HSE Health & Safety Executive

Further investigation of possible musculoskeletal and cognitive deficit due to welding in divers identified in the ELTHI diving study

Prepared by the University of Aberdeen for the Health and Safety Executive 2005

RESEARCH REPORT 390 HSE Health & Safety Executive

Further investigation of possible musculoskeletal and cognitive deficit due to welding in divers identified in the ELTHI diving study

Jennie I Macdiarmid PhD John AS Ross MB ChB PhD FRCA Sean Semple PhD *Liesl M Osman MA PhD Stephen J Watt MBBS FRCPEd AFOM **John R Crawford PhD FBPsS CclinPsychol Departments of Environmental and Occupational Medicine *Medicine & Therapeutics and **Psychology University of Aberdeen Aberdeen AB25 2ZP

A questionnaire survey was conducted to assess the health of divers and welders. The aim of the study was to determine if the health impact of a career in diving and welding was greater than welding alone. The response rate was 48%, with 182 diver welders, 108 non-diver welders and 252 non-diver non- welders completing the questionnaire. Complaints of cognitive and musculoskeletal symptoms were more common in diver welders than non-diver welders, and non-diver non-welders. In spite of this observation there was no convincing correlation between symptom complaint and welding exposure or diving experience. Welding fume exposure was significantly higher in non-divers than divers. Sixty percent of divers had welded at pressure, but the prevalence of symptom complaints in these divers did not differ from divers who had not welded at pressure. Further analysis of the ELTHI Diving study showed that among divers other factors, such as diving accidents and chemical exposures (e.g. contaminated breathing gas, petrochemicals and hydrogen sulphide), were associated with reported symptoms. Welding remained a significant predictor of cognitive but not musculoskeletal complaint after adjusting for diving experience, diving accidents, chemical exposure and lifestyle factors. In summary, these studies suggest that welders who dive have an increased the risk of cognitive complaint which is not observed in absence of diving. The cause of this remains unclear. This report and the work it describes were funded by the Health and Safety Executive (HSE). Its contents, including any opinions and/or conclusions expressed, are those of the authors alone and do not necessarily reflect HSE policy.

HSE BOOKS © Crown copyright 2005

First published 2005

ISBN 0 7176 6169 5

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means (electronic, mechanical, photocopying, recording or otherwise) without the prior written permission of the copyright owner.

Applications for reproduction should be made in writing to: Licensing Division, Her Majesty's Stationery Office, St Clements House, 2-16 Colegate, Norwich NR3 1BQ or by e-mail to [email protected]

ii Acknowledgements

The authors of this report would like to thank Miss Eve Clarke and Mrs Linda Hamill (project secretaries) for their work on this project. We would also like to thank Dr David Caughey from CAPITA Health Solutions (Aberdeen) for collaborating in the study.

iii iv EXECUTIVE SUMMARY

INTRODUCTION Within a welding environment there is potential for exposure to various chemical and physical hazards. Welding generates toxic gases and high levels of fine particles containing a complex variety of metals, varying with the type of welding and material used. Welding has been associated with a number of acute and chronic health related hazards, including respiratory, cognitive, musculoskeletal and hearing complaints. The results of the ELTHI Diving study support these findings, showing that an occupational history of welding is associated with an increased prevalence of cognitive, musculoskeletal and hearing complaints. There were, however, insufficient numbers of non-diver welders in the ELTHI Diving study to determine whether diver welders were more at risk than non-divers welders. It is proposed that there could be an interaction between diving and welding, with divers being at greater risk to adverse health effects since many of them weld at pressure in small, poorly ventilated welding habitats.

AIM To determine if the health impact of diving and welding is greater than welding alone, with particular interest in the effect on cognitive and musculoskeletal symptom complaint.

METHODS The study consisted of a postal questionnaire survey. The 5 page questionnaire assessed welding and diving exposure, symptom complaints, diagnosed medical conditions, lifestyle history, cognitive failure (Cognitive Failure Questionnaire (CFQ)) and health related quality of life (SF-12).

Study population: Three age matched study groups were targeted in the survey; diver welders, non-diver welders and non-diver non-welders. Non-divers were men who had worked offshore but never as a diver. Diver welders and non-diver non-welders were selected from men who had previously participated in the ELTHI Diving study. Due to an insufficient number of non-diver welders in the ELTHI Diving study, a new sample of non-diver welders were recruited through CAPITA Health Solutions. These were men who had had an offshore medical examination between 1990-2002, stated their occupation as a welder and had not dived. 361 diver welder, 352 non-diver welders and 503 non-diver non-welders were traced and sent a questionnaire. Non-responders were sent a maximum of 3 questionnaires at 4 week intervals.

Re-analysis of the ELTHI Diving study data: In addition to the questionnaire survey, data from the ELTHI Diving study were re-analysed to include only divers. Using hierarchical logistic regression analysis, diving experience was related to symptoms complaint, adjusting for lifestyle factors, welding, diving accidents and chemical exposures. A separate model was used for each compliant, which were ‘forgetfulness or loss of concentration’, ‘joint pain or muscle stiffness’, ‘back or neck pain’ and ‘impaired hearing’

RESULTS The questionnaire survey had a response rate of 48%, with response being higher among those who had previously completed the ELTHI Diving study (53%) than newly recruited non-diver welders (30%). 182 diver welders, 108 non-diver welders and 252 non-diver non- welders were included in the analysis.

v Cognitive complaint: The prevalence of complaint of ‘forgetfulness or loss of concentration’ was significantly higher among diver welders (33%) than both non-diver welders (10%) and non-diver non-welders (14%). Consistent with this, diver welders scored significantly higher on the CFQ than both non-diving groups, reflecting greater cognitive symptomatology. Non-diver welders and non-diver non-welders did not differ significantly on either of these measures of cognitive complaint.

Musculoskeletal complaint: As observed for cognitive complaint, the complaint of ‘joint pain or muscle stiffness’ and ‘back or neck pain’ was significantly more common among diver welders than both non-diver welders and non-diver non-welders.

The only other symptom to differ between the groups was ‘impaired hearing’, which was more common in diver welders (29%) and non-diver welders (32%), than non-diver non- welders (12%). Physical and mental health related quality of life did not differ significantly between the 3 study groups.

Welding and diving exposure related to symptom complaint: Divers tended to have shorter welding careers and had been exposed to less welding fume than non-divers. The divers, on average, had been exposed to the equivalent of 11 years of welding fume at a concentration of 5mg.m-3 for 240 days per year, while in non-divers equivalent exposure was 36 years. Divers were more likely than non-divers to have welded aluminium (29% vs. 8%). Despite differences between diver and non-diver welders in welding experience, there were no convincing correlations between welding exposure and symptom complaint within this population. Divers who welded at pressure had longer diving careers and greater welding fume exposure than divers who had not welded at pressure, but they were no more likely to complain of cognitive or musculoskeletal symptoms.

Re-analysis of the ELTHI Diving study data: Further analysis of the ELTHI Diving study data in divers, revealed other factors, such as diving accidents and chemical exposures, were related to cognitive, musculoskeletal and hearing complaint. Neurological decompression illness, exposure to contaminated gas while diving, exposure to petrochemicals and work as a welder were associated with the complaint of ‘forgetfulness or loss of concentration’. ‘Back or neck pain’ was associated with contaminated gas, exposure to petrochemicals and neurological decompression illness. ‘Joint pain or muscle stiffness’ was associated with pain only decompression illness, exposure to contaminated gas and exposure to petrochemicals. ‘Impaired hearing’ was associated with exposure to contaminated gas and noise. With respect to diving techniques, after adjusting for these other factors, only mixed gas bounce diving was associated with reported ‘forgetfulness or loss of concentration’ and reported ‘joint pain or muscle stiffness’, and saturation diving related to reported ‘impaired hearing’.

SUMMARY The main finding of the study confirms the previous observations that the health impact of welding and diving is greater than that of welding alone. Cognitive and musculoskeletal complaints were found to be more common in diver welders than non-diver welders. Welding, therefore, should be taken into consideration when assessing the long term health impact of diving at work. In addition, further analysis of the ELTHI Diving study indicates that in addition to welding, other work related factors, such as accidents and chemical exposures, are associated with the complaints studied. Further investigation is needed to determine a causal link. If a causal link were established with chemical exposures then greater focus would need be placed on occupational hygiene and interventions introduced to control possible toxic exposures within the diving sector.

vi CONTENTS

1 INTRODUCTION ...... 1

2 OBJECTIVES ...... 3

3 METHODOLOGY ...... 4 3.1 QUESTIONNAIRE SURVEY ...... 4 3.2 STUDY POPULATION ...... 4 3.3 STATISTICAL ANALYSIS...... 5 3.3.1 Re-analysis of questionnaire data from ELTHI Diving Study ...... 6 3.4 ETHICAL APPROVAL...... 6

4 RESULTS ...... 7 4.1 RESPONSE RATE ...... 7 4.2 RESPONSE BIAS ...... 8 4.3 DEMOGRAPHIC AND LIFESTYLE CHARACTERISTICS...... 8 4.4 OCCUPATIONAL HISTORY...... 8 4.4.1 3-day lost time accidents ...... 10 4.4.2 Welding experience ...... 10 4.4.3 Diving experience ...... 12 4.5 GENERAL HEALTH ...... 12 4.5.1 Symptoms...... 12 4.5.2 Diagnosed medical conditions...... 13 4.5.3 Cognitive Failure Questionnaire ...... 13 4.6 HEALTH RELATED QUALITY OF LIFE ...... 15 4.7 WELDING FUME EXPOSURE AND COGNITIVE COMPLAINT ...... 16 4.8 WELDING FUME EXPOSURE AND MUSCULOSKELETAL COMPLAINT ...... 16 4.9 DIVING EXPOSURE AND COGNITIVE COMPLAINT ...... 17 4.10 DIVING EXPOSURE AND MUSCULOSKELETAL COMPLAINT ...... 18 4.11 THE INTERACTION BETWEEN DIVINGAND WELDING...... 18 4.11.1 Relationship of cognitive complaint with diving related accidents in diver welders……………...... 19 4.11.2 Relationship of musculoskeletal complaint with diving related accidents in diver welders...... 20 4.12 CHANGE IN REPORTED COMPLAINTS AFTER 4 YEARS ...... 20 4.12.1 Cognitive complaint ...... 20 4.12.2 Musculoskeletal complaint...... 21

vii 4.13 RE-ANALYSIS OF QUESTIONNAIRE DATA FROM THE ELTHI DIVING STUDY...... 22 4.13.1 Cognitive Complaint ...... 22 4.13.2 Back or neck pain ...... 22 4.13.3 Joint pain or muscle stiffness ...... 23 4.13.4 Impaired hearing ...... 23 4.13.5 Relationship between saturation diving and work as a welder...... 29 4.13.6 Prevalence of diagnosed medical conditions in divers in relationship to work as a welder ...... 29

5 DISCUSSION ...... 30

6 APPENDICES ...... 33 APPENDIX 1:QUESTIONNAIRE ...... 33 APPENDIX 2:CALCULATION OF WELDING FUME EXPOSURE...... 38 APPENDIX 3: A DESCRIPTION OF THE METALS WELDED UNDER THE CATEGORY OF ‘EXOTICS’ ...... 41

7 REFERENCES ...... 42

viii 1 INTRODUCTION

Welding is associated with a number of health related hazards. The general morbidity and mortality of welders has been reported to be worse than other groups of shipyard workers (1). Respiratory illness is the most commonly reported health hazard of welding, but non- respiratory health problems such as neurological and dermatological effects have been reported (2). Adverse health effects of welding might be attributed to various chemical exposures (e.g. fumes, gases) mainly through inhalation or ingestion, physical exposures (e.g. heat, electricity, noise, vibration) and radiation hazards. Welding is a diverse occupation with respect to hazards depending on the composition of material being welded, the method of welding and the environment in which the welding is taking place. The composition of welding fume is complex and varies according to the type of welding being performed. However, it often contains neurotoxic metals, such as manganese, aluminium, and lead, and toxic gases such as carbon monoxide and nitrogen dioxide. In addition, welders frequently work in confined, poorly ventilated spaces and with an awkward work posture and this also can contribute to the health effects.

WELDING AND COGNITIVE COMPLAINT Welders are exposed to a plethora of inhaled toxins. Metal fume, fumes from the flux used, nitrogen oxides and carbon monoxide are all potential hazards in welding and, for divers, the inert gases used for shielding also pose a hazard due to their narcotic properties at pressure. Carbon monoxide exposure has a potential for both acute and chronic neurological effects and the permitted limits for CO exposure in a hyperbaric welding habitats have been higher (70 ppm 8hour TWA) than the industry norm for working at atmospheric pressure (30ppm 8hour TWA). A study on welding in Manitoba, in addition to identifying higher levels of exposure to carbon monoxide, also documented high exposure levels to manganese and iron (3). Aluminium fume inhalation is also a potential hazard. Both manganese (4) and aluminium (5) exposure have been identified with cognitive impairment in welders and manganese exposure also effects mood and autonomic function (6). Neuropsychiatric symptoms are more frequent in welders with prolonged exposure to lead, manganese or aluminium (7) (8).

WELDING AND MUSCULOSKETETAL COMPLAINT Musculosketetal complaint is more common in welders than in other workers especially in the upper limbs (9). This may be associated with a requirement to work in restricted spaces and the same position for long periods of time. Burdorf et al (10) reported that rate of sickness absence due to musculoskeletal complaints was higher in welders than the general population of workers. These findings are supported by the prevalence of musculoskeletal complaints reported by welders in the examination of the long term health impact of diving study (ELTHI Diving study) (11)(see Table 1).

WELDING AND HEARING COMPLAINT Welders are exposed to high noise levels and a degree of noise induced hearing loss might be expected. Welders are also at risk of hearing loss caused by sparks of metal perforating the eardrum. An increased prevalence of complaint of hearing loss was seen in welders in the questionnaire survey for the ELTHI Diving study (Table 1).

INTERACTION BETWEEN DIVING AND WELDING The adverse effects of welding might be increased by diving. In addition to exposure to welding fumes in an enclosed underwater welding habitat, the effect of a high-pressure environment must be considered. For example, argon commonly used as an inert shielding gas is a narcotic at high pressure. Furthermore there is the potential of high hydrostatic pressure to amplify the toxicity of already hazardous chemicals. In cell culture work, the toxicity of the welding fume component chromate was amplified by hydrostatic pressure (12).

1 THE ELTHI DIVING STUDY The ELTHI Diving study (11) identified differences in symptom complaint between divers and an age matched group of offshore workers. Complaint of ‘forgetfulness or loss of concentration’, ‘joint pain or muscle stiffness’ and ‘hearing loss’ were found to be more common in divers than offshore workers, after adjustments for confounding lifestyle factors. Furthermore, within this population, the prevalence of all these complaints was significantly higher in the welders than non-welders. Controlling for welding experience greatly reduced the difference between divers and offshore workers. The ELTHI Diving study suggests that working as a welder, as either a diver or an offshore worker, increased the risk of the complaints more commonly reported by divers (Table 1). The prevalence of musculoskeletal complaint was of similar proportion in both diver and non-diver welders. The incidence of reported cognitive complaint was 13% higher in diver welders (28%) than diver non-welders (15%), but only 6% higher in non-diver welders than non-diver non-welders. As illustrated in Table 1, however, there were only 47 non-diver welders in the study (representing only 5% of the offshore worker population). With this small number it was not possible to conclude with certainty whether diver welders are at higher risk of suffering these complaints than non-diver welders. Since this is an important consideration in the investigation into the long term health impact of diving, the current study was designed to survey a larger number of non-diver welders to try and resolve this issue.

Table 1 The prevalence (%) of symptom complaints in divers and offshore workers related to working as a welder Offshore workers Divers non-welder welder non-welders welder (n=941) (n=47) (n=1152) (n=351)

Joint pain or muscle stiffness 20% 40% 27% 37%

Back or neck pain 27% 54% 29% 40%

Forgetfulness or loss of concentration 6% 11% 15% 28%

2 2 OBJECTIVES

The aim of this study was to determine whether the health impact of diving and welding is greater than welding alone.

Specific objectives were:

x to identify the effects of diving on the musculoskeletal and cognitive complaint in welders

x to examine any effect of welding on musculoskeletal and cognitive complaint in non- divers

x to re-analyse the ELTHI Diving study data for effects of factors other than diving in relation to cognitive, musculoskeletal complaint and hearing.

3 3 METHODOLOGY

3.1 QUESTIONNAIRE SURVEY

The Questionnaire: The questionnaire used in this study was similar to that used in the ELTHI Diving study (11), with the addition of a section detailing welding experience and the inclusion of the Cognitive Failure Questionnaire. The questionnaire was divided into 5 sections, assessing the following (see Appendix 1):

Demographics and lifestyle: Data of birth, alcohol consumption (including frequency of ‘binge’ drinking1), smoking habit, educational attainment and employment status.

Occupational History: Welding, diving and offshore (topside) experience.

General Health: Health complaints in the form of symptoms regularly suffered reported on a 4 point scale (‘not at all’, ‘slightly’, ‘moderate’ or ‘severe’) and diagnosed medical conditions.

Health Related Quality of Life (HRQOL) - SF12: a 12 item questionnaire used to assess physical and mental HRQOL (13). A lower score reflected lower HRQOL.

Cognitive Failure Questionnaire (CFQ) (14): The CFQ is a 25 item questionnaire used to assess self-reported failures in perception, memory and motor function. It assesses how cognitive failures affect daily routine. A higher score reflects more frequent occurrence of cognitive failures. This was used to give a continuous variable reflecting subjective complaint. It was also thought that the simple report of degree of “forgetfulness or loss of concentration” might be biased in the study sample who had previous experience of the question from previously participating in the ELTHI Diving study and an alternative measure of subjective cognitive report was needed.

The questionnaire was piloted on a small group of diver welders based at the National Hyperbaric Centre, Aberdeen.

Postal Survey: The questionnaire was sent out to all traced participants with a letter explaining the purpose of the study, a recent newsletter from the ELTHI Diving study and a reply paid envelope. The newsletter informed all subjects about the outcome of the ELTHI Diving study and had details of the Health and Safety Executive website on which the study report is published. This information was communicated in order to compensate for bias in the study occasioned by some subjects knowing the result of the ELTHI Diving study and others not. Non-responders were sent a maximum of 3 questionnaires at 4-week intervals.

3.2 STUDY POPULATION The study population consisted of 3 groups; diver welders, non-diver welders and non-diver non-welders.

All the diver welders and non-diver welders who previously participated in the ELTHI Diving study were selected. An age matched random sample of non-diver non-welders was also selected from participants of the ELTHI Diving study. Since there were insufficient non- diver welders from the ELTHI Diving study, a new sample was identified through CAPITA

1 ‘binge’ drinking was defined as the consumption of 8 or more units of alcohol on a single occasion

4 Health Solutions, Aberdeen (formally Liberty Occupational Health). These men had undergone an offshore medical examination with CAPITA Health Solutions between 1990- 2002 and stated their occupation as welder, but not diver. The aim was to identify 500 people in each group. All the participants in the study were men.

Identification and tracing of participants: The accuracy of postal addresses was checked using a data verification company, Data Discovery Ltd, Edinburgh, UK. This company uses publicly available information to verify and trace individuals’ addresses. Current addresses were traced for 1216 participants, all of whom were included in the study.

3.3 STATISTICAL ANALYSIS Univariate analysis was conducted as appropriate to compare lifestyle, welding and diving experience and health complaints between the 3 groups. Where different statistical techniques are used this is described in the results section. In the results, where the mean is reported this is expressed as the mean with 95% confidence interval for the mean (95% CI). All statistical analysis was conducted using SPSS version 11 (SPSS Inc, Minnesota).

Lifestyle characteristics: Body mass index (BMI) was calculated from reported height and weight (weight (kg) / height2 (m 2)). Smoking was expressed in pack years, calculated from the following equation: (‘number of cigarettes smoked per day’ x ‘number of years smoked’) / 20.

Assessment of welding fume exposure: A single variable for welding fume exposure was calculated incorporating welding experience in different environments, the use of different welding techniques and the use of respiratory protective equipment (see Appendix 2). This variable was expressed as mg.m-3 days (number of days exposed to a time weighted average concentration of 1 mg.m-3 welding fume).

Classification of welding metals: Welders were asked to name the 3 metals they most commonly used when welding. The welders were then classified into 4 groups according to the metals they had welded: 1. mild or carbon steel only 2. stainless steel only (not exotic metals) 3. exotic metals only (not stainless steel) 4. stainless steel and exotic metals

It should be noted that welders in groups 2 to 4 would have also welded with mild or carbon steel. The metals included in the group termed ‘exotics’ are described in Appendix 3. Special note was made of experience with aluminium.

Symptoms: Reported symptoms were categorised into two groups; grouping ‘not at all’ or ‘slightly’, and ‘moderate’ or ‘severe’ together. These groupings for cognitive complaints were supported by results from the objective neuropsychological tests in the ELTHI Diving study.

The effect size was calculated for the difference in CFQ scores between the 3 study groups: Cohen’s d = (mean group1 – mean group2) /pooled standard deviation. A Cohen’s d of 0.2 is considered small, 0.5 is moderate and 0.8 is large.

Diving experience and accidents: Divers were asked to report whether they had used each of five different diving techniques and to estimate the number of dives they had carried out

5 using each of the techniques. Diving experience was expressed as the number of dives conducted using SCUBA, surface supplied gas, mixed gas bounce or surface decompression using oxygen. Experience of saturation diving was taken as the number of days spent in saturation.

Accidents associated with decompression illness (pain only, neurological and cerebral gas embolism) and drilling mud burns were reported in the current survey. Reports of having suffered loss of consciousness at pressure, exposure to contaminated gas or underwater explosion while diving were taken from the ELTHI Diving study questionnaire. These reports were quantified as ‘not at all’, ‘once’ or ‘more than once’.

3.3.1 Re-analysis of questionnaire data from the ELTHI Diving study In order to assess the overall importance of work as a welder in health outcomes associated with diving, data for the prevalence of moderate to severe complaint of ‘forgetfulness or loss of concentration’, ‘joint pain or muscle stiffness’, ‘back pain or neck pain’ and ‘hearing impairment’ were re-analysed for divers only. Hierarchical logistic regression analyses were used to assess the effect of diving experience on complaint after adjustment for lifestyle factors (age, smoking, alcohol) and work as a welder. Diving accidents and reported chemical exposures were then added to the model. For hearing complaint reported noise exposure was included and for musculoskeletal complaint body mass index was included. Diving accidents were reported as ‘not at all’, ‘once’ or ‘more than once’. Exposures to chemicals and noise were reported as ‘not at all’, ‘a little’ or ‘a lot’.

Diving experience of each of the five techniques was entered as a continuous variable as described for the ELTHI Diving study. Briefly, data for each of the diving techniques was converted from a categorical into a continuous variable. This was achieved by taking the mid-value of the range in each category. For the highest category, which had no upper limit (e.g. >1000), the value was estimated from information based on a diving history interview with all subjects in a 10% random sample of the questionnaire population since the experience reported at interview correlated well with the data obtained from the questionnaire data. The median value was taken for all those divers who fell into the upper category.

SCUBA >1000 dives (n=44) 1688 dives surface oxygen decompression >1000 dives (n=8) 1584 dives other air/nitrox dives > 500 dives (n=76) 1020 dives saturation >1000 days (n=24) 1500 days

For mixed gas bounce diving none of the divers in the random sample reported to have done more than 500 dives (i.e. in the highest category used in the questionnaire), therefore the midpoint (300 dives) of the previous category was used as the top category.

The prevalence of diagnosed medical conditions was compared between divers who had and had not worked as a welder.

3.4 ETHICAL APPROVAL The study was approved by the Grampian Research Ethics Committee.

6 4 RESULTS

4.1 RESPONSE RATE The response to the questionnaire survey by group is illustrated in Figure 1. Five hundred and forty two completed questionnaires were returned, to give a valid response rate of 48%. The response rate was higher among those who had previously taken part in the ELTHI Diving study (53%) than the newly recruited non-diver welders (30%). Not everyone who returned a completed questionnaire met the criteria for inclusion into the study groups. A final sample of 153 diver welders, 108 non-diver welders and 252 non-diver non-welders were included in the analysis. The 29 divers were excluded from the analysis since they stated in this questionnaire that they had not worked as a welder, despite previously reporting that they had in the ELTHI Diving study.

Diver Non diver Non-diver Welder Welder Non welder

Questionnaires sent 361 352 503

Response to the survey 202 143 274

Completed questionnaires 182 108 252

Excluded: Dead 1 1 2 Wrong address 20 34 15 Abroad 1 0 1

ANALYSIS 153 108 252 response rate* 54% 34% 53%

* exclude those who were dead, sent to the wrong address and living abroad Figure 1 Response rate to the questionnaire survey

Recreational divers: 12 non-diver welders (11%) and 14 non-diver non-welders (6%) reported some recreational diving activity (all except 3 people had done less than 10 SCUBA dives). Consistent with the results from the ELTHI Diving study there was no association found between recreational diving and reported ‘forgetfulness or loss of concentration’ (F2 (1)=0.24 p=0.62) or CFQ score (t (336)=0.1 p=0.91). In the present study, therefore, these recreational divers were included in the non-divers groups.

Non-welder divers: Twenty nine divers who reported to have worked as a welder in the ELTHI diving study, stated in the current study that they had not worked as a welder. These divers were therefore not included in the analysis of the study. They were, however, compared with the diver welders in the main study to identify whether excluding these 29 divers introduced a bias. No difference in age, smoking habit, alcohol consumption or education was found between the non-welder divers and diver welders included in the study. The prevalence of reported complaints did not differ significantly between these groups.

7 Twenty eight percent of non-welder divers compared to 33% of diver welders reported suffering from ‘forgetfulness or loss of concentration’ (p=0.58). The cognitive failure questionnaire score did not differ significantly between groups (p=0.39). Forty eight percent of both non-welder divers and diver welders reported ‘joint pain or muscle stiffness’ (p=1.0). There was no difference between these two dive groups in the duration of their diving career (p=0.42). Excluding these divers does not introduce bias in the main diver welder group. It is quite possible that these subjects were in fact diver welders but were not inclined to complete the detailed welding experience questionnaire.

4.2 RESPONSE BIAS Non-responders: Response bias was assessed by comparing responders and non-responders in the current study using the information they provided in the ELTHI Diving study. Non- responders were on average 2 years younger than responders (p=0.001) but did not differ in body mass index (BMI), educational attainment, employment status or frequency of binge drinking of alcohol. Non-responders were more likely to be current than ex-smokers (p=0.007). There was no difference in diagnosed medical conditions, symptoms or health related quality of life, with the exception of complaint of ‘skin rash or itch’, and diagnosed ‘eczema or hayfever’ and ‘migraine’ which were more common in responders than non- responders.

Late responders: Using previously described methods to identify characteristics of non- responders (15) of comparing those who returned the first questionnaire (‘early responders’) with those who returned the third questionnaire (‘late responders’). The order of questionnaire response did not differ between the 3 study groups (p=0.58). Comparing early and late responders, found no difference in age, body mass index (BMI), educational attainment, employment status or frequency of binge drinking alcohol. Smokers were more likely than non-smokers to be late responders (p=0.02). There was no difference between early and late responders in reported symptoms, health related quality of life, cognitive failure scores or diagnosed medical conditions, with the exception of ‘arthritis’ (p=0.03) and ‘dermatitis’ (p=0.03) which were more common in early responders.

4.3 DEMOGRAPHIC AND LIFESTYLE CHARACTERISTICS There were very few differences in lifestyle characteristics between the study groups (Table 2). The mean age of this population was 50 years old, ranging from 30 to 81 years. The groups did not differ in age, smoking habit, alcohol consumption or BMI. The educational achievements differed significantly between groups, with fewer non-diver welders (15%) than non-diver non-welders (53%) and diver welders (32%) having educational qualifications to A level or above (p<0.001). This difference is likely to reflect their chosen career requiring different academic qualifications rather than a difference in pre-morbid IQ. This has been observed in other occupational studies (16). In the ELTHI Diving study, where the same differences in educational achievements was observed, there was no difference in pre-morbid IQ between divers and non-divers.

4.4 OCCUPATIONAL HISTORY Eighty five percent of the participants were employed at the time of the study (Table 3), although not necessarily in the diving, welding or offshore industry. There was no difference between the 3 groups in the proportion of people employed at the time of the study (p=0.14).

8 Table 2 Lifestyle and demographic characteristics Non-diver Non-diver Diver non-welder welder welder (n=252) (n=108) (n=153)

Age mean (95%CI) 49.9 (48.9-50.9) 48.8 (47.2-50.3) 50.9 (49.6-52.2) p=0.12

Smoking n (%) Current 115 (46%) 42 (39%) 59 (39%) p=0.43 Ex-smoker 88 (35%) 38 (35%) 63 (41%) Non-smoker 49 (19%) 27 (25%) 30 (20%)

Pack years median (IQR) 3 (0-21) 6 (0-23) 4 (0-19) p=0.81

Alcohol units / per week median (IQR) 14 (6-24) 12 (6-20) 11 (5-25) p=0.44

Binge drinking per month never n (%) 30 (12%) 12 (11%) 29 (19%) p=0.20 less than once 80 (32%) 35 (32%) 47 (31%) 1-9 times 109 (43%) 50 (46%) 52 (34%) 10-20 times 20 (8%) 5 (5%) 15 (10%) more than 20 times 8 (3%) 4 (4%) 9 (6%)

Body mass index (kg.m-2 ) mean (95% CI) 27.2 (26.6-27.8) 26.9 (26.3-27.6) 26.8 (26.4-27.3) p=0.91

Educational qualifications none n (%) 31 (12%) 24 (22%) 26 (17%) p=<0.001 O level, Standard Grades 86 (34%) 66 (61%) 75 (49%) A levels, Highers 18 (7%) 2 (2%) 10 (7%) HNC or HND 60 (24%) 12 (11%) 16 (11%) University degree 53 (21%) 2 (2%) 22 (15%)

Currently working in the industry: It was not a requirement of this study for people to be working in their respective industry at the time of the study. Fifty seven percent of diver welders were still working as either a diver or a welder (31% were working as a welder and a diver), while 69% of non-diver welders were still working as a welder and 59% of non-diver non-welders were still working offshore at the time of the study.

Table 3 Employment status Non-diver Non-diver Diver non-welder welder welder (n=252) (n=108) (n=153)

Employed n (%) 211 (83%) 96 (89%) 127 (83%) Unemployed 6 (2%) 6 (5%) 3 (2%) Not working/retired (due to ill health) 12 (5%) 3 (3%) 10 (7%) Retired 23 (9%) 3 (3%) 12 (8%)

9 4.4.1 3-day lost time accidents Diver welders were more likely to report having suffered a 3-day lost time accident (LTA) at work than both non-diver welders and non-diver non-welders. 52% of diver welders reported at least one LTA compared to 31% of non-divers welders and 26% of non-diver non-welders (p<0.001).

4.4.2 Welding experience Table 4 summarizes the welding experience of both the divers and non-divers. Diver welders had shorter welding careers and were less likely to still be working as a welder at the time of the study than non-diver welders. Non-diver welders started welding at an earlier age, with 70% starting to weld professionally before the age of 18 years compared to only 35% of diver welders.

Non-diver welders had used a wider range of welding techniques than diver welders (Table 4). A higher proportion of non-diver welders had used MIG, TIG and FCA welding.

Table 4 Welding experience Non-diver welder Diver welder (n=108) (n=153)

Years welded median (IQR) 26.5 (18.0-34.0) 15.0 (6.0-27.0) p<0.001 Age started welding median (IQR) 16 (16-19) 21 (17-28) p<0.001 Still welding n (%) 72 (69%) 70 (49%) p=0.002

Welding techniques n (%) Manual Metal Arc (MMA) 104 (96%) 137 (90%) p=0.47 Metal Inert Gas (MIG) 96 (89%) 103 (67%) p=<0.001 Tungsten Inert Gas (TIG) 84 (78%) 85 (56%) p=0.02 Oxyfuel 83 (77%) 115 (75%) p=0.52 Flux Cored Arc (FCA) 67 (62%) 58 (38%) p=0.01 Flame or Arc Metal cutting (FAM) 94 (87%) 118 (77%) p=0.37

Welding environments: The majority of divers and non-divers had welded both ‘outdoors’ and ‘indoors in well ventilated areas’ (Table 5). Non-divers were more likely to have worked in ‘small, poorly ventilated environments’ (p<0.001). Seventy percent of non-divers and 56% of divers had welded in all three of these environments. Sixty percent of divers had welded in a pressurised environment (while diving). Four non-divers reported welding in a pressurised environment, but this accounted for less than 20% of their total welding experience and none of these people had ever dived. This type of work is likely to have been conducted while working in a compressed air caisson.

Table 5 Welding work areas Non-diver Diver welder welder Chi2 test

Outdoors 94 (87%) 136 (88%) p=0.64 Indoors in a well ventilated area 99 (92%) 125 (82%) p=0.01 Small, poorly ventilated area 89 (82%) 100 (65%) p=0.002 Pressurised environment 4 (4%)* 90 (60%) - * not while diving

10 Welding fume exposure: Non-diver welders had significantly higher welding fume exposures than diver welders (p<0.001). The median (IQR) welding fume exposure for non- diver welders was 43,623 (15,933-82,485) mg.m-3 .days and for divers was 13,471 (3,602- 39,470) mg.m-3.days. This exposure, in non-diver welders, is equivalent to 36 years of welding at an average exposure of 5 mg.m-3 for 240 days per year and in diver welders to 11 years at the same exposure.

Metals used in welding: Metals commonly used for welding differed between divers and non-divers (Table 6). Non-divers were more likely than divers to have welded stainless steel (p<0.001), while divers were more likely to have only welded using mild or carbon steel. The same proportion in both groups had welded using exotic metals (p=0.35).

Table 6 Metals commonly used in welding Non-diver Diver welder welder

Mild or carbon steel only 13 (12%) 32 (21%) Stainless steel only 27 (25%) 28 (18%) Exotic metals only 9 (8%) 33 (22%) Stainless steel & exotic metals 56 (52%) 43 (28%)

Exotic metals: Table 7 shows the different types of ‘exotic’ metals welded by divers and non-divers (see Appendix 3 for the metals included in each of these groups). Aluminium was the most commonly welded ‘exotic’ metal, and was welded by more diver welders than non- diver welders (p<0.001). Divers more commonly used brass than non-divers. Non-diver welders were more likely to weld copper and chromium than divers.

Table 7 The type of exotic metals welded by non-divers and divers Non-diver Diver Chi2 test Metal groups welder welder

Aluminium 9 (8%) 44 (29%) p=<0.001 Copper 22 (20%) 10 (7%) p=0.002 Zinc 9 (8%) 6 (4%) p=0.19 Brass 5 (5%) 17 (11%) p=0.04 Chromium 18 (17%) 4 (3%) p<0.001 Miscellaneous 6 (6%) 11 (7%) p=0.48

Welding related accidents: Non-diver welders were more likely than diver welders to have suffered metal fume fever (55% vs. 21%, p<0.001). Adjusting the data for the longer welding careers of non-divers did not alter this relationship. Furthermore, this relationship remained statistically significant after removing from the sample divers who had welded at pressure (n=90). Thirty eight percent of non-diver welders compared to only 7% of diver welders had suffered metal fume fever more than 5 times during their welding career.

Welding ‘exotic’ metals and metal fume fever: Metal fume fever was more common among welders who had welded zinc based metals (p=0.04), chromium based metals (p=0.01) and copper based metals (p=0.05). It was not possible to compare the use of different metals and metal fume fever in divers and non-divers separately because of the small numbers in the

11 individual groups. The same proportion of non-divers (19%) and divers (20%) reported to have suffered a major electric shock while welding (p=0.57).

4.4.3 Diving experience

Duration of diving career: The mean (95% CI) diving career was 20.0 (17.6-20.3) years.

Diving techniques: The majority of divers had used SCUBA, surface demand and surface oxygen decompression diving techniques (Table 8).

Table 8 Diving techniques Used the Number of dives* technique (median (IQR)) n (%)

SCUBA 148 (97%) 300 (100-1000) n =102 Surface demand 142 (93%) 500 (100-1000) n =99 Surface oxygen decompression 130 (85%) 250 (60-575) n =88 Mixed gas bounce 87 (57%) 45 (12-100) n =58 Saturation 96 (61%) 500 (108- n =61 1100)** * based on those divers having used the technique (n refers to the number of divers with information for the number of dives for each dive type) ** days in saturation

Diving accidents: Seventy-four divers (48%) had suffered decompression illness (DCI) in the form of pain only DCI, neurological DCI or cerebral gas embolism, at least once (Table 9). There was, however, no difference in the prevalence of diving related accidents between divers who welded at pressure and those who had not welded at pressure.

Table 9 Diving related accidents Welded at pressure n (%) Number of divers (%) No Yes Chi2 (n=59) (n=90) test

Pain only DCI 71 (46%) 27 (46%) 42 (47%) p=0.91 Neurological DCI 17 (11%) 6 (10%) 11 (12%) p=0.70 Cerebral gas embolism 9 (6%) 6 (10%) 3 (3%) * Drilling mud burn 37 (24%) 12 (20%) 25 (28%) p=0.30 * it was not possible calculate a p-value due to the small number of people in one of the groups

4.5 GENERAL HEALTH Reported head injury was more common in divers (24%) than both non-diver welders (8%) and non-diver non-welders (8%) (p<0.001).

4.5.1 Symptoms The percentage in each group reporting ‘moderate’ or ‘severe’ symptoms is summarized in Table 10. The symptoms that differed significantly between groups included ‘forgetfulness or

12 loss of concentration’, ‘joint pain or muscle stiffness’, ‘back or neck pain’ and ‘impaired hearing’. Diver welders more commonly reported ‘forgetfulness or loss of concentration’ than both non-diver welders and non-diver non-welders. The prevalence of ‘forgetfulness or loss of concentration’ did not differ between non-diver welders and non-diver non-welders. The same pattern was observed for ‘joint pain and muscle stiffness’ and ‘back pain or neck pain’, with more divers than non-divers complaining of these symptoms, with no difference between non-divers welders and non-diver non-welders. ‘Impaired hearing’ was more common in welders (both diver and non-diver) than non-welders, but the reliability of this self-reported complaint is questioned. Results from the ELTHI Diving study found, unlike other reported symptoms, that self-reported complaint of hearing loss did not correlate well with audiogram assessment of hearing loss.

Table 10 Symptoms (‘moderate’ or ‘severe’) suffered regularly Non-diver Non-diver Diver Chi2 test non-welders welders welders

Forgetfulness or loss of concentration 34 (14%) 11 (10%) 50 (33%) p<0.001 Joint pain or muscle stiffness 66 (26%) 34 (32%) 72 (47%) p<0.001 Back pain or neck pain 74 (29%) 36 (33%) 72 (47%) p=0.001 Impaired hearing 31 (12%) 34 (32%) 44 (29%) p<0.001 Impaired vision (not corrected by spectacles) 16 (6%) 13 (12%) 20 (13%) p=0.07 Breathlessness 20 (8%) 10 (9%) 10 (7%) p=0.70 Cough or wheeze 18 (7%) 6 (6%) 10 (7%) p=0.83 Abdominal pain, diarrhoea, constipation or nausea 17 (7%) 9 (8%) 11 (7%) p=0.91 Skin rash or itch 23 (9%) 8 (7%) 16 (11%) p=0.71 Muscle weakness or tremor 11 (4%) 3 (3%) 12 (8%) * Unsteadiness when walking, dizziness or poor 9 (4%) 0 (0%) 6 (4%) * balance * it was not possible calculate a p-value due to the small number of people in one of the groups

4.5.2 Diagnosed medical conditions Table 11 summarizes the diagnosed medical conditions reported by each group. The only conditions that differed between groups were arthritis and vibration white finger. Reported arthritis was more common in diver welders than non-diver non-welders, but not non-diver welder. This relationship was retained after controlling for age. Welders (both diver and non-diver) were more likely to report suffering vibration white finger than non-welders.

4.5.3 Cognitive Failure Questionnaire Divers scored significantly higher on the CFQ than both non-diver non-welders and non-diver welders (Table 12), reflecting a greater degree of cognitive failures among diver-welders. Post-hoc tests revealed that there was no difference between non-diver welders and non-diver non-welders (Tables 13).

13 Table 11 Diagnosed medical conditions Non-diver Non-diver Diver Chi2 test non-welders welders welders

Arthritis 21 (8%) 13 (12%) 26 (17%) p=0.05 Vibration white finger 5 (2%) 7 (7%) 18 (12%) p=0.001 Bronchitis 11 (4%) 5 (5%) 10 (7%) p=0.66 High blood pressure 40 (16%) 14 (13%) 23 (15%) p=0.66 Asthma 22 (9%) 9 (8%) 12 (8%) p=0.89 Dermatitis 28 (11%) 9 (8%) 16 (11%) p=0.65 Eczema 44 (18%) 16 (15%) 23 (15%) p=0.59 Depression 24 (10%) 11 (10%) 23 (15%) p=0.31 Ulcer 13 (5%) 8 (7%) 17 (11%) p=0.14 Heart disease 12 (5%) 8 (7%) 6 (4%) p=0.42 Migraine 18 (7%) 4 (4%) 14 (9%) * Cancer 7 (3%) 1 (0.9%) 4 (3%) * Diabetes 6 (3%) 5 (5%) 1 (0.7%) * Stroke 1 (0.4%) 0 (0%) 1 (0.7%) * Epilepsy 0 0 0 *

Taking medication 40 (33%) 14 (32%) 23 (43%) p=0.12 * it was not possible calculate a p-value due to the small number of people in one of the groups

Table 12 Cognitive failure questionnaire (CFQ) scores Non-diver Non-diver Diver non-welder welder welder CFQ score mean (95% CI) 32.8 (31.1-34.5) 35.7 (32.8-38.5) 40.3 (37.8-42.7) p<0.001

Table 13 Cognitive failure questionnaire (CFQ) scores: Post Hoc Analysis (method)

Group comparison Mean difference in 95% Confidence CFQ interval of difference (effect size) (effect size) non-diver non-welder -2.9 -6.7 - +1.0 p=0.20 vs (-0.20) (-0.47 - +0.07) non-diver welder

non-diver non-welder -7.4 -10.9- -4.0 p=<0.001 vs (-0.52) (-0.77 - -0.28) diver welder

non-diver welder -4.6 -8.9 - -0.3 p=0.03 vs (-0.32) (-0.63 - -0.02) diver welder

While there were significant differences at less than the 5% level between diver welders and the other two groups, the comparison between non-diver welders and non-diver non-welders was not conclusive. While the null hypothesis of no difference between the two groups cannot

14 be rejected, the 95% confidence interval of the difference between the two groups indicates that the presence of a small to medium effect cannot be statistically ruled out.

The effect size of the difference between diver welders and non-diver non-welders was 0.52 indicating a medium effect. Another way of putting this is that 33% of the diver welder population fell outside the range of CFQ scores seen in non-diver non-welders.

Compliant of ‘forgetfulness or loss of concentration’ and CFQ: There was a strong relationship between the complaint of ‘forgetfulness or loss of concentration’ and CFQ score (ANOVA: F (3,466) = 94.8, p<0.001), as illustrated in Figure 2. Reporting of ‘severe’ ‘forgetfulness or loss of concentration’ was more common in divers welders (n=10 (7%)) than non-diver non-welders (n=3 (1%)) and non-diver welders (n=1 (1%)).

70

60

50

40 CFQ score (mean 95%CI) 95%CI) score (mean CFQ

30

20

not at all s lightly m oderately sev erely Fo rgetfu lness o r lo ss of con cen tratio n

Figure 2 Relationship between complaint of ‘forgetfulness or loss of concentration’ and CFQ scores (mean (95%CI))

4.6 HEALTH RELATED QUALITY OF LIFE Physical (PCS) and mental (MCS) component HRQOL scores are shown in Table 14. PCS and MCS do not differ between groups. The mean PCS and MCS in these groups is similar to the population norm of 51.2 and 51.4, respectively, for men of a similar age in the UK (17).

Table 14 Health related quality of life (PCS and MCS scores) Non-diver Non-diver Diver non-welder welder welder

PCS mean (95%CI) 50.7 (49.5-51.8) 50.1 (48.7-51.5) 49.2 (47.6-50.7) p=0.3 MCS 51.0 (49.8-52.3) 52.0 (50.5-53.5) 52.0 (50.5-53.5) p=0.5

15 4.7 WELDING FUME EXPOSURE AND COGNITIVE COMPLAINT

Complaint of ‘forgetfulness or loss of concentration’: Table 15 shows the median welding fume exposure of diver welders and non-diver welders complaining of ‘forgetfulness or loss of concentration’. Welding fume exposure was higher among those complaining of ‘forgetfulness or loss of concentration’ for both divers and non-divers.

Table 15 Welding fume exposure of divers and non-divers complaining of ‘forgetfulness or loss of concentration’ Welding fume exposure (mg.m-3. days) Mann Whitney

Not Forgetful Forgetful

All welders median (IQR) 21,158 (4,890-52,479) 42,565 (12,258-73,374) p=0.01 n=150 n=46

Non-diver welders 39,646 (14,776-78,021) 78,246 (61,540-102,584) p=0.03 n=75 n=9

Diver welders 7,402 (2,272-24,788) 24,888 (11,664-54,462) p<0.001 n=75 n=37

Cognitive Failure Questionnaire: CFQ score was not related to welding fume exposure in either divers (Pearson’s correlation: r =0.09, p=0.37) or non-divers (Pearson’s correlation: r =0.03, p=0.78).

Welding ‘exotic’ metals and cognitive complaint: The prevalence of reported ‘forgetfulness or loss of concentration’ did not differ significantly for any of the groups of exotic metals with the exception of the group of miscellaneous metals. Welders who had welded metals from this group were more likely than other welders to report this cognitive complaint (44% vs. 23%, p=0.05). This association was apparent in divers but none of the non-divers who had used this group of metals reported the complaint. However, the small number of the divers in this group and wide variety of metals included in the miscellaneous groups made the interpretation of this observation difficult. It was observed that divers who had welded aluminium were less likely to report a cognitive complaint than divers who had not welded aluminium (23% vs. 41%, p=0.03).

4.8 WELDING FUME EXPOSURE AND MUSCULOSKELETAL COMPLAINT Musculoskeletal complaints, including the complaints of ‘joint pain or muscle stiffness’ and ‘back pain or neck pain’, and diagnosed arthritis, were not associated with welding fume exposure in this population (Table 16). A weak relationship among non-diver welders was observed between welding fume exposure and ‘back pain or neck pain’, but not diver welders (p=0.04). It should be noted that the same relationships with musculoskeletal complaint were

16 observed when welding exposure was based on the total number of days welded rather than welding fume exposure.

Table 16 Welding fume exposure of divers and non-divers with musculoskeletal complaints

Welding fume exposure (mg.m-3. days) Mann Whitney

No joint pain or muscle Joint pain or muscle stiffness stiffness All welders median (IQR) 23,519 (6,156-58,247) 23,271 (6,699-68,661) p=0.69 n=122 n=73

Non-diver welders 40,213 (15,261-78,820) 63,964 (18,724-100,187) p=0.27 n=61 n=24

Diver welders 12,984 (2,307-28,789) 14,310 (4,856-45,660) p=0.32 n=61 n=49

No back or neck pain Back or neck pain All welders 23,138 (6,107-52,341) 24,659 (7,321-72,138) p=0.35 n=116 n=78

Non-diver welders 39,159 (14,206-66,017) 74,497 (23,729-102,112) p=0.04 n=56 n=28

Diver welders 14,098 (2,289-28,836) 13,720 (5,018-44,320) p=0.37 n=60 n=50

No arthritis Arthritis All welders 22,213 (6,215-61,126) 28,883 (13,196-91,719) p=0.14 n=162 n=29

Non-diver welders 40,212 (14,396-80,476) 61,120 (43,609-113,427) p=0.15 n=73 n=10

Diver welders 12,984 (3,228-41,213) 24,392 (5,052-50,870) p=0.16 n=89 n=19

4.9 DIVING EXPOSURE AND COGNITIVE COMPLAINT The duration of a divers career was not related to the complaint of ‘forgetfulness or loss of concentration’ (t (142) = 1.01, p=0.31) or CFQ score (Pearson’s correlation r=-0.13, p=0.13). Experience using different diving techniques was not related to CFQ scores or complaint of ‘forgetfulness or loss of concentration’ in this group of diver welders (Table 17). Similarly specific dive experience was not related to CFQ scores.

17 Table 17 Dive experience of ‘forgetful’ and ‘non-forgetful’ diver welders Used the diving technique n (%) Median (IQR) dives for those divers who have used the technique non-forgetful forgetful Chi2 non-forgetful forgetful Mann (n=102) (n=50) test Whitney SCUBA 98 (97%) 49 (98%) * 300 (100-1000) 175 (58-1000) p=0.09 Surface demand 92 (92%) 49 (98%) * 600 (100-1000) 400 (100-1000) p=0.37 Surface oxygen 83 (84%) 46 (92%) p=0.17 200 (50-500) 200 (60-500) p=0.69 decompression Mixed gas bounce 53 (55%) 33 (66%) p=0.14 55 (15-100) 30 (11-100) p=0.30 Saturation 59 (61%) 33 (66%) p=0.54 400 (93-1150) 440 (200-1000) p=0.49 * it was not possible calculate a p-value due to the small number of people in one of the groups

4.10 DIVING EXPOSURE AND MUSCULOSKELETAL COMPLAINT

Diving exposure was not related to the complaint of ‘joint pain or muscle stiffness’ in this group of divers. The total number of years they had worked as a diver was not related to this complaint (t (141) = 0.66, p=0.51). Furthermore, specific dive techniques were not related to musculoskeletal complaint (Table 18).

Table 18 Dive experience of diver welders with and without ‘joint pain or muscle stiffness’ Used the diving technique n (%) Median (IQR) dives for those divers who have used the technique

no joint pain joint pain Chi2 no joint pain joint pain Mann (n=78) (n=72) test Whitney

SCUBA 76 (97%) 70 (97%) * 300 (100-1000) 250 (100-1000) p=1.00 Surface demand 69 (90%) 71 (99%) * 500 (100-1000) 600 (100-1000) p=0.13 Surface oxygen 64 (83%) 64 (90%) p=0.21 200 (50-563) 200 (55-500) p=0.60 decompression Mixed gas bounce 48 (63%) 38 (55%) p=0.32 34 (10-100) 75 (20-100) p=0.95 Saturation 52 (68%) 39 (56%) p=0.11 500 (200-1000) 300 (111-1000) p=0.55 * it was not possible calculate a p-value due to the small number of people in one of the groups

4.11 THE INTERACTION BETWEEN DIVING AND WELDING

The interaction between diving and welding was explored by comparing divers who had welded at pressure (n=90 (60%)) with those who had not (n=59 (40%)). There was no difference in age between these 2 groups, which both had a mean age of 50.7 year (p=0.99).

Welding fume exposure: Divers who had welded at pressure had overall significantly greater welding fume exposure than those who had not welded at pressure (welded at pressure

18 (median (IQR) = 16,191 (5,548-45,660) mg.m-3.days, not welded at pressure = 9,339 (1,998- 21,464) mg.m-3.days, p=0.02).

Diving exposure: Divers who welded at pressure had slightly longer diving careers (mean (95%CI) =19.8 (18.1-21.5) years) than divers who had not welded at pressure (17.2 (15.0- 19.5) years) (p=0.06). With the exception of surface decompression diving, which was more common among divers who had welded at pressure, there were no differences in the type of dive techniques used.

Health complaints: As illustrated in Table 19, no difference in the prevalence of reported symptoms or diagnosed medical conditions were seen between divers who had and had not welded at pressure. Comparing only those who had saturation dived, there was no difference in complaint of ‘forgetfulness or loss of concentration’ between divers who had welded at pressure (37%) and those who had not welded at pressure (35%) (p=0.88).

Table 19 Health complaint of divers who have and have not welded at pressure Divers welded at pressure No (n=59) Yes (n=90)

Forgetfulness or loss of concentration 16 (27%) 33 (37%) p=0.25 Joint pain or muscle stiffness 31 (53%) 38 (42%) p=0.22 Back pain or neck pain 31 (53%) 38 (42%) p=0.19

Arthritis 13 (22%) 12 (13%) p=0.13 Vibration white finger 7 (12%) 11 (12%) p=0.96

CFQ mean (95% CI) 38.5 (34.6-42.3) 41.7 (38.4-45.0) p=0.21

The proportion of divers’ total welding experience carried out at pressure ranged from 1 to 100% (median (IQR) = 20% (5-50%)). Twenty divers (22%) had performed more than 50% of their welding at pressure. There was no relationship between the percent of welding done at pressure and CFQ score (Pearson’s correlation r = -0.07, p=0.52). Health complaints of ‘forgetfulness or loss of concentration’, ‘joint pain or muscle stiffness’ and ‘back or neck pain’ were not related to the percentage of welding done at pressure. In six divers, welding career and diving career did not overlap in time. One of these divers had cognitive complaint.

4.11.1 Relationship of cognitive complaint with diving related accidents in diver welders The only diving related accident associated with the complaint of ‘forgetfulness or loss of concentration’ was cerebral gas embolism (Table 20). The prevalence of this in divers complaining of forgetfulness or loss of concentration was five times higher than in divers not complaining, but due to the small numbers it was not possible to determine the significance of the observation. The relationships between complaint of forgetfulness and diving related accidents were confirmed using the cognitive failure questionnaire scores. The only significant difference in CFQ scores was for those who had suffered a cerebral gas embolism (CGE), which was significantly higher than for those who had not suffered a CGE (Student’s

19 t (n=137) = 2.2, p=0.03, mean difference 11.3). Forgetful diver welders were more likely than non-forgetful diver welders to have had a drilling mud burn (38% vs. 17%, p=0.005).

Table 20 Prevalence of diving related accidents in forgetful and non-forgetful diver welders Non-forgetful diver Forgetful diver Chi2 test welders welders

Pain only DCI 47 (47%) 23 (46%) p=1.00 Neurological DCI 11 (11%) 6 (12%) p=0.71 Cerebral gas embolism 3 (3%) 5 (10%) ** Exposure to contaminated gas* 36 (36%) 23 (47%) p=0.22 Loss of consciousness under pressure* 10 (10%) 7 (14%) p=0.55 Underwater explosion* 21 (21%) 17 (35%) p=0.10 * information about these accidents was taken from the ELTHI Diving study as it was not reported in the current. ** it was not possible to calculate a p-value due to the small number of people in one of the groups

4.11.2 Relationship of musculoskeletal complaint with diving related accidents in diver welders Diving related accidents (decompression illness and non-decompression illnesses) were not related to the complaint of ‘joint pain or muscle stiffness’ (Table 21).

Table 21 Prevalence of diving related accidents in diver welders with and without ‘joint pain or muscle stiffness’ No Joint pain Joint pain Chi2 test

Pain only DCI 37 (47%) 29 (40%) p=0.38 Neurological DCI 10 (13%) 7 (10%) p=0.55 Cerebral gas embolism 2 (3%) 7 (10%) ** Exposure to contaminated gas* 27 (37%) 30 (42%) p=0.52 Loss of consciousness under pressure* 6 (9%) 12 (17%) p=0.12 Underwater explosion* 19 (27%) 19 (38%) p=0.92 * information about these accidents was taken from the ELTHI Diving study as it was not reported in the current study. ** it was not possible calculate a p-value due to the small number of people in one of the groups

4.12 CHANGE IN REPORTED COMPLAINTS AFTER 4 YEARS Since the majority of subjects in this study also completed the ELTHI Diving study it was possible to observe the change in prevalence of the symptom complaints in these group over time. The current study was conducted 4 years after the ELTHI Diving study.

4.12.1 Cognitive complaint Four hundred and four subjects had complete data for ‘forgetfulness or loss of concentration’ in both studies. Table 22 shows the group prevalence of the complaint of ‘forgetfulness or loss of concentration’ (moderate or severe) at baseline (ELTHI Diving study) and after 4 years (current study). The prevalence of this complaint increased among both the non-diver non-welders and the diver welders, but remained stable among the non-diver welders.

20 The change in severity of complaint was also assessed at an individual level. Subjects were classified as ‘more forgetful’ when they reported this complaint as ‘moderate or severely’ in the current study and previously reported ‘not at all or slightly’, and ‘less forgetful’ when they reported ‘not at all or slightly’ in the current study and previously ‘moderate or severely’ forgetful (Table 22). Diver welders were more likely than the other groups to change ‘forgetful’ status (p<0.001). Of those who reported a change in status, diver welders were less likely than non-diver non-welders to become ‘more forgetful’ (p=0.04), but the overall prevalence of this complaint remained significantly higher among diver welders.

Table 22 Prevalence of complaint of ‘forgetfulness or loss of concentration’ at baseline and after 4 years Non-diver Non-diver Diver non-welder welder welder (n=229) (n=29) (n=146)

Baseline 8% 7% 32% After 4 years 14% 7% 34%

Individual change in complaint More forgetful 9% 0% 14% Less forgetful 2% 0% 12%

No change 89% 100% 74%

4.12.2 Musculoskeletal complaint The change in complaint of ‘joint pain or muscle stiffness’ was again assessed at a group and individual level (Table 23). The group prevalence of this complaint increased among both the non-diver non-welders and the diver welders, but decreased among the non-diver welders. At an individual level, non-diver welders were more likely than non-diver non-welders and divers welders to change status over time (p=0.002). The prevalence of ‘joint pain’ decreased in non-diver welders, with a higher proportion reporting having ‘less joint pain’ after 4 years. The inverse was observed for both diver welders and non-diver non-welders where, of those changing status, a higher proportion complained of ‘more joint pain’ after 4 years. Of those changing status, however, there was no significant difference between groups in the proportion complaining of ‘more joint pain’ (p=0.15).

There was no relationship with age and the groups changing status, or those reporting ‘more joint pain’. There was no association between the change in status and the proportion who were currently working as either a diver or a welder and those who no longer worked in these professions.

21 Table 23 Prevalence of complaint of ‘joint pain or muscle stiffness’ at baseline and after 4 years Non-diver Non-diver Diver non-welder welder welder (n=235) (n=33) (n=145)

Baseline 23% 36% 42% After 4 years 28% 24% 48%

Individual change in complaint More joint pain 11% 15% 15% Less joint pain 6% 27% 9%

No change 83% 58% 76%

4.13 RE-ANALYSIS OF QUESTIONNAIRE DATA FROM THE ELTHI DIVING STUDY

Hierarchical logistic regression analysis was conducted to relate diving experience to symptom complaints. For each of the complaint models diving experiences was entered first, then adjusted for lifestyle factors (age, smoking, alcohol) and work as a welder, and finally adjusted for diving accidents and chemical exposures. Tables 24-28 show the final model including all these factors for each of the symptom complaints.

4.13.1 Cognitive complaint In the first step of the model, adjusting for lifestyle and work as a welder, saturation diving (p=0.006) and mixed gas bounce diving (p=0.006) was found to be associated with cognitive complaint. Surface oxygen decompression (p=0.063) could not be ruled out as having an effect. As shown in Table 24, in the final model adjusted for diving accidents and chemical exposure as well as lifestyle and welding, only mixed gas bounce diving experience was associated with ‘forgetfulness or loss of concentration’. Neurological decompression illness, contaminated gas accidents and exposure to petrochemicals were also significant, as was age and binge drinking. Work as a welder remained statistically significant after adjusting for these other factors.

When the model for cognitive complaint was fitted for only diver welders adjusted only for lifestyle factors no association with experience of any of the dive techniques studied was found. Although there was an increased odds ratio for risk of forgetfulness associated with binge drinking more than 20 times per month (p=0.018), there was no association with the levels of binge drinking when considered as a group. In the model adjusted for diving accidents and chemical exposures, complaint was associated with neurological decompression illness and exposure to contaminated gas and hydrogen sulphide. Age was also significant (Table 25).

4.13.2 Back or neck pain In the logistic regression model relating diving experience to back or neck pain adjusting only for lifestyle factors and work as a welder, saturation diving (p=0.01) was associated with complaint along with cigarette smoking (p<0.001) and work as a welder (p=0.002). In the final model adjusted for diving accidents and chemical exposure, only the association with

22 cigarette smoking remained, the other significant associations included pain only decompression illness, contaminated gas accidents, exposure to petrochemicals and head injury. Association with work as a welder (p=0.08) and neurological decompression illness (p=0.06) could not be ruled out (Table 26).

4.13.3 Joint pain or muscle stiffness In the model relating diving experience to joint or muscle stiffness adjusting only for lifestyle factors and work as a welder, mixed gas bounce diving (p=0.003) was associated with complaint along with cigarette smoking (p=0.018), work as a welder (p=0.007) and BMI (p=0.028). In the final model adjusted for diving accidents and chemical exposure, the association with mixed gas bounce diving and BMI remained with other significant associations including pain only decompression illness, contaminated gas accidents, exposure to petrochemicals and head injury. An association with hydrogen sulphide exposure was also seen but its significance is doubtful as report of low level exposure indicated a lower risk of complaint than no exposure at all (Table 27). Work as a welder did not remain statistically significant (p=0.11) in the final model including diving accidents and chemical exposure.

4.13.4 Impaired hearing In the logistic regression model relating diving experience to hearing complaint adjusting only for lifestyle factors and work as a welder, saturation diving (p=0.005) was associated with complaint along with work as a welder (p=0.013) and age (p<0.001). In the final model adjusted for diving accidents and chemical exposure, the association with age remained with other significant associations identified with noise and contaminated gas exposures (Table 28). The association with saturation diving could not be ruled out (p=0.06), but in this final model there was no significant association with welding (p=0.17).

23 Table 24 Cognitive Complaint in all divers from the ELTHI Diving study Factor Mean Odds Ratio 95% CI for OR P Age (years) 0.97 0.95-0.99 0.01 Smoking (pack years) 1.00 0.99-1.01 0.62 Binge drinking 0.02 never 1.00 less than monthly 1.81 1.02-3.21 0.04 1-9 times per month 1.54 0.0.90-2.64 0.12 10-20 times per month 2.03 1.10-3.77 0.02 > 20 times per month 3.21 1.56-6.61 0.002 Work as a welder 1.54 1.11-2.12 0.009

Diving techniques (per 100 dives/days): SCUBA 0.99 0.97-1.01 0.41 Other air/nitrox dives 0.96 0.92-1.00 0.07 Surface decompression 1.01 0.97-1.05 0.59 Mixed gas bounce 1.32 1.11-1.56 0.002 Saturation days 1.03 0.99-1.06 0.12

Accidents Neurological DCI 0.005 never 1.00 once 1.46 0.89-2.39 0.13 >once 2.87 1.45-5.66 0.002 Pain only DCI 0.26 never 1.00 once 1.31 0.89-1.93 0.17 > once 1.3 0.87-1.95 0.21 Underwater explosion 0.13 never 1.00 once 1.33 0.78-2.26 0.29 > once 1.62 0.98-2.69 0.06 Loss of consciousness at pressure 0.73 never 1.00 once 1.19 0.71-3.13 0.51 > once 0.75 0.18-3.13 0.69 Contaminated gas 0.01 never 1.00 once 1.19 0.81-1.73 0.39 > once 1.77 1.21-2.60 0.003 Head injury 1.21 0.84-1.73 0.31

Chemical exposure Petrochemicals 0.02 never 1.00 a little 1.09 0.70-1.68 0.72 a lot 1.79 1.06-3.01 0.03 Hydrogen sulphide 0.18 never 1.00 a little 1.03 0.74-1.43 0.88 a lot 1.94 0.95-3.97 0.68 n=1442 with 98 cases excluded because of missing data. 262 subjects had cognitive complaint. Binge drinking was defined as the consumption of more than 8 units of alcohol in one session.

24 Table 25 Cognitive complaint in diver welders from the ELTHI Diving study Factor Mean Odds Ratio 95% CI for OR P Age (years) 0.96 0.92-0.99 0.03 Smoking (pack years) 1.00 0.98-1.02 0.64 Binge drinking 0.15 never 1.00 less than monthly 1.81 0.51-6.39 0.36 1-9 times per month 2.10 0.66-6.65 0.21 10-20 times per month 1.72 0.47-6.24 0.41 > 20 times per month 5.39 1.35-21.57 0.02

Diving techniques (per 100 dives/days): SCUBA 1.00 0.95-1.04 0.82 Other air/nitrox dives 0.98 0.91-1.05 0.52 Surface decompression 0.96 0.96-1.04 0.33 Mixed gas bounce 0.96 0.66-1.38 0.81 Saturation days 1.05 0.99-1.11 0.13

Accidents Neurological DCI 0.01 never 1.00 once 2.00 0.83-4.70 0.12 >once 8.42 1.72-41.31 0.009

Pain only DCI 0.64 never 1.00 once 0.98 0.46-2.08 0.96 > once 0.69 0.32-1.52 0.36

Underwater explosion 0.48 never 1.00 once 1.05 0.42-2.59 0.92 > once 1.67 0.72-3.86 0.23

Loss of consciousness at pressure 0.72 never 1.00 once 1.03 0.40-2.64 0.96 > once 2.47 0.27-22.37 0.42 Contaminated gas 0.02 never 1.00 once 1.31 0.63-2.74 0.47 > once 2.60 1.32-5.11 0.006 Head injury 1.03 0.52-2.04 0.92

Chemical exposure Petrochemicals 0.12 never 1.00 a little 0.49 0.19-1.23 0.13 a lot 0.85 0.31-2.37 0.76 Hydrogen sulphide 0.049 never 1.00 a little 1.06 0.58-1.95 0.84 a lot 7.12 1.48-34.98 0.016 n=334 with 25 cases excluded because of missing data. 91 subjects had cognitive complaint. Binge drinking was defined as the consumption of more than 8 units of alcohol in one session.

25 Table 26 Back or neck pain in all divers from the ELTHI Diving study Factor Mean Odds Ratio 95% CI for OR P Age (years) 1.01 0.99-1.03 0.36 Smoking (pack years) 1.01 1.00-1.02 0.01 Binge drinking 062 less than monthly 1.01 0.67-1.54 0.96 1-9 times per month 1.02 0.70-1.50 0.92 10-20 times per month 1.23 0.77-1.97 0.39 > 20 times per month 1.44 0.79-2.62 0.24 Body mass index 1.02 0.99-1.05 0.29 Work as a welder 1.29 0.97-1.70 0.08

Diving techniques (per 100 dives/days): SCUBA 1.00 0.98-1.02 0.90 Other air/nitrox dives 1.01 0.97-1.04 0.71 Surface decompression 0.98 0.95-1.02 0.31 Mixed gas bounce 0.97 0.83-1.42 0.74 Saturation days 1.02 0.99-1.04 0.29 Accidents Neurological DCI 0.06 never 1.00 once 1.68 1.10-2.55 0.02 > once 1.04 0.53-2.06 0.91 Pain only DCI 0.03 never 1.00 once 1.25 0.90-1.74 0.18 > once 1.59 1.27-2.25 0.08 Underwater explosion 0.55 never 1.00 once 1.11 0.68-1.82 0.68 > once 1.29 0.81-2.05 0.29 Loss of consciousness at pressure 0.16 never 1.00 once 1.16 0.73-1.85 0.54 > once 0.24 0.05-1.17 0.08

Contaminated gas 0.01 never 1.00 once 1.34 0.98-1.83 0.07 > once 1.68 1.20-2.34 0.002 Head injury 1.42 1.04-1.93 0.03

Chemical exposure Petrochemicals 0.01 never 1.00 a little 1.25 0.89-1.74 0.20 a lot 1.87 1.22-2.88 0.004 Hydrogen sulphide 0.14 never 1.00 a little 0.88 0.66-1.16 0.22 a lot 1.68 0.87-3.26 0.28 n=1411 with 129 cases excluded because of missing data. 441 subjects had cognitive complaint. Binge drinking was defined as the consumption of more than 8 units of alcohol in one session.

26 Table 27 Joint pain or muscle stiffness in all divers from the ELTHI Diving study Factor Mean Odds Ratio 95% CI for OR P Age (years) 1.01 0.99-1.02 0.55 Smoking (pack years) 1.01 1.00-1.02 0.10 Binge drinking 0.17 less than monthly 1.06 0.69-1.61 0.80 1-9 times per month 1.01 0.69-1.50 0.95 10-20 times per month 1.06 0.66-1.72 0.80 > 20 times per month 1.99 1.09-3.65 0.026 Body mass index 1.04 1.00-1.07 0.03 Work as a welder 1.27 0.95-1.69 0.11 Diving techniques (per 100 dives/days): SCUBA 0.99 0.97-1.01 0.17 Other air/nitrox dives 1.03 0.99-1.06 0.11 Surface decompression 1.01 0.97-1.04 0.77 Mixed gas bounce 1.19 1.02-1.39 0.03 Saturation days 0.99 0.96-1.02 0.35 Accidents Neurological DCI 0.60 never 1.00 once 1.10 0.71-1.71 0.67 > once 1.38 0.71-2.67 0.35 Pain only DCI 0.001 never 1.00 once 1.48 1.06-2.06 0.02 > once 1.90 1.34-2.68 <0.001 Underwater explosion 0.75 never 1.00 once 1.02 0.61-1.68 0.94 > once 1.20 0.75-1.92 0.45 Loss of consciousness at pressure 0.08 never 1.00 once 1.69 1.06-2.67 0.03 > once 1.23 0.37-4.16 0.74

Contaminated gas 0.004 never 1.00 once 1.40 1.02-1.92 0.04 > once 1.69 1.20-2.37 0.003 Head injury 1.39 1.01-1.90 0.04

Chemical exposure Petrochemicals 0.04 never 1.00 a little 1.12 0.80-1.58 0.50 a lot 1.66 1.07-2.57 0.02 Hydrogen sulphide 0.004 never 1.00 a little 0.66 0.49-0.89 0.005 a lot 1.45 0.75-2.82 0.27 n=1417 with 123 cases excluded because of missing data. 424 subjects had complained of moderate to severe joint pain or muscle stiffness. Binge drinking was defined as the consumption of more than 8 units of alcohol in one session.

27 Table 28 Impaired hearing in all divers from the ELTHI Diving study Factor Mean Odds Ratio 95% CI for OR P Age (years) 1.07 1.05-1.09 <0.001 Smoking (pack years) 1.01 1.0-1.01 0.34 Binge drinking 0.23 less than monthly 0.84 0.50-1.41 0.50 1-9 times per month 0.84 0.52-1.36 0.48 10-20 times per month 1.36 0.77-2.39 0.29 > 20 times per month 1.18 0.56-2.52 0.66 Work as a welder 1.28 0.90-1.81 0.17 Exposure to noise <0.001 never 1.00 a little 1.02 0.50-2.09 0.96 a lot 4.26 2.06-8.77 <0.001 Diving techniques (per 100 dives/days): SCUBA 1.01 .98-1.03 0.64 Other air/nitrox dives 0.99 0.95-1.03 0.65 Surface decompression 0.99 0.95-1.03 0.80 Mixed gas bounce 1.00 0.82-1.20 0.99 Saturation days 1.03 1.00-1.07 0.06 Accidents Neurological DCI 0.66 never 1.00 once 1.20 0.72-2.01 0.48 > once 1.31 0.60-2.87 0.51 Pain only DCI 0.59 never 1.00 once 0.92 0.59-1.43 0.71 > once 1.19 0.78-1.81 0.42 Underwater explosion 0.48 never 1.00 once 1.23 0.70-2.15 0.48 > once 0.78 0.44-1.38 0.39 Loss of consciousness at pressure 0.66 never 1.00 once 1.23 0.71-2.14 0.46 > once 1.44 0.42-4.99 0.57 Contaminated gas 0.04 never 1.00 once 1.12 0.75-1.69 0.58 > once 1.68 1.12-2.51 0.01 Head injury 1.33 0.90-1.96 0.15

Chemical exposure Petrochemicals 0.17 never 1.00 a little 0.64 0.41-1.02 0.06 a lot 0.64 0.36-1.14 0.13 Hydrogen sulphide 0.66 never 1.00 a little 0.84 0.59-1.22 0.36 a lot 0.92 0.39-2.17 0.85 n=1444 with 96 cases excluded because of missing data. 424 subjects had complained of moderate to severe impaired hearing.

28 Table 29 Relationships between cognitive complaint and musculoskeletal or hearing complaint Moderate-severe cognitive complaint No Yes Moderate – severe musculoskeletal complaint 409 (33.7%) 194 (72.7%) Moderate – severe hearing complaint 140 (11.5%) 90 (33.2%) Note: Musculoskeletal complaint is either ‘neck and back pain’ or ‘joint pain and muscle stiffness’. Percentages refer to the percent of the category with respect to cognitive complaint.

Musculoskeletal complaint was associated with cognitive complaint with 73% of subject with cognitive complaint also having musculoskeletal complaint. The same association was found with hearing complaint but to a lesser degree with 33% of subjects having cognitive complaint also reporting hearing impairment.

4.13.5 Relationship between saturation diving and work as a welder Since only saturation divers will have welded in a saturation hyperbaric welding habitat, the relationship between work as a welder and saturation diving was examined. Forty nine percent (174) of diving welders had used saturation diving techniques and therefore had the opportunity to work in a hyperbaric welding habitat. Of diver welders who had used saturation diving techniques 31% had cognitive complaint, a proportion similar to that found in diver welders as a group (28%) and not significantly greater than in diving welders who had not used saturation diving at the 5% level (23%) (p=0.095).

4.13.6 Prevalence of diagnosed medical conditions in divers in relation to work as a welder The prevalence of reports of arthritis, vibration white finger, drilling mud burn, dermatitis, peptic ulcer and migraine were higher in divers who had worked as a welder (Table 30). Table 30 Diagnosed medical conditions Non- Welder Welder Diver Chi2 test Diver

Arthritis 95 (8%) 41 (12%) p=0.04 Vibration white finger 23 (2%) 15 (4%) p=0.01 Bronchitis 44 (4%) 15 (4%) p=0.64 High blood pressure 120 (10%) 36 (10%) p=1.00 Asthma 54 (5%) 24 (7%) p=0.10 Dermatitis 100 (9%) 45 (13%) p=0.02 Drilling mud burn 197 (18%) 91 (20%) p=<0.001 Eczema or hay fever 184 (16%) 53 (15%) p=0.87 Depression 102 (9%) 38 (11%) p=0.21 Peptic ulcer 58 (5%) 33 (10%) p=0.003 Heart disease 23 (2%) 8 (2%) p=0.67 Migraine 69 (6%) 36 (10%) p=0.008 Cancer 22 (2%) 8 (2%) p=0.66 Diabetes 15 (1%) 4 (1%) * Stroke 4 (0.3%) 3 (0.9%) * Epilepsy 6 (0.5%) 1 (0.3%) * * it was not possible calculate a p-value due to the small number of people in one of the groups

29 5 DISCUSSION

The aim of this study was to determine whether the health impact of diving and welding is greater than welding alone.

Specific objectives were:

x To identify the effects of diving on the musculoskeletal and cognitive complaint in welders

x To examine any effect of welding on musculoskeletal and cognitive complaint in non-divers

x To analyse the ELTHI Diving study data for effects of factors other than diving in relation to cognitive and musculoskeletal complaint.

Diver welders were more likely to report a cognitive symptom than non-diver welders and have higher scores on the cognitive failure questionnaire. This confirms the observation of higher levels of cognitive complaint in diver welders from the ELTHI Diving study. Diver welders were more likely to report musculoskeletal complaint than welders who did not dive and there was a higher prevalence of arthritis in diver welders.

Although non-diving welders had more musculoskeletal complaint than subjects who neither welded nor dived, there was no significant difference between them in the prevalence of cognitive complaint.

Divers in the current study who worked as welders were more likely to complain of forgetfulness than divers in the ELTHI Diving study who did not weld. From the observations made in these two studies it might be deduced that the practice of welding significantly increases the risk of cognitive symptoms in divers but does not necessarily have a risk associated with it alone.

Welding in a hyperbaric habitat might be argued to put the diver more at risk from toxic fume inhalation than welding in a non-hyperbaric workplace. Conditions may be more cramped, ventilation might be poorer in such a space and high pressure might interact with the fume toxicity. In addition, deep saturation diving has been associated with liver dysfunction which may increase susceptibility to inhaled or ingested toxins (18) (19). Divers who weld in a habitat while diving almost always use saturation diving techniques. In the ELTHI Diving study, however, only 49% of welding divers would have had the opportunity to work in a hyperbaric saturation welding habitat and there was little evidence that cognitive complaint was commoner in saturation divers who had welded. In the present study, 60% of diver welders had welded in pressurised conditions. They had a higher welding fume exposure but there was no difference in the prevalence of cognitive complaint between divers who had welded at pressure and those who had not. So, there is no evidence from these studies to support the hypothesis that welding at pressure is the cause of increased cognitive complaint in diver welders.

There was little relationship identified in this study between either diving or welding exposure and the symptoms studied. It should be remembered, however, that establishing these relationships was not the objective of the study and the subject numbers are relatively small. It is more reasonable to look for relationships in divers within the larger ELTHI Diving study. This analysis showed that, although different diving techniques were associated with

30 symptom complaints after allowing for lifestyle factors, the complaints could equally be explained by other factors in the divers’ life such as decompression illness, exposure to contaminated gas, petrochemicals or hydrogen sulphide. In terms of factors unavoidably associated with diving and work under pressure, work as a welder and the practice of mixed gas bounce diving remained independently associated with cognitive complaint. Saturation diving and work as a welder were similarly associated with hearing complaint after full adjustment. If these work techniques are to be improved to avoid complaint then further, prospective work needs to be done to identify causative factors.

The other factors identified as being associated with complaint are all avoidable and not necessary parts of the divers job. Relationships between musculoskeletal complaint and other accidents such as head injury and pain only decompression illness might be expected. Similarly an association between cognitive complaint and neurological decompression illness might be expected. Associations between complaint and exposure to contaminated breathing gas and petrochemicals have not been made before but may be equally reasonable. The major gases contaminating breathing gas supplies in diving are hydrocarbons and carbon monoxide. Carbon monoxide has a known central nervous system effect and single incidences of exposure can lead to long term health effects even when consciousness is not lost (20). Carbon monoxide has also been associated with hearing loss (21) and muscle damage (22). Although low level exposure to organic solvents such as hydrocarbons have not convincingly been related to long term health effects (23), higher exposure to organic solvents are associated with effects (24) (25) (26) and it could be argued that petrochemical exposure might have a similar effect. A history of exposure to petrochemicals in divers, however, may also be an indicator of exposure to a generally contaminated worksite with a multiplicity of other toxins. Hydrogen sulphide was associated with cognitive complaint rather than petrochemicals in welders and again, like carbon monoxide, exposure to this gas can lead to long term health effects even in the absence of unconsciousness (27).

In the further analysis of the ELTHI Diving study data we found an association between reported petrochemical exposure and musculoskeletal symptoms. The literature does not indicate that petrochemical exposure is directly associated with musculoskeletal complaint. We found, however, that 73% of subjects with cognitive symptoms also had moderate to severe musculoskeletal symptoms and this may explain the association here.

Unsurprisingly, subjective report of exposure to noise was associated with reported impaired hearing. Less expected was an association with reported contaminated gas exposure. Carbon monoxide, however, can cause hearing loss (21) as can exposure to solvents (28) (29). Combined exposure to noise and solvents is more likely to cause hearing loss than either alone (30) and this effect has been seen in petroleum refinery workers who were exposed to noise and solvents at levels generally taken to be safe (31).

The association between symptoms and the possibly toxic exposures of welding, contaminated gas, petrochemicals and hydrogen sulphide in divers leads to consideration of other possible effects. Drilling mud burns were reported by 288 (19%) of divers in the ELTHI Diving study and were commoner in welder divers. This condition is a contact dermatitis with the hydrocarbons in the mud causing defatting of the skin and can be triggered by sensitisation to polyamines, the dried mud and low boiling point petrochemicals (32) (33). A report of drilling mud burn by a diver indicates skin contact with the substance in question most probably in the water. If skin contact with ambient water occurs during diving then dermal uptake of petrochemicals and other agents is a distinct possibility.

The ELTHI Diving Study indicated that a reported diagnosis of dermatitis was commoner in diver welders than in non-welding divers and in the present study it was commoner in diver

31 welders than in non-diving welders. Welding has known associations with dermatitis due to radiation and welding fume exposure. Divers have the additional risk factor of prolonged water contact which softens the skin and makes it more susceptible to irritation.

In a longitudinal study welders were found to lose lung function more rapidly with age when compared to non-diving controls (34). Although this study allowed for smoking habit, no allowance was made for welding fume, contaminated breathing gas or exposure to other possible inhaled toxins. The question arises as to whether the effect identified was due to diving at all or to toxic exposures. Petrochemcals and hydrogen sulphide are mucosal irritants and hydrogen sulphide may have long term effects on lung function (35). Aerosolised hydrocarbon contamination of divers’ breathing gas can result in lipid pneumonitis with potentially severe long term pulmonary effect (36). Welding fume has recognised effects on the lung (2) related to inflammation (37) and pneumonia. In the present study and in the ELTHI Diving study, however, there was no indication of pulmonary abnormality in either divers or diver welders.

A series of studies have identified an effect of deep saturation diving on liver function (18) (19) (38) manifest as an increase in transaminase enzymes and glycoproteins. There is, however, no mention of atmospheric quality control in the saturation system used during these trials and contamination may be a possible explanation of the effect.

We have found that welding is associated with a greater degree of hearing complaint, and that diver welders are more likely to have cognitive and musculoskeletal symptoms in addition. In spite of these observations there were no convincing correlations between these complaints and measures of welding or diving experience. Further analysis of data from the ELTHI Diving study indicate that other factors, notably those associated with toxic exposures, are associated with the complaints studied and these now should be investigated for a causal link. An important observation is that if toxic exposures are the cause of these complaints, occupational hygiene measures will be a powerful intervention.

32 6 APPENDICES

APPENDIX 1: QUESTIONNAIRE

Please answer all questions by placing a cross in the box or a number, as appropriate.

All the information you provide will be STRICTLY CONFIDENTAL

SECTION 1 Lifestyle

Date of birth (day/month/year)? …….…. / ….…….. / 19…….… Male Female

What is your current marital status? Never married Married Divorced or widowed

What is the highest educational qualification you have gained?

None O Level, Standard Grades, A Levels, Scottish HNC/HND University School Certificates Highers degree

Have you smoked more than 100 cigarettes IN TOTAL in your life? No Yes if yes, please complete the following: Current Smokers Ex-Smokers in what year did you stop smoking? ……….. How many years in total have you smoked? ….……..… How many years in total did you smoke? .……..………….. How many cigarettes do you smoke per day? ………… How many cigarettes did you smoke per day? ……………

On average, how many UNITS of alcohol do you drink per WEEK? ……………… units per week (Note: 1 UNIT is equivalent to a ½ pint of normal strength beer, lager or cider OR 1 small glass of wine OR 1 shot of spirit)

During the last 12 months, how often have you drunk 8 units or more on any one occasion? never less than once a month 1-9 times a month 10-20 times a month more than 20 times a month

How much do you weigh? ..……… kg or ……..st …..…lb How tall are you? ………m or ……ft …...in

Have you ever done recreational underwater diving? No Yes if yes, approximately how many recreational dives have you done? ……………….

SECTION 2 Occupational History

Which of the following best describes your CURRENT work status? Employed Self-employed Self-employed Unemployed Not working (on sickness benefits) Retired & working & between jobs or retired through ill health

What is your current or most recent job description? …………………………………………………………………..

Have you ever had a lost time (more than 3 days) accident or injury at work? No Yes

if yes, how many lost time (more than 3 days) accidents have you had in total? …………….

33 In the course of your career, have you been exposed to any of the following? No, not at all Yes, a little Yes, a lot Petrochemicals, solvents or non-lead based paints Lead based paints (including paint removal activities) Drilling mud Asbestos Hydrogen sulphide Rock or coal dust Noise (loud enough to make conversation difficult)

2.1 Have you ever worked as a WELDER?

Have you ever worked as a welder? No Yes if no please go to section 2.2 In what year did you START welding at work? ……………… In what year was your MOST RECENT welding job? ……..………. How many years in total have you carried out welding work? ………….. years Have you permanently stopped welding at work? No Yes

Name the 3 metals you used most commonly when welding? 1……..………... 2…..………..…. 3………..………

What percent of your welding was done in the following work areas? tick this box if ‘never’ 1. Outdoors? …..……% 2. Indoors in a well ventilated area? …..……% 3. In a small, poorly ventilated area? …..……% 4. In a pressurised environment (i.e. while diving)? …..……%

Over what percent of your career have you used the following personal protective breathing equipment while welding? tick this box if ‘never’ 1. Simple dust mask ……..…% 2. Filter respirator ……..…% 3. Atmosphere supply respirator (e.g. Aga mask) ……..…% 4. Other, please name: …………………………….. ……..…%

At work, have you used the following welding techniques?

if yes: how many years have on average, how many days per No Yes you used this technique? year have you used it?

Manual Metal Arc (MMA) …………………years ……………… days per year

Metal Inert Gas (MIG) …………………years ……………… days per year

Tungsten Inert Gas (TIG) …………………years ……………… days per year

Oxyfuel (e.g. oxyacetylene) …………………years ……………… days per year

Flux Cored Arc (FCW) …………………years ……………… days per year

Flame or Arc Metal cutting …………………years ……………… days per year

Other, please name …………………… …………………years ……………… days per year

While welding have you ever suffered from the following? NoYes if yes, how many times Metal fume fever ………………… Major electric shock …………………

34 2.2 Have you ever worked OFFSHORE (topside)?

Have you ever worked topside (not as a diver) offshore? No Yes if no please go to section 2.3

In what year did you START working offshore? ………………

In what year did you MOST RECENTLY work offshore? ……..…..…

How many years in total have you worked offshore? ………….. years

Have you permanently stopped working offshore? No Yes

2.3 Have you ever carried out work as a professional DIVER (include military diving)?

Have you ever carried out work as a professional diver? No Yes if no please go to section 3

In what year did you START professional diving? …….…….

In what year was your MOST RECENT professional dive? …………..…

How many years in total have you worked as a diver? ………….. years

Have you permanently stopped working as a diver? No Yes

Which of the following diving techniques have you used at work (include your training)? NoYes if yes, how many? SCUBA diving ………… dives Surface supplied (air or nitrox) diving ………… dives Surface oxygen decompression diving ………… dives Mixed gas bounce diving ………… dives Saturation diving ………… days in saturation

Have you suffered from any of the following: while WORKING while RECREATIONAL DIVING No Yes how many times? No Yes how many times? Pain only decompression illness ……………. ……………. Neurological decompression illness ……………. ……………. Cerebral gas embolism ……………. ……………. Drilling mud burn ……………. …………….

SECTION 3 General Health

Do you regularly suffer from any of the following? not at all slightly moderately severely

Joint pain or muscle stiffness

Back pain or neck pain Breathlessness Cough or wheeze Abdominal pain, diarrhoea, constipation or nausea Muscle weakness or tremor Unsteadiness when walking, dizziness or poor balance Forgetfulness or loss of concentration Poor eye sight (not corrected by spectacles) Impaired hearing Skin rash or itch

35 Are you currently receiving any medical treatment or medication? No Yes

if yes please provide details:…………………………………………………………………………………….

Have you ever been diagnosed with any of the following conditions? No Yes No Yes Asthma Ulcer (stomach or peptic) Chronic bronchitis or other lung diseases Heart attack or disease Arthritis Dermatitis Depression or anxiety Head injury Stroke Eczema or hayfever High blood pressure Epilepsy Cancer (including leukaemia) Migraine Diabetes Vibration white finger

In general, would you say your health is: excellent very good good fair poor

The following questions are about activities you might do during a typical day. Does your health limit you in these activities? If so, how much? Yes, a lot Yes, a little No, not at all

a. Moderate activities such as moving a table, pushing a vacuum cleaner, bowling, playing golf b. Climbing several flights of stairs

During the past 4 weeks, have you had any of the following problems with your work or other regular daily activities as a result of your physical health? Yes No a. Accomplished less than you would like

b. Were limited in the kind of work or other activities

During the past 4 weeks, have you had any of the following problems with your work or other regular daily activities as a result of any emotional problems (such as feeling depressed or anxious)? Yes No a. Accomplished less than you would like

b. Didn’t do work or other activities as carefully as usual

During the past 4 weeks, how much did pain interfere with your normal work (including both work outside the home and housework)? not at all a little bit moderately quite a bit extremely

These questions are about how you feel and how things have been with you during the past month. For each question, please give the one answer that comes closest to the way you have been feeling.

How much during the past month: all the most of a good bit some of a little of none of time the time of the time the time the time the time a. Have you felt calm and peaceful? b. Did you have a lot of energy? c. Have you felt downhearted and depressed? d. Has your health limited your social activities (like visiting friends or close relatives)?

36 The following questions are about minor mistakes which everyone makes from time to time, but some of which happen more often than others. We want to know how often these things have happened to you in the last 6 months. Please circle the appropriate number.

Very Quite Very often often Occasionally rarely Never

Do you read something and find you haven’t been thinking 43 2 10 about it and must read it again?

Do you find you forget why you went from one part of the 43 2 10 house to the other? Do you fail to notice signposts on the road? 4 3 2 1 0 Do you find you confuse right and left when giving 43 2 10 directions? Do you bump into people? 4 3 2 1 0

Do you find you forget whether you’ve turned off a light or 43 2 10 a fire or locked the door?

Do you fail to listen to people’s names when you are 43 2 10 meeting them?

Do you say something and realise afterwards that it might 43 2 10 be taken as insulting?

Do you fail to hear people speaking to you when you are 43 2 10 doing something else? Do you lose your temper and regret it? 4 3 2 1 0 Do you leave important letters unanswered for days? 4 3 2 1 0

Do you find you forget which way to turn on a road you 43 2 10 know well but rarely use?

Do you fail to see what you want in a supermarket 43 2 10 (although it’s there)?

Do you find yourself suddenly wondering whether you’ve 43 2 10 used a word correctly? Do you have trouble making up your mind? 4 3 2 1 0 Do you find you forget appointments? 4 3 2 1 0

Do you forget where you have put something like a 43 2 10 newspaper or a book? Do you find you accidentally throw away the thing you want and keep what you meant to throw away – as in the 43 2 10 example of throwing away the matchbox and putting the used match in your pocket? Do you daydream when you ought to be listening to 43 2 10 something? Do you find you forget people’s names? 4 3 2 1 0

Do you start doing one thing at home and get distracted 43 2 10 into doing something else (unintentionally)?

Do you find you can’t quite remember something although 43 2 10 it’s ‘on the tip of your tongue’?

Do you find you forget what you came to the shops to 43 2 10 buy? Do you drop things? 4 3 2 1 0 Do you find you can’t think of anything to say? 4 3 2 1 0

37 APPENDIX 2: CALCULATION OF WELDING FUME EXPOSURE

Welding fume exposure (mg.m-3 days) was estimated using an equation, devised by Dr S. Semple, taking into account the welding environment, the use of respiratory protective equipment, different welding techniques and the time spent welding.

Crude base estimates for the welding fume exposure (mg.m-3) from each of the different welding types were generated using mean data from van der Wal (1990) (39). These figures were then modified depending on the work environment, degree of ventilation and the use of respiratory protective equipment. Work by Cherrie (1999) has shown the impact of general ventilation and room size on personal exposure to generated fume (40). For welding in a pressurised environment data were used from a hyperbaric welding trial where a total of 20 hours of welding was monitored. The time weighted average for total inhalable dust (PM10) was 2.58 mg.m-3 (see section A2.1). This is about twice the value compared to atmospheric pressure.

The following ‘semi-subjective’ factors were used in the equation to estimate welding fume exposure levels (Table 31).

Table 31 Weighting factors used in estimating welding fume exposure Factor Welding environment Outdoors 0.1 Indoors in a well ventilated area 1 In a small, poorly ventilated area 10 In a pressurised environment (while diving) 2

Respiratory protective equipment Simple dust mask 1 Filter respirator 0.1 Atmosphere supply respirator 0.01

Welding techniques mg.m-3 Tungsten inert gas (TIG) 1 Metal inert gas (MIG) 4 Flux cored arc (FCA) 4 Manual metal arc (MMA) 8 Oxyfuel (Oxy) 8

Welding fume exposure equation: (((tech*(0.1*out)) + (tech*(1*in)) + (tech*(10*small)) + (tech*(2*dive))) *yrs*days * (1*dust)) + (((tech*(0.1*out)) + (tech*(1*in)) + (tech*(10*small)) + (tech*(2*dive))) *yrs*days * (0.1*filter)) + (((tech*(0.1*out)) + (tech*(1*in)) + (tech*(10*small)) + (tech*(2*dive))) *yrs*days * (0.01*atmos))

Key: tech exposure factor for each welding technique out proportion of welding done outdoors in proportion of welding done indoors in a well ventilated area small proportion of welding done in a small, poorly ventilated area dust proportion of welding done using a dust mask filter proportion of welding done using filter respirator atmos proportion of welding done using atmosphere supply respirator yrs number of years having used the welding technique days number of days per year welded using the welding techniques

38 The above equation was used for each of the different welding techniques then summed to give a total welding fume exposure.

For example: A diver welder reported the following welding experience:

For 13 years he used MMA (22 days per year), MIG (5 days per year), Oxyfuel (1 day per year), FAM (30 days per year)

70% of his welding was done outdoors, 10% indoors in a well-ventilated area and 20% in a pressurised environment. While welding he used a filter respirator for 10% of the time and atmosphere supplied respirator for 20%.

MMA = (((8*(0.1*0.7)) + (8*(1*0.1)) + (8*(10*0)) + (8*(2*0.2)))*13*22 *(1*(1-(0.1+0.2)))) + (((8*(0.1*0.7)) + (8*(1*0.1)) + (8*(10*0)) + (8*(2*0.2))) *13*22 * (0.1*0.1)) + (((8*(0.1*0.7)) + (8*(1*0.1)) + (8*(10*0)) + (8*(2*0.2))) *13*22 * (0.01*0.2)) = 929 mg.m-3 days

MIG = (((4*(0.1*0.7)) + (4*(1*0.1)) + (4*(10*0)) + (4*(2*0.2))) *13*5 * (1*(1-(0.1+0.2)))) + (((4*(0.1*0.7)) + (4*(1*0.1)) + (4*(10*0)) + (4*(2*0.2))) *13*5 * (0.1*0.1)) + (((4*(0.1*0.7)) + (4*(1*0.1)) + (4*(10*0)) + (4*(2*0.2))) *13*5 * (0.01*0.2)) = 106 mg.m-3 days

Oxyfuel =(((8*(0.1*0.7)) + (8*(1*0.1)) +(8*(10*0)) + (8*(2*0.2))) *13*1* (1*(1-(0.1+0.2)))) + (((8*(0.1*0.7)) + (8*(1*0.1)) + (8*(10*0)) + (8*(2*0.2))) *13*1 * (0.1*0.1)) + (((8*(0.1*0.7)) + (8*(1*0.1)) + (8*(10*0)) + (8*(2*0.2))) *13*1 * (0.01*0.2)) = 42 mg.m-3 days

FAM = (((8*(0.1*0.7)) + (8*(1*0.1)) + (8*(10*0)) + (8*(2*0.2))) *13*30 * (1*(1-(0.1+0.2))) + (((8*(0.1*0.7)) + (8*(1*0.1)) + (8*(10*0)) + (8*(2*0.2))) *13*30 * (0.1*0.1)) + (((8*(0.1*0.7)) + (8*(1*0.1)) + (8*(10*0)) + (8*(2*0.2))) *13*30 * (0.01*0.2)) = 1266 mg.m-3 days

TOTAL welding fume exposure = MMA + MIG + Oxyfuel + FAM = 2343 mg.m-3 days

This is equivalent to being exposed for 2 years at a concentration of 5mg.m-3 for 240 days a year.

39 A2.1 Measurement of dust levels during hyperbaric welding In order to establish an exposure measure for welding fume in a hyperbaric habitiat, dust levels were studied during a hyperbaric weld conducted at 10 atmospheres absolute in a helium oxygen environment.

-3 PM10 (particles of 10 Pm or less), were measured in mg.m using the TSI Incorporated DustTrak Aerosol Monitor, during a hyperbaric welding trial at 40m. A gas sample flow was emitted from the work chamber during welding operations using a penetration in the chamber hull plumbed with 0.75 inch diameter tungum pipe and controlled using a quarter turn valve outside the chamber. The gas inlet for the sample line was approximately 1 metre away from the main weld piece and was at waist level. The length of pipe descended on the outside of the chamber and allowed gas sampling from it into the analysers. The analyser sample tubes were inserted into the open end of the tungum pipe and the quarter turn valve adjusted to give an obvious free flow from the end of the pipe. Sampling was therefore performed at one atmosphere pressure. All readings were multiplied by chamber pressure in atmospheres absolute in order to allow for the dilution of sample decompression.

Four welding sessions were monitored over a total period of 20 hours. Time weighted exposures were calculated by dividing the sum of all observations by the number of observations. Observations were taken regularly at 5 minute intervals. The mean time -3 weighted exposure for PM10 dust over this period was 2.58 mg m (Table 32) and a typical analysis trace is shown in Figure 3.

Table 32 Time weighted PM10 exposure during hyperbaric welding operations Duration of Time Weighted Exposure session (hrs) mg m-3 22 .0 61 .86 63 .92 62 .53 Total 20 hrs 20 hr time weighted average 2.58

Welding Dust 25/05/04 40 m

10

8

6

4

2 mg per cubic metre 0 10 11 12 13 14 15 16 17 Time (24 hour)

PM 2.5 PM 10

Figure 3 PM10 and PM2.5 levels during a hyperbaric welding session

40 APPENDIX 3: A DESCRIPTION OF THE METALS WELDED UNDER THE CATEGORY OF ‘EXOTICS’

Table 33 shows the different metals classified as ‘exotic’ metals. These metals were reported as commonly welded by the both diver welders and non-divers welders.

Table 33 Grouping of metals classified as ‘exotic’ metals Grouping Metals included in this category Number of welders using these metals

Aluminium Aluminium 53

Copper Copper, Cunifer, Cuni, corten 32

Zinc Galvanised steel, zinc 15

Brass Brass, bronze 22

Chromium Chrome, chrome moly, duplex, Stellite 22 A, inconel Miscellaneous Alloys, tungsten, tin, titanium, nickel, 17 lead, magnesium

41 7 REFERENCES

1 Newhouse ML, Oakes D, Woolley AJ. Mortality of welders and other craftsmen at a shipyard in NE England. British Journal of Industrial Medicine 1985; 42:406-410.

2 Antonini JM. Health effects of welding. Critical Reviews in Toxicology 2003; 33(1):61-103.

3 Korczynski RE. Occupational health concerns in the welding industry. Applied Occupational and Environmental Hygiene 2000; 15(12):936-945.

4 Lucchini R, Apostoli P, Perrone C, Placidi D, Albini E, Migliorati P et al. Long term exposure to 'low levels' of manganese oxides and neurofunctional changes in ferroalloy workers. Neurotoxicology 1999; 20(2-3):287-297.

5 Riihimaki V, Hanninen H, Akila R, Kovala T, Kuosma E, Paakkulainen H et al. Body burden of aluminum in relation to central nervous system function among metal inert- gas welders. Scandinavian Journal of Work, Environment & Health 2000; 26(2):118- 130.

6 Barrington WW, Angle CR, Willcockson NK, Padula MA, Korn T. Autonomic function in manganese alloy workers. Environmental Research 1998; 78(1):50-58.

7 Sjogren B, Gustavsson P, Hogstedt C. Neuropsychiatric symptoms among welders exposed to neurotoxic metals. British Journal of Industrial Medicine 1990; 47:704-707.

8 Bowler RM, Gysens S, Diamond E, Booty A, Hartney C, Roels HA. Neuropsychological sequelae of exposure to welding fumes in a group of occupationally exposed men. International Journal of Hygiene and Environmental Health 2003; 206(6):517-529.

9 Torner M, Zetterberg C, Anden U, Hansson T, Lindell V. Workload and musculoskeletal problems: a comparison between welders and office clerks (with reference also to fishermen). Ergonomics 1991; 34(9):1179-1196.

10 Burdorf A, Naaktgeboren B, Post W. Prognostic factors for musculoskeletal sickness absence and return to work among welders and metal workers. Occupational and Environmental Medicine 1998; 55(7):490-495.

11 Macdiarmid JI, Ross JAS, Taylor CL, Watt SJ, Adie W, Osman LM et al. Co-ordinated investigation into the possible long term health effects of diving at work: The ELTHI Diving Study. ed. Norwich: HSE, 2004.

12 Syversen T, Jenssen J. High hydrostatic pressure potentiation of the toxic effects of chromate in cell culture. Undersea Biomedical Research 1987; 14(1):11-19.

13 Ware JE, Kosinski M, Keller SD. SF-12: How to score the SF-12 physical and mental summary scores. Third ed. Lincoln, RI: Quality Metric Incorportation, 1998.

42 14 Broadbent DE, Cooper PF, FitzGerald P, Parkes KR. The Cognitive Failures Questionnaire (CFQ) and its correlates. British Journal of Clinical Psychology 1982; 21:1-16.

15 Unwin C, Blatchley N, Coker W, Ferry S, Hotopf M, Hull L et al. Health of UK servicemen who served in Persian Gulf War. The Lancet 1999; 353:169-178.

16 Sarter M, Bruno JP. Developmental origins of the age-related decline in cortical cholinergic function and associated cognitive abilities. Neurobiology of Aging 2004; 25:1127-1139.

17 Jenkinson C, Layte R, Wright L, Coulter A. The U.K. SF-36: an analysis and interpretation manual. Oxford: Health Services Research Unit, 1996.

18 Ikeda M, Nakabayashi K, Shinkai M, Hara Y, Kizaki T, Oh-ishi S et al. Supplementation of antioxidants prevents oxidative stress during a deep saturation dive. Tohoku Journal of Experimental Medicine 2004; 203:353-357.

19 Goldinger JM, Nakayama H, Takeuchi H, Hong SK. Seadragon VI: a 7-day dry saturation dive at 31 ATA. VIII. Plasma enzyme profiles. Undersea Biomedical Research 1987; 14(5):455-459.

20 Thom SR, Taber RL, Mendiguren II, Clark JM, Hardy KR, Fisher AB. Delayed neuropsychologic sequelae after carbon monoxide poisoning: prevention by treatment with hyperbaric oxygen. Annals of Emergency Medicine 1995; 25(4):474-480.

21 Shahbaz Hassan M, Ray J, Wilson F. Carbon monoxide poisoning and sensorineural hearing loss. The Journal of Laryngology & Otology 2003; 117:134-137.

22 Wolff E. Carbon monoxide poisoning with severe myonecrosis and acute renal failure. American Journal of Emergency Medicine 1994; 12(3):347-349.

23 Gamble JF. Low-level hydrocarbon solvent exposure and neurobehavioural effects. Occupational Medicine 2000; 50(2):81-102.

24 Nilson LN, Backman L, Sallsten G, Hagberg S, Barregard L. Dose-related cognitive deficits among floor layers with previous heavy exposure to solvents. Archives of Environmental Health 2003; 58(4):208-217.

25 Morrow LA, Stein L, Bagovich GR, Condray R, Scott A. Neuropsychological assessment, depression, and past exposure to organic solvents. Applied Neuropsychology 2001; 8(2):65-73.

26 Bockelmann I, Pfister EA, Peters B, Duchstein S. Psychological effects of occupational exposure to organic solvent mixtures on printers. Disability & Rehabilitation 2004; 26(13):798-807.

27 Hirsch AR. Hydrogen sulfide exposure without loss of consciousness: chronic effects in four cases. Toxicology & Industrial Health 2002; 18(2):51-61.

28 Sulkowski WJ, Kowalska S, Matyja W, Guzek W, Wesolowski W, Szymczak W et al. Effects of occupational exposure to a mixture of solvents on the inner ear: a field study. International Journal of Occupational Medicine & Environmental Health 2002; 15(3):247-256.

43 29 Sliwinska-Kowalska M, Zamyslowska-Szmytke E, Szymczak W, Kotylo P, Fiszer M, Dudarewicz A et al. Hearing loss among workers exposed to moderate concentrations of solvents. Scandinavian Journal of Work, Environment & Health 2001; 27(5):335- 342.

30 Sliwinska-Kowalska M, Zamyslowska-Szmytke E, Szymczak W, Kotylo P, Fiszer M, Wesolowski W et al. Effects of coexposure to noise and mixture of organic solvents on hearing in dockyard workers. Journal of Occupational & Environmental Medicine 2004; 46(1):30-38.

31 Morata TC, Engel T, Durao A, Costa TR, Krieg EF, Dunn DE et al. Hearing loss from combined exposures among petroleum refinery workers. Scandinavian Audiology 1997; 26(3):141-149.

32 Ormerod AD, Wakeel RA, Mann TA, Main RA, Aldridge RD. Polyamine sensitization in offshore workers handling drilling muds. Contact Dermatitis 1989; 21(5):326-329.

33 Ormerod AD, Dwyer CM, Goodfield MJ. Novel causes of contact dermatitis from offshore oil-based drilling muds. Contact Dermatitis 1998; 39(5):262-263.

34 Skogstad M, Thorsen E, Haldorsen T, Kjuus H. Lung function over six years among professional divers. Occupational and Environmental Medicine 2002; 59:629-633.

35 Buick JB, Lowry RC, Magee TR. Is a reduction in residual volume a sub-clinical manifestation of hydrogen sulfide intoxication? American Journal of Industrial Medicine 2000; 37(3):296-299.

36 Kizer KW, Golden JA. Lipoid pneumonitis in a commercial abalone diver. Undersea Biomedical Research 1987; 14(6):545-552.

37 Kim JY, Chen JC, Boyce PD, Christiani DC. Exposure to welding fumes is associated with acute systemic inflammatory responses. Occupational and Environmental Medicine 2005; 62:157-163.

38 Doran GR, Chaudry L, Brubakk AO, Garrard MP. Hyperbaric liver dysfunction in saturation divers . Undersea Biomedical Research 1985; 12(2):151-164.

39 van der Wal JF. Exposure of welders to fumes and gases in Dutch industries: summary of results. Annal of Occupational Hygiene 1990; 34(1):45-54.

40 Cherrie JW. The effect of room size and general ventilation on the relationship between near and far-field concentrations. Applied Occupational and Environmental Hygiene 1999; 14(8):539-546.

Printed and published by the Health and Safety Executive C1.10 10/05 ISBN 0-7176-6169-5

RR 390 £10.00 9 78071 7 661 695