C67)

Ovary (C56) 10 11.0 0.91 0.34 1.95 6,106

Brain and 6 8.3 0.73 0.19 1.90 4,366 nervous system (C71- C72)

Liver (C22) 9 7.6 1.19 0.41 2.64 0.35 5,335 CL – Confidence Limit 1 - Based on comparison with age, sex and deprivation matched population of Scotland 2 - Calculated using exact Poisson distribution where O/E≥1

The overall O/E ratio for cancers with an uncertain relationship with radiation, where any increased cancer risk would be difficult to ascribe to radiation, was 0.84 (99% CI 0.63-1.08) – Table 3.

9

Information Services Division

Table 3. Incidence of cancers rarely associated with radiation with uncertain risk estimates, Dalgety Bay, 2008-2017

Cancer Observed Expected O/E 99% CL 99% CL 1-sided p Scotland (ICD-10) Cases (O) Cases (E) 1 Lower 2 Upper 2 value 2 Cases Kidney 12 15.1 0.80 0.33 1.60 9,191 (C64)

Salivary * * 0.91 0.00 6.79 557 glands (C07-C08)

Non- 22 20.2 1.09 0.58 1.85 0.37 10,382 Hodgkin lymphoma (C82-C86)

Myeloma 9 9.1 0.98 0.34 2.19 4,386 (C90)

Malignant 22 31.4 0.70 0.38 1.19 12,323 Melanoma (C43)

NMSC 279 249.4 1.12 0.95 1.30 0.03 110,000 (C44)

Rectum 20 22.3 0.90 0.46 1.56 12,098 (C19-C20)

Uterus 10 14.0 0.71 0.26 1.53 7,638 (C54-C55)

Bone (C40- * * 0.98 0.00 7.31 493 C41)

Connective * * 0.33 0.00 2.44 1,417 tissues (C47, C49) CL – Confidence Limit 1 - Based on comparison with age, sex and deprivation matched population of Scotland 2 - Calculated using exact Poisson distribution where O/E≥1 * - values less than 5 suppressed in accordance with ISD’s Data Disclosure Protocol

The overall O/E ratio for cancers that have never, or only sporadically, been associated with radiation was 0.90 (99% CI 0.68-1.17) – Table 4.

10

Information Services Division

Table 4. Incidence of cancers never or sporadically associated with radiation with no risk estimates, Dalgety Bay, 2008-2017

Cancer Observed Expected O/E 99% CL 99% CL 1-sided p Scotland (ICD-10) Cases (O) Cases (E) 1 Lower 2 Upper 2 value 2 Cases Chronic * * 0.70 0.12 2.19 2,693 lymphocytic leukaemia (C91.1)

Pancreas 10 12.9 0.78 0.29 1.66 7,823 (C25)

Hodgkin * * 1.82 0.39 5.16 0.14 1,670 lymphoma (C81)

Prostate 71 75.1 0.95 0.68 1.27 32,662 (C61)

Testis (C62) * *    2,083

Cervix (C53) * * 1.22 0.26 3.46 0.39 3,303

Certain * *    174 childhood cancers (aged 0-14) CL – Confidence Limit 1 - Based on comparison with age, sex and deprivation matched population of Scotland 2 - Calculated using exact Poisson distribution where O/E≥1 * - values less than 5 suppressed in accordance with ISD’s Data Disclosure Protocol  - numbers too small to calculate

The O/E ratio for childhood leukaemias was 2.58 (99% CI 0.13-11.94, p=0.18) – Table 5.

Table 5. Incidence of childhood leukaemias, Dalgety Bay, 2008-2017

Cancer Observed Expected O/E 99% CL 99% CL 1-sided p Scotland (ICD-10) Cases (O) Cases (E) 1 Lower 2 Upper 2 value 2 Cases Childhood * * 2.58 0.13 11.94 0.18 410 leukaemias

(aged 0-14) CL – Confidence Limit 1 - Based on comparison with age, sex and deprivation matched population of Scotland 2 - Calculated using exact Poisson distribution where O/E≥1 * - values less than 5 suppressed in accordance with ISD’s Data Disclosure Protocol

11

Information Services Division

Discussion We have described cancer incidence in the Dalgety Bay area between 2008 and 2017. Cancer incidence – or risk – was not significantly higher than the rest of Scotland overall, nor was it significantly higher in any individual group of cancers. This includes the two types of cancer (liver and non-Hodgkin lymphoma) that were found to have higher than expected incidence in the previous report (from 2000 to 2009).

The limitations of epidemiological surveys based on residency as a proxy for exposure to risk factors should be acknowledged. The period of time between exposure and the diagnosis of cancer, and the mobility of the population, mean that a true association might not be apparent. Some exposed individuals may have moved elsewhere before a cancer was diagnosed, but conversely, some individuals may have just moved into the area and not been exposed to any local risk factors yet have been diagnosed with cancer.

The risk of environmental exposure leading to a localised increase in cancer rates is difficult to determine from cancer registration statistics alone. Where statistical studies of the type we have conducted highlight a cause for concern, other types of epidemiological study may be carried out involving actual measurement of exposure levels and the exposure histories of those with and without a cancer diagnosis. A fact sheet9 for the non-statistician, produced by the UK and Ireland Association of Cancer Registries, includes a discussion of some of these issues.

12

Information Services Division

References

1 Black R J, Sharp L, Finlayson A R and Harkness E F (1994). Cancer incidence in a population potentially exposed to radium-226 at Dalgety Bay, Scotland. Br J Cancer 69, 140–43

2 COMARE (2014). Fifteenth Report. Radium contamination in the area around Dalgety Bay. PHE. Accessed 22/05/2018 https://www.gov.uk/government/publications/radium-contamination-in- the-area-around-dalgety-bay

3 SEPA (2011). Accessed 22/05/2018 https://www.sepa.org.uk/regulations/radioactive- substances/dalgety-bay/

4 Information Services Division. Cancer Incidence and Prevalence in Scotland (to December 2017). 30th April 2019. https://www.isdscotland.org/Health-Topics/Cancer/Publications/2019-04- 30/2019-04-30-Cancer-Incidence-Report.pdf

5 Boice JD (2006). Ionizing radiation. In Schottenfeld D and Fraumeni Jr, JF (eds). Cancer Epidemiology and Prevention. Third Edition, Oxford University Press; Table 15-1, page 260.

6 Little MP, Wakeford R, Borrego D, et al. Leukaemia and myeloid malignancy among people exposed to low doses (<100 mSv) of ionising radiation during childhood: a pooled analysis of nine historical cohort studies. Lancet Haematology 2018; 5(8);e346-e358. https://doi.org/10.1016/S2352- 3026(18)30092-9

7 Liddell FDK (1984) Simple exact analysis of the standardised mortality ratio. JECH, 38: 85-88

8 Breslow NE, Day NE (1987) Volume II: The Design and Analysis of Cohort Studies: The Design and Analysis of Cohort Studies Vol 2 ... Research on Cancer Scientific Publications) IARC Scientific publications no 82, IARC: Lyon 9 White C, Hounsome L, Gavin A, McConnell H, Witt J. and Ireland Association of Cancer Registries Cancer Cluster Factsheet. June 2017. http://www.ukiacr.org/publication/ukiacr- cancer-cluster-factsheet

13

Information Services Division

Glossary Confidence Limit An estimate of the effect of chance, or random error, on the results. With 99% Confidence Limits, a result that is lower than the lower confidence limit or higher than the higher confidence limit is very unlikely (although not impossible) to be due to chance alone. Between the two Limits, there is a greater likelihood that chance may be responsible for the result.

Datazone Data Zones are the primary geography for the release of small area statistics in Scotland. They are composed of Census Output Areas and are large enough that statistics can be presented accurately without fear of disclosure and yet small enough that they can be used to represent communities.

ICD-10 The 10th revision of the International Classification of Diseases produced by the World Health Organisation (WHO). It assigns codes to particular diseases and conditions.

Incidence Incidence refers to the number of new cases of a condition in a defined population during a defined period

The ratio of the observed (actual number) to the expected (in a O/E population of the same size, in the same calendar year, and with the same age, sex and socio-economic characteristics). In this publication, it is the same as the Standardised Incidence Ratio

Percentage The number per hundred.

P value The probability that a result of this size or more extreme would have occurred by chance alone.

The Scottish Index of Multiple Deprivation (SIMD) provides a relative Scottish Index of Multiple ranking of the data zones in Scotland from 1 (most deprived) to 10 Deprivation (least deprived) based on a weighted combination of data in the domains of Income; Employment; Health; Education, Skills and Training; Geographic Access to Services; Crime; and Housing.

14

Information Services Division

Contact Dr David Morrison, Director Scottish Cancer Registry, Information Services Division Phone: 0131 275 6087 Email: [email protected]

Catherine Thomson, Service Manager Population Health, Information Services Division Phone: 0131 275 7198 Email: [email protected]

Further Information Further Information can be found on the ISD website. For more information on Cancer see the Cancer section of our website. The next release of this publication will be determined by COMARE, but it is anticipated to be no earlier than 2024.

Rate this publication Please provide feedback on this publication to help us improve our services.

15

Information Services Division

Appendices Appendix 1 – List of radiation-associated with cancers CANCERS FREQUENTLY ASSOCIATED WITH RADIATION WITH AUTHORITATIVE RISK ESTIMATES Leukaemia, especially Myeloid (C92) Thyroid (C73) – little risk if exposed >20 yr age Female breast (C50) – little risk if exposed >40 yr age

CANCERS OCCASIONALLY ASSOCIATED WITH RADIATION WITH ROBUST RISK ESTIMATES Lung (C34) – interaction of smoking Stomach (C16) Colon (C18) Oesophagus (C15) Bladder (C67) Ovary (C56) Brain and nervous system (C71-C72) – mainly after high-dose childhood exposure Liver (C22)

CANCERS RARELY ASSOCIATED WITH RADIATION WITH UNCERTAIN RISK ESTIMATES Limited evidence: Kidney (C64) Salivary glands (C07-C08) Non-Hodgkin lymphoma (C82-C86) Myeloma (C90) Effect may be limited to high doses: Skin (C43-C44) Rectum (C19-C20) Uterus (C54-C55) Bone (C40-C41) Connective tissues (C47, C49)

CANCERS NEVER OR SPORADICALLY ASSOCIATED WITH RADIATION WITH NO RISK ESTIMATES Chronic lymphocytic leukaemia (C91.1) Pancreas (C25) Hodgkin lymphoma (C81) Prostate (C61) Testis (C62) Cervix (C53) Certain childhood cancers (ICD-O M8960, 9490, 9500, 9510-9513, age 0-14)

Childhood Leukaemias (C91-C95, aged 0-14)

16

Information Services Division

Appendix 2 – Publication Metadata

Metadata Indicator Description Publication title Analysis of cancers associated with radiation. Dalgety Bay. 2008-2017. Description Analyses of incidences of cancer in Dalgety Bay were carried out by comparing observed and expected numbers of cases of cancer in Dalgety Bay (compared to Scotland as a whole). Theme Health and Social Care Topic Conditions and Diseases Format Excel workbooks Data source(s) Scottish Cancer Registry (SMR06) Date that data are acquired 01 March 2018 Release date 10 December 2019 Frequency Ad hoc Timeframe of data and Data up to 31 December 2016. No delays between data availability and timeliness processing of data for publication. Continuity of data Report include data from 2008 to 2017. Coding of cancer registrations moved from ICD-9 to ICD-10 and from ICD-O to ICD-O2 in incidence year 1997, then to ICD-O3 in incidence year 2006. ICD codes have been back-mapped to 1989 for continuity of reporting. The range of statistics provided does mean that the continuity will vary, and while considered to be very high, any notable discontinuities (eg for specific conditions) will be highlighted within the published data. Revisions statement As with other population-based cancer registries, the Scottish Cancer Registry is dynamic, with ongoing updating of records. Each year's release includes a refresh of the previous years, and as new registrations from previous years come to light, or changes in the coding are taken into account, the numbers may change. The timing of the release is intended to balance the likelihood of significant revision with timeliness of data. Revisions relevant to this None publication Concepts and definitions See the Cancer Information FAQs Relevance and key uses of The number and type of cancer registrations, by sex and geography, allow the statistics planning for provision of cancer treatment services and palliative care planning. Permits indirect measure of success of public health measures and interventions over the longer term. Key uses include: public health surveillance; health needs assessment, planning and commissioning of cancer services; evaluation of the impact of interventions on incidence and survival; clinical audit and health services research; epidemiological studies; and providing information to support genetic counselling and health promotion. Accuracy Registry data are subject to validation at data entry and quality assurance procedures. See the Cancer Information FAQs. Reported data are compared to previous years' figures and to expected trends. Completeness At time of extraction, data for the most recent year are estimated to be at least 100% complete. See above note on Revisions. Routine indicators of data quality are compared to the rest of the UK and to other countries, and are available on the UKIACR website. There have been ad hoc studies of data completeness in the past. See the Cancer Information FAQs.

17

Information Services Division

Comparability COMARE 15th Report 1 Accessibility It is the policy of ISD Scotland to make its web sites and products accessible according to published guidelines. Coherence and clarity The data tables are available upon request. Value type and unit of The observed number and the expected numbers of cases of cancer. measurement Disclosure The ISD protocol on Statistical Disclosure Protocol is followed. Official Statistics Official Statistics designation UK Statistics Authority Not assessed. Assessment Last published Not applicable Next published To be determined by COMARE, but it is anticipated to be no earlier than 2024. Date of first publication 10/12/2019 Help email [email protected] Date form completed 11 November 2019

18

Information Services Division

Appendix 3 – Early access details

Pre-Release Access Under terms of the "Pre-Release Access to Official Statistics (Scotland) Order 2008", ISD is obliged to publish information on those receiving Pre-Release Access ("Pre-Release Access" refers to statistics in their final form prior to publication). The standard maximum Pre-Release Access is five working days. Shown below are details of those receiving standard Pre- Release Access.

Standard Pre-Release Access: Scottish Government Health Department Fife NHS Board Chief Executive Fife NHS Board Communication lead

Early Access for Quality Assurance These statistics will also have been made available to those who needed access to help quality assure the publication: Committee on Medical Aspects of Radiation in the Environment (COMARE) Committee Scottish Environment Protection Agency (SEPA)

19

Information Services Division

Appendix 4 – ISD and Official Statistics

About ISD Scotland has some of the best health service data in the world combining high quality, consistency, national coverage and the ability to link data to allow patient based analysis and follow up. Information Services Division (ISD) is a business operating unit of NHS National Services Scotland and has been in existence for over 40 years. We are an essential support service to NHSScotland and the Scottish Government and others, responsive to the needs of NHSScotland as the delivery of health and social care evolves. Purpose: To deliver effective national and specialist intelligence services to improve the health and wellbeing of people in Scotland. Mission: Better Information, Better Decisions, Better Health Vision: To be a valued partner in improving health and wellbeing in Scotland by providing a world class intelligence service.

Official Statistics Information Services Division (ISD) is the principal and authoritative source of statistics on health and care services in Scotland. ISD is designated by legislation as a producer of ‘Official Statistics’. Our official statistics publications are produced to a high professional standard and comply with the Code of Practice for Official Statistics. The Code of Practice is produced and monitored by the UK Statistics Authority which is independent of Government. Under the Code of Practice, the format, content and timing of statistics publications are the responsibility of professional staff working within ISD. ISD’s statistical publications are currently classified as one of the following: • National Statistics (ie assessed by the UK Statistics Authority as complying with the Code of Practice) • National Statistics (ie legacy, still to be assessed by the UK Statistics Authority) • Official Statistics (ie still to be assessed by the UK Statistics Authority) • other (not Official Statistics) Further information on ISD’s statistics, including compliance with the Code of Practice for Official Statistics, and on the UK Statistics Authority, is available on the ISD website.

20