Trends in exposure of veterinarians to physical and chemical hazards and use of

Trends in exposure of veterinarians to physical and chemical hazards and use of protection practices Lin Fritschi 1 Adeleh Shirangi 2 Ian D Robertson 3 Lesley M Day 4 1. Laboratory for Cancer Medicine, Western Australian Institute for Medical Research, Perth, Australia 2. School of Population Health, University of Western Australia, Perth, Australia 3.School of Veterinary and Biomedical Sciences, Murdoch University, Perth, Australia 4. Accident Research Centre, Monash University, Melbourne, Australia Corresponding author Associate Professor Lin Fritschi Western Australian Institute for Medical Research Ground Floor, B block, Hospital Avenue, Nedlands 6009, Australia email: fritschi@waimr.uwa.edu.au Phone: +61-8-9346 3473 1

ABSTRACT Objectives: To determine whether exposure to physical and chemical occupational hazards and use of protective practices has changed in recent veterinary graduates, and to describe trends in exposure to occupational hazards and use of protective practices over time Methods: This paper reports on a retrospective cohort study of veterinarians who graduated from any of the four Australian veterinary schools between 1960 and 2000 and were currently in clinical practice. A self-completed postal questionnaire was used to collect personal details, professional history since graduation, and details of occupational hazards and protective practices used. The prevalence of occupational hazards and use of protective practices was examined by decade of graduation adjusting for gender, type of practice and number of hours worked. Results: After adjusting for other factors, recent graduates tended to take more X-rays than early graduates graduates since 1990 were 2.59 times more likely to take more than 7 X- rays a week compared with graduates before 1970. Recent graduates were also more likely to personally restrain animals during X-rays but were more likely to use X-ray protection. Of those who undertook surgery, recent graduates were more likely to use anesthetic waste gas scavengers. Over time, veterinary jobs have become more likely to involve longer hours of surgery, but shorter periods of driving. The use of scavengers for waste anaesthetic gas has increased markedly over time from 3.8% of jobs commencing in the 1960s to over 70% for jobs commencing since 1997. Conclusions: This survey is, to our knowledge, the first to examine trends in the occupational health and safety practices of veterinarians. We have shown that occupational health issues are still important in veterinary practice, with most veterinarians exposed to a number of physical and chemical hazards and many using inadequate protection. 2

KEYWORDS: radiation, occupation, veterinary medicine, anesthetics 3

INTRODUCTION Veterinarians are exposed to a number of potential hazards in their professional lives including ionizing radiation, anaesthetics, motor vehicle driving and pesticide use (Jeyaretnam, 2000b). There are established behaviours or equipment which can be used to mitigate the effect of these exposures, such as waste anaesthetic gas scavengers during surgery (Torda, 1978) and wearing lead aprons and avoidance of use of staff to restrain animals when taking X-rays. Studies in the 1980s and 1990s found that many veterinarians were occupationally exposed to hazards and use of protective practices was not widespread. For example, studies in Australia and the US found that 30% of veterinary practices did not have a waste anaesthetic gas scavenger (Jeyaretnam, 2000a; Wiggins, 1989) and 76% of American female veterinarians reported physically restraining animals for X-ray procedures (Wiggins, 1989). With more attention being paid in veterinary schools to occupational health and safety, we were interested to know whether this effort has paid off in terms of better safety behaviour over time and in the more recent graduates. A survey of Australian veterinarians was conducted to determine whether exposure to physical and chemical occupational hazards and use of protective practices has changed in recent veterinary graduates, and to describe trends in exposure to occupational hazards and the use of protective practices over time. METHODS This cohort study of Australian veterinarians has been described previously (Fritschi, 2006; Shirangi, 2007). All veterinarians who graduated from any of the four Australian veterinary schools between 1960 and 2000 were eligible for the study (n = 7928). Lists of graduates were obtained from the alumni organizations of each university. Current addresses of veterinarians were obtained from State Veterinary Registration Board lists and national and overseas organizations. Of the nearly eight thousand veterinarians, 69 (0.1%) were known to 4

be deceased, 501 (6%) were known to be living overseas, and no current address was found for 1612 (22%). Invitations to participate in the study, including an information sheet, a consent form, and the questionnaire, were sent to 5746 subjects for whom a current Australian address could be found. Subjects who did not respond to the first mail out were sent a postcard reminding them to return their questionnaire. The questionnaire included personal information (date of birth, gender, university and year of graduation) and details of veterinary work undertaken since graduation. This professional history included details of each job held for more than 6 months including start date, end date, job type (clinical, government, academic, drug company, other), and type of practice. For this analysis, we only included those veterinarians who stated that they worked in large, mixed, small or specialist clinical practice (n = 2562). This excluded those who were primarily not clinicians and were employed by government, research organizations or drug companies. Since there were few specialist practitioners and they were almost all working with large animals, they were combined with the large animal practitioners for the analysis. Prevalence of current occupational hazards: For the current job we asked the number of hours worked, the approximate number of X-ray films taken per week, how many hours of surgery and driving were done, how often pesticides were used, and how many kilometres the veterinarian drove to see patients each week. Contingency tables and the chi-squared statistic were used to examine the prevalence of occupational hazards in the current job by decade of graduation. The mean for each of the continuous variables (hours worked, X-ray films taken, hours surgery, hours driving) was calculated for those who undertook these practices and differences between decade of graduation were assessed using one-way ANOVA analyses. Logistic regression was taken to predict the odds ratio for doing more than the mean number 5

for each of the continuous variables adjusting for gender, type of practice and number of hours worked. Protective practices: For the current job, we asked how frequently the veterinarian restrained animals during X-rays, whether an anaesthetic gas scavenger was used during surgery, the frequency of use of various forms of radiation protection, how often inadequate restraint of animals was used, and whether the veterinarian always used a seatbelt when driving a car. For these analyses we examined the proportion of veterinarians using the protective practice, of those exposed to the hazard. This consisted of use of anaesthetic scavenger in those doing surgery; use of radiation protection in those taking X-rays; and use of seatbelt in those driving to see patients. Contingency tables and the chi-squared statistic were used to examine the prevalence of appropriate protection use by decade of graduation. Logistic regression was undertaken to examine the likelihood of using the protective practice by graduation decade adjusting for gender and practice type. The four protective practices examined were: use of anesthetic waste gas scavenger in those doing surgery; use of either apron or screen in those doing X-rays; never or almost never working with inadequate restraint; and always wearing seatbelt in those driving for work. Trends over time: We also obtained a full professional history from each respondent. This included years starting and finishing in that practice, job type and practice type. For each job we asked the average number of hours worked, average number of X-rays taken per week, whether they personally restrained animals during X-rays, average number of hours in surgery per week, whether they used a waste anaesthetic gas scavenger, average number of hours driving a week, and how often pesticides were usually used. We classified each job according to the year the veterinarian started in that job. Jobs were then grouped by year starting into five subgroups: 1960-9, 1970-9, 1980-9, 1990-6, 1997-2002. One-way ANOVA analyses were used to compare means of the practice characteristics 6

for each of these five sub-groups. To adjust for the changing demographics of veterinary practice we performed logistic regression to predict the odds ratio for doing more than the mean number for each of the continuous variables adjusting for gender, type of practice, decade of graduation and number of hours worked. RESULTS Completed questionnaires were received from 2800 veterinarians, 35% of the whole cohort and 48.7% of those to whom questionnaires were sent. Of these, 2562 worked in clinical positions and were included in this analysis. There have been a number of major changes in the makeup of veterinary practice over time. Female veterinarians went from a minority (8.1% of those graduating in the 1960s) to a majority (65.6% of those graduating since 1990). Practice types have also changed (Figure 1). Veterinarians graduating in the 1960s were equally as likely to currently be working in large, mixed or small animal practice. Veterinarians graduating since 1990 were most likely to be currently working in small animal practice (57.6%) with only a minority working in large animal practices (9.3%). On average, veterinarians worked about 44 hours a week with a range of 1 to over 100 hours per week. Earlier graduates tended to work more hours than recent graduates (Table 1). More recent graduates were more likely to take X-rays and undertake surgery but the number of hours in surgery was less in more recent graduates. Earlier graduates were most likely to drive for work, spend more hours driving for work (an average of nearly 9 hours a week) and drive many more kilometres per week. Pesticides were used more often by recent graduates. After adjusting for gender, type of practice and hours of work, there was a clear trend towards veterinarians who graduated more recently being more likely to take more than the mean number of X-rays per week (Table 2). All graduation year groups were similar in their likelihood of undertaking many hours of surgery or driving. The type of practice was the major influence on exposure to occupational hazards as large animal veterinarians were most 7

likely to spend more than 7 hours driving a week, while small animal veterinarians were most likely to take more than 7 X-rays per week and spend longer in surgery. Overall, the use of protective practices was higher in recent graduates compared to earlier graduates (Table 3). Of those graduates after 1990 who took X-rays, nearly 95% used either a lead apron or a physical radiation screen compared with just over three quarters of those who graduated in the 1960s. However, the percent of veterinarians who personally restrained animals during X-rays and who worked generally with inadequate animal restraint was higher in recent graduates. After adjusting for gender and practice type, more recent graduates were much more likely than graduates from the 1960s to use an anesthetic waste gas scavenger and to use either an apron or a screen when taking X-rays (Table 4). There was no statistically significant difference in the use of animal restraints by decade of graduation or seatbelt use. Large animal veterinarians were less likely to use all types of protective practice. Trends over time The most marked change over time in the jobs was the change in the gender balance from the 1980s (Table 5). Over time, there were more jobs in small animal practice than large animal or mixed practice. Regarding occupational hazards, over time, veterinary jobs were more likely to involve taking X-rays and undertaking surgery, but less likely to involve driving for work purposes. Of those jobs involving taking X-rays, the number of X-rays taken per week tended to increase, but there was no obvious pattern of personal restraining of animals during X-rays. Of those jobs involving surgery, hours spent in surgery tended to decrease. Repeating these analyses for only those jobs which started and finished in the same year grouping did not change the patterns seen. 8

The use of scavengers for waste anaesthetic gas has increased markedly over time. Only 3.8% of jobs commencing in the 1960s had a scavenger present, but one was present in 70.5% of jobs commencing since 1997. After adjusting for gender, practice type, decade of graduation and hours of work, over time, jobs became less likely to involve more than 9 hours surgery per week, and much less likely to require more than 7 hours driving per week (Table 6). There was no change in the number of jobs which required more than 7 X-rays to be taken per week. DISCUSSION This survey is, to our knowledge, the first to examine trends in occupational health and safety practices by veterinarians. We have shown that occupational health issues are still important in veterinary practice, with most veterinarians exposed to a number of physical and chemical hazards and many using inadequate protection. Others have reported high levels of injuries and have highlighted the physical risks to which veterinarians are exposed in routine work within the profession (Jeyaretnam, 2000a and 2000b; Landercasper, 1988). Recent graduates were more likely to take X-rays than earlier graduates even after adjusting for practice type, gender, and hours of work. It is not clear why this should be so, but may be due to them having less confidence in their clinical judgement than more experienced veterinarians. In addition, there may have been changes in training and practice which have more recently emphasized the use of tests such as X-rays in order to make more accurate diagnoses more quickly. X-ray protection was more likely to be used by recent graduates, suggesting that perhaps professional education needs to target older veterinarians, who may not be adequately protected from ionizing radiation. In addition, jobs since the 1970s were more likely than those commencing in the 1960s to involve taking of X-rays, possibly 9

because the X-ray equipment has become more available and affordable. However, the number of X-rays taken by those taking X-rays has not changed after adjustment for confounding variables. Similar results were found in a US study of female veterinarians (Wiggins, 1989) who reported an increase in the use of protective radiographic equipment with a concurrent increase in the use of X-rays in the diagnostic processes. Jacobsen and VanFarowe (1964) highlighted that although veterinarians were aware of the dangers in using radiographic equipment many did not apply their knowledge or use preventive measures to reduce exposure to ionizing radiation. More recent graduates than early graduates undertook surgery. For those who did surgical procedures, the amount of time spent in surgery was not different by decade of graduation, after adjusting for type of practice, gender, and hours of work. There was a major difference in the use of waste anaesthetic gas scavenging with recent graduates much more likely to work with one present than older practitioners. In addition, the proportion of jobs with a scavenger present has increased markedly over the years and now is close to 70%. However, another smaller survey of veterinary clinics in Australia (Jeyaretnam, 2000a) reported that only 30% of clinics had extractor fans or scavenging systems (although 37% of respondents did not answer the question). The previously mentioned study of female veterinarians showed no trend in regard to use of WAG scavenging with year of graduation from veterinary school but reported a marked increase in the use of WAG scavenging units for pregnancies that occurred between 1960 and 1987(Wiggins, 1989). Driving was more common in large animal practitioners, and large animal practice is becoming less common. After adjusting for practice type, there were no differences between recent and earlier graduates in the likelihood of doing >7hours driving a week, but there had been change over time, with fewer jobs involving driving to see patients. House calls are 10

more time consuming than seeing patients at surgeries and now that most clients have cars, it is probable that animal owners are less likely to require a house call. We saw the recognized changes in the profile of veterinary practice with recent graduates more likely to be female and more likely to work in small animal practice (Slater, 2000). The proportion of veterinarians exposed to anaesthetic gases and ionizing radiation seems much higher in Australia than in Finland (Reijula, 2003). In Finland 60% of veterinarians reported never being exposed to anaesthetic gases compared to 17% in Australia in our study and 12% in another smaller Australian study (Jeyaretnam, 2000b). Just over half of Finnish veterinarians (52%) reported never taking X-rays compared to 15% in our study and 6% in the other Australian study (Jeyaretnam, 2000b, Reijula, 2003) Exposure to waste anaesthetic agents and ionizing radiation has been linked to adverse reproductive outcome (Ward, 1982) and impaired mental and physical performance (Korczynski, 1999). This study has demonstrated that the use of protective equipment for ionizing radiation and scavenging systems for anaesthetic gases has increased in recent Australian veterinary graduates. This coincides with an increase in the number of radiographs taken and the time spent performing surgeries by recent graduates compared to earlier graduates. This survey is, to our knowledge, the first to examine trends in the occupational health and safety practices of veterinarians. The participants included veterinarians from all areas of Australia and is one of the largest surveys which have been done of the profession. Practice conditions in Australia are probably similar to those in other developed countries. Although our response rate was relatively low, we included large numbers of veterinarians from all types of clinical practice. The participation fraction was slightly better among females and those graduating in the 1990s but the actual differences were relatively small 11

(albeit statistically significant) suggesting that our study provided a reasonable representation of veterinarians. In addition, we excluded those we knew to be overseas. We do not have information on the practice types of those who did not participate, however it may be that those not in clinical practice or veterinary work were less likely to participate. Practice profiles have changed over time with work involving fewer hours of work, more time in surgery but less driving. Safety has increased in some areas over time, such as the use of anesthetic waste gas scavengers and most veterinarians appear to use adequate radiation protection in their current jobs. However, the potential for exposure to a number of physical and chemical hazards is still high and inadequate restraint of animals is still very common. There is still scope for increasing the use of gas scavengers, and for improving protective practices among earlier graduates and large animal veterinarians. ACKNOWLEDGEMENTS This study was funded by the Cancer Council of Western Australia and the University of Western Australia Research Grants. We would like to thank the alumni organisations of Queensland University, Murdoch University and the University of Sydney for providing lists of graduates, the Australian Veterinary Association, the New Zealand Veterinary Registration Board and the veterinary registration boards of the Australian States for their assistance. Lin Fritschi and Lesley Day are supported by the National Health and Medical Research Council. REFERENCES Korczynski RE Anesthetic gas exposure in veterinary clinics. App Occup Env Hyg 1999; 14:384-90 12

Fritschi L, Day LM, Shirangi A, et al. Injury in Australian veterinarians. Occup Med (Lond). 2006; In press 56:199-203. Jacobsen A, Van Farowe DE Survey of X-ray protection practices among Michigan veterinarians J Am Vet Med Assoc 1964:145:793-7 Jeyaretnam J, Jones H, Phillips M. Disease and injury among veterinarians. Aust Vet J. 2000;78:625-9. Jeyaretnam J, Jones H. Physical, chemical and biological hazards in veterinary practice. Aust Vet J. 2000b;78:751-8. Landercasper J, Cogbill T, Strutt, et al. Trauma and the veterinarian. J Trauma 1988; 28:1255-8. Reijula K, Rasanen K, Hamalainen M, et al. Work environment and occupational health of Finnish veterinarians. Am J Ind Med. 2003;44:46-57. Shirangi A, Fritschi L, Holman CD. Prevalence of occupational exposures and protective practices in Australian female veterinarians. Aust Vet J. 2007;85:32-8. Slater MR, Slater M. Women in veterinary medicine. J Am Vet Med Assoc. 2000;217:472-6. Torda TA, Jones R, Englert J. A study of waste gas scavenging in operating theatres. Anaesth Intensive Care. 1978;6:215-21. Ward GS and Byland RR Concentration of halothane in veterinary operating and treatment rooms J Am Vet Med Assoc 1982;180:174-7. Wiggins P, Schenker MB, Green R, et al. Prevalence of hazardous exposures in veterinary practice. Am J Ind Med 1989; 16:55-66 13

Table 1. Prevalence of occupational hazards by decade of graduation in current veterinary practitioners Hazard Number 1960-9 1970-9 1980-9 1990-2000 p* Mean no. hrs worked/wk 2531 47.7 48.8 41.5 42.8 Take X-rays Mean no. X-rays in those taking X-rays Do surgery No Yes No Yes Mean no. hrs/wk surgery in those doing surgery Drive for work Mean no hrs/wk driving in those who drive No Yes No. km driven /wk 0 <50 50-100 389 2014 36.0 64.0 21.4 78.6 18.7 81.3 6.2 93.8 2014 6.5 8.0 6.5 7.3 0.03 358 2095 28.8 71.2 19.9 80.1 17.3 82.7 5.5 94.5 2095 10.2 10.0 8.6 8.4 497 2028 15.1 84.9 13.9 86.1 22.1 77.9 22.6 77.4 2028 8.7 7.8 5.8 5.8 256 1102 486 10.2 28.7 19.2 7.2 38.4 21.5 11.1 49.1 17.1 11.2 46.8 19.5 200+ 681 41.9 32.9 22.6 22.5 Pesticide use Never 433 19.5 18.1 21.3 13.6 Rarely 609 48.5 45.6 44.4 33.2 Weekly 1025 18.0 21.7 23.4 28.4 Daily 438 13.9 14.6 10.9 24.9 * p-value from chi-square test for categorical variables and one-way ANOVA for continuous variables. 14

Table 2. Odds ratio (OR) and 95% confidence intervals (95%CI) for current exposure to occupational hazards adjusted for all other variables in current veterinary practitioners. Hazard 7> X-rays/wk 9> hrs surgery /wk Graduation yr 1960-9 p-value* 1970-9 1980-9 1990-2000 Type of practice Small animal p-value* Gender p-value* Mixed Large animal Male Female Hours of work (continuous) p-value* 1.59 (7-2.35) 1.51 (2-2.24) 2.59 (1.75-3.83) 0.81 (0.65-0) 0.38 (0.28-0.51) 9 (0.87-1.35) 0.47 6 (5-6) *p-value for term in logistic regression model 0.96 (0.67-1.37) 0.86 (0.60-1.24) 0.86 (0.61-1.24) 0.78 0.51 (0.41-0.63) 0.05 (0.04-0.08) 0.85 (0.69-6) 0.14 6 (5-7) 7> hrs driving /wk 0.91 (0.64-1.30) 0.98 (0.68-1.41) 8 (0.75-1.57) 0.71 8.09 (6.17-10.6) 16.1 (11.9-21.7) 0.49 (0.38-0.63) 4 (3-5) 15

Table 3. Prevalence of protective practices by decade of graduation in respondents in current clinical veterinary practice. Protective practice Anaesthetic waste gas scavenger# Number 1960-9 1970-9 1980-9 1990-2000 p- value** 2095 34.3 58.1 72.3 76.1 Lead apron* 2014 75.5 87.9 86.7 91.9 Thyroid protector* 2014 19.9 36.9 44.7 52.4 Lead gloves* 2014 56.3 54.1 49.4 44.1 Physical radiation screen* 2014 13.9 13.5 15.2 18.9 X-ray film holders* Either apron or screen* Restrain animals during X- rays* Work with inadequate restraint 2014 24.5 34.5 20.9 18.3 2014 77.5 88.7 89.0 94.7 2014 66.9 75.2 64.4 76.1 2365 17.3 21.8 20.5 26.1 Seatbelt~ 1813 98.3 95.3 96.0 95.7 # In those doing surgery * In those taking X-rays ~ In those driving for work who answered this question ** p-value from chi-square test 16

Table 4. Use of protective practices by participants in current clinical veterinary practice adjusted for all variables in the table. Use anaesthetic waste gas scavenger 1 N=2095 Use either apron or screen 2 N=2014 Always work with adequate animal restraint N=2365 Always wear seatbelt 3 N=1813 Graduation 1960-9 1970-9 2.70 (1.87-3.90) 2.56 (1.56-4.19) 0.82 (0.57-1.19) 0.34 (0.10-1.14) 1980-9 4.76 (3.28-6.92) 3.17 (1.92-5.23) 0.80 (0.55-1.16) 0.32 (0.09-1.12) 1990-2000 5.80 (4.00-8.42) 7.61 (4.42-13.1) 0.48 (0.81-1.29) 0.28 (0.08-0) p-value* Gender Male Female 0.96 (0.78-1.20) 0.53 (0.37-0.75) 2 (0.81-1.29) 1.28 (0.73-2.21) p-value* 0.74 0.85 0.001 Practice type Small Mixed 0.63 (0.51-0.78) 0.82 (0.58-1.15) 0.37 (0.28-0.48) 0.35 (0.15-0.59) Large 0.34 (0.24-0.47) 0.37 (0.23-0.61) 0.67 (0.51-0.88) 0.29 (0.19-0.64) p-value* 1. In those doing surgery 2. In those doing X-rays 3. In those driving for work who answered this question * p-value from term in logistic regression model 17

Table 5. Prevalence (as %) of occupational hazards by year group in which job started for Australian veterinarians. Hazard 1960-9 1970-9 1980-9 1990-6 1997-2002 p* N 287 1151 2013 2107 1170 Gender Male 90.2 86.3 58.3 43.0 35.3 Female 9.8 13.7 41.7 57.0 64.7 Practice type Large Mixed Small 30.1 51.2 18.8 17.7 51.6 30.7 9.9 45.1 45.0 8.4 38.4 53.2 7.8 30.3 61.9 Mean no. years in job 5.7 6.7 4.9 3.1 1.7 (Range) (<1-39) (<1-32) (<1-22) (<1-12) (<1-5) Percent of jobs start and 51.2 57.5 60.5 55.1 100 finish in same time period Mean no. hrs worked/wk 53.0 53.2 48.5 45.1 41.5 Take X-rays No 26.2 7.9 5.5 4.4 5.7 Yes 73.8 92.1 94.5 95.6 94.3 Mean no. X-rays* 6.0 7.2 7.3 7.3 7.1 0.22 Restrain animals* Rare/no Daily/wkly 27.6 71.4 18.0 82.0 19.1 80.9 19.1 77.9 28.0 72.0 Do surgery No Yes 2.1 97.9 1.8 98.2 2.6 97.4 2.9 97.1 4.0 96.0 0.006 Mean no. hrs/wk surgery^ 10.5 10.6 10.0 9.3 8.5 Waste gas scavenger^ No 96.2 91.3 78.4 51.4 29.5 Yes 3.8 8.7 21.6 48.6 70.5 Drive for work No Yes 6.4 93.6 7.0 93.0 13.7 86.3 18.0 82.0 24.8 75.2 Mean no hrs/wk driving# 15.8 11.7 8.4 7.0 6.3 Pesticide used Never/rarely Weekly/daily 50.2 49.8 49.3 50.7 48.9 51.1 50.4 49.6 48.3 51.7 0.76 * in those jobs which involve taking X-rays ^ in those jobs which involve doing surgery # in those jobs which involve driving for work * p-value from chi-square test for categorical variables and one-way ANOVA for continuous variables. 18

19

Table 6. Odds ratio (OR) and 95% confidence intervals (95%CI) for exposure to occupational hazards in each veterinary job over time adjusted for all other variables. Hazard 7> X-rays/wk n = 7071 jobs Yr job started 1960-9 1970-9 1.19 (0.81-1.75) 1980-9 1.22 (0.80-1.84) 1990-96 1.19 (0.78-1.83) 1997-2002 1.18 (0.76-1.83) 0.9 Type of practice Small animal Mixed 0.63 (0.57-0.71) Large animal 0.52 (0.43-0.63) Gender Male Female 1.17 (4-1.32) 0.01 Graduation 1960-9 1970-9 1.35 (3-1.76) 1980-9 1.37 (1-1.84) 1990-2000 2.05 (1.49-2.83) Hours of work (continuous) 6 (5-6) * p-value from term in logistic regression model 9> hrs surgery /wk n = 7149 jobs 0.94 (0.66-1.33) 0.84 (0.58-1.22) 0.70 (0.47-2) 0.55 (0.37-0.82) 0.51 (0.46-0.57) 0.11 (0.09-0.14) 5 (0.94-1.18) 0.39 0.98 (0.77-1.26) 0.88 (0.66-1.16) 1.14 (0.84-1.54) 0.01 6 (5-6) 7> hrs driving /wk n = 6620 jobs 0.66 (0.43-0) 0.41 (0.26-0.64) 0.33 (0.21-0.53) 0.29 (0.18-0.48) 8.7 (7.5-10.0) 22.8 (18.1-28.6) 0.60 (0.52-0.69) 0.92 (0.68-1.23) 7 (0.78-1.50) 0.96 (0.67-1.38) 0.40 5 (4-5) 20