Industrial Sectors With High Risk of Women s Hospital-Treated Injuries

Transcription

1 AMERICAN JOURNAL OF INDUSTRIAL MEDICINE 50:13 21 (2007) Industrial Sectors With High Risk of Women s Hospital-Treated Injuries Pete Kines, PhD, Harald Hannerz, PhD, Kim Lyngby Mikkelsen, MD, PhD, and Finn Tüchsen, MSc Background Women s occupational injury rates are converging with those of males. Associations between female workers hospital treated injury rates, industrial sector and injured body area were analyzed to provide for better-focused injury prevention of women s hazardous jobs. Methods Females standardized hospital treatment ratios (SHR) and the excess fraction for five body regions (head/neck, thorax, back, upper and lower extremities) were calculated for 58 industrial sectors for Results Five industrial sectors, Cleaning, laundries and dry cleaners, Transport of passengers, Hotels and restaurants, Hospitals and Transport of goods had significantly high SHRs for all five body regions. The excess fraction for upper extremity injuries revealed that 14% 27% of injuries could theoretically have been avoided. Conclusions There is strong evidence for an association between women s hospital treated injuries and industrial sector. The results justify the need for gender-sensitive analyses to orient injury prevention programs. Am. J. Ind. Med. 50:13 21, ß 2006 Wiley-Liss, Inc. KEY WORDS: industry; injury; hospitalization; injury surveillance; injury prevention INTRODUCTION Workforce and Gender Studies in the literature reveal that women are underserved by the occupational safety and health research community [Chen and Hendricks, 2001; Messing et al., 2003]. Over the past 50 years women have accounted for an increasing proportion of paid economic activities in National Institute of Occupational Health, LersÖ Parkalle, Copenhagen, Denmark *Correspondence to: Pete Kines, National Institute of Occupational Health Denmark, LersÖ Parkalle 105, 2100 Copenhagen, Denmark. pk@ami.dk Accepted18 September 2006 DOI /ajim Published online in Wiley InterScience ( developed countries. In the US females accounted for 34% of the labor force in 1962, and it is predicted that they will account for 48% in the year 2008 [Fullerton, 1999]. Similar trends are seen in Europe, where females in 2004 accounted for 42% of the workforce [Wilén, 2006]. In 2005 the European country with the highest female employment rate was Denmark [Bautier, 2006], where women accounted for 47.5% of the workforce [Statistics Denmark, 2006]. Alongside the converging rates of men and women s labor force participation, there continues to be a large degree of industrial and occupational gender segregation. This is seen with men s dominance in specific industries such as construction, farming, and transport, whereas women are dominant in industries and occupations such as healthcare, homecare, public school teaching, food service, retail trade, clerical and cashier service. In gender mixed industries, there is also evidence for job task gender segregation, such as women s dominance in light, repetitive work tasks in ß2006Wiley-Liss,Inc.

2 14 Kines et al. manufacturing processes [Lindqvist et al., 1999; Chen and Hendricks, 2001; Islam et al., 2001; Messing et al., 2003; Smith and Mustard, 2004]. Women s Occupational Injuries At the same time as women are accounting for an increasing proportion of the labor force, there has been similar gender convergence for occupational injury rates [Smith and Mustard, 2004]. Much of the increasing injury rates for women stem from industries and occupations with traditionally high injury rates such as in healthcare [Lindqvist et al., 1999; Islam et al., 2001; Mustard et al., 2003]. In spite of industry specific high injury rates, it is well-documented that women have lower fatal and nonfatal injury rates than males [Layne and Landen, 1997; McCaig et al., 1998; Lindqvist et al., 1999; Jackson, 2000; Islam et al., 2001; Mustard et al., 2003; Tyler and Jackson, 2003a,b]. In one study the authors estimated that, given the female injury rates in each industry, if female labor force participation was identical to male labor force participation by industry and physical work demands, women would experience an approximate 90% increase in injuries [Smith and Mustard, 2004]. Women s Serious Occupational Injuries Reduction of injuries, and in particular fatal and serious nonfatal injuries, is often a goal for injury prevention programs at international, national and company levels, and analysis of occupational injury surveillance data is essential for documenting, planning, implementing and evaluating prevention programs. However, measures for serious nonfatal occupational injuries vary widely and provide challenges in making national and international comparisons, as definitions for serious injuries vary from, for example, lost working time, type of injury, numerical injury scales and scores, hospitalization, degree of disability, employability, compensation revenues paid out, as well as total costs to the injured, company, or society. An additional challenge is that injury statistics are often not dichotomized into the two genders [Forst et al., 1999], thus making it difficult to target gender specific safety initiatives. Few epidemiological studies have analyzed occupational injury risks in a national population of women across all industrial sectors. Studies regarding women s serious nonfatal injuries are often based on local registries, with relatively few observations for women to provide reliable and conclusive evidence at industry or occupational group level. Many studies focus on sub-populations of women [Chen and Hendricks, 2001], specific industries and injuries [Zwerling et al., 1993; Ashbury, 1995; Kelsh and Sahl, 1996; Gluck and Oleinick, 1998; Hannerz and Tüchsen, 2002], and explanations for injury risk differences within a specific industry or job have been provided [Zwerling et al., 1993; Wohl et al., 1995; Lindqvist et al., 1999]. There is a need for studying women s industry specific injuries, and a number of sources of injury data in Denmark exist for carrying out these analyses. Two of these contain self-reported work injury data [Statistics Denmark, 1999; Burr et al., 2003], and a third is based on emergency department contacts [Nielsen and Frimodt-Møller, 2005]. All three of these contain representative samples of the labor force, but too few observations for gender and industrial sector specific analyses. A fourth source, the National Working Environment Authority [2005], provides national data on injuries resulting in a minimum of 1 day s absence from work. Preliminary analyses of these data reveal that although the number and incidence rates of serious occupational injuries (i.e. deaths, amputations or fractures) declined significantly for both women and men in the period , they were declining at a significantly slower rate for women than for men. In this 4-year period relative injury severity rate ratios (severe vs. minor injury) [Kines, 2001] increased significantly for women, while they decreased for men (borderline significant). These results justify the need for gender specific analyses, however, in spite of compulsory reporting to the national authorities, these data are subject to underreporting, with a great degree of variation between industrial sectors and trades [National Work Environment Authority, 1996; Burr, 1997; Statistics Denmark, 1999]. There is therefore a need for a more reliable source of data in studying industrial sectors with high risks of injury. Current Study The purpose of this study was to test the hypothesis that there are associations between female workers injuries, their industrial sector and their subsequent injured body area. Such gender-sensitive analyses may be helpful in programs aiming to make workplaces safe for all workers; young and old, men and women. MATERIALS AND METHODS Data Source The present study is based on data derived from the Danish Occupational Hospitalization Register (OHR), a research register obtained through a record-linkage between three national registers the centralized civil register, the patient register, and the employment classification module [Soll-Johanning et al., 2004]. A personal identification number (PIN) from the national patient register was used in the cross-linking procedure and for information on gender, date and year of birth. Age and gender are encoded in the PIN and recorded practically without error. The completeness and

3 Women s Injuries and Industry 15 accuracy of these data are shown by the fact that matching of various registers on PIN was 100% complete. The centralized civil register contains information on gender, addresses and dates of birth, death, and migrations for every person who is or has been an inhabitant of Denmark. The patient register contains information (inpatient, outpatient, emergency ward visits), and discharge diagnosis from all hospitals in Denmark [Tüchsen et al., 1996; Baarts et al., 2000]. A person s industry, occupation and employment status are registered annually in the employment classification module [Soll-Johanning et al., 2004]. Fifty-eight industrial codes were derived from a Danish aggregate of the EU classification of industries NACE [Statistics Denmark, 1996]. Classification into these 58 industrial groups offers the opportunity to reveal the diversity in the frequency and rates of injuries between different industrial sectors, and as to which injuries result in hospitalization. Studies using hospital or trauma registry data have often concluded that there is a need for variables such as type of industry and occupation [McCaig et al., 1998; Forst et al., 1999] in order to enhance a register s value in occupational injury surveillance. Other registries have these variables, yet provide no gender specific results [Husberg et al., 1998]. The OHR provides both these latter two types of data, but does not contain reliable information to differentiate between leisure time and work injuries for each person. It does, however, allow for calculation of the fraction of injuries, poisonings and other consequences of external causes that could be attributed to work. The limitations of using emergency department and hospital records for studying occupational injuries have been documented in many previous studies [Tüchsen et al., 1996; Geidenberger et al., 1997; Layne and Landen, 1997; McCaig et al., 1998; Forst et al., 1999]. There can be gender as well as industrial differences in behavior as regards to seeking medical treatment for injury, that is, females and health care workers are more likely to seek treatment due to their knowledge and proximity to health care treatment, respectively [Layne and Landen, 1997]. On the other hand, onsite medical facilities can reduce the number of workers presenting injuries in hospitals, but such on site facilities are very rare in Denmark [Spangenberg et al., 2005]. Follow-Up All economically active women in Denmark aged years on January 1, 1999 (N ¼ 1,165,898) were followed in the period for first hospital contacts with the principal diagnoses according to the International Classification of Diseases [World Health Organization, 2003]: Injuries to the head and neck (ICD-10 ¼ S00-S19), injuries to the thorax (ICD-10 ¼ S20-S29), injuries to the abdomen, lower back, lumbar spine and pelvis (ICD-10 ¼ S30-S39), injuries to the upper extremities (ICD-10 ¼ S40-S69), and injuries to the lower extremities (ICD-10 ¼ S70-S99). These diagnoses all belong to the chapter Injury, poisoning and certain other consequences of external causes, while musculoskeletal diseases like epicondylitis, carpal tunnel syndrome, or prolapsed disks have not been included. Dates of deaths, emigrations and hospital treatment were used to calculate person years at risk for each individual. The validity of injury diagnoses in the national patient register in 2002 was found to be between 84.9% and 98.4% at a three-digit level [Nickelsen, 2002], and there is no reason to believe that the diagnoses are less valid today. Statistical Methods Indirect standardization was used to adjust for 5-year age groups, with all economically active women in the total population of Denmark as the standard population. Age standardized hospital treatment ratios (SHR) and hospitalization (inpatient) ratios were calculated by dividing the observed number of hospital treatment referrals for a given industrial sector by the expected number and then multiplying this by 100. As the age structure may vary between industries, the expected number was based on corresponding age-specific rates, in 5-year classes, for all gainfully employed women in Denmark. In the present study the term industrial sector association is used to describe the relationship between industrial sector and injury category. If there is only random variation in the industrial sectors specific SHRs, the injury category is inferred not to be associated with industrial sectors. However, if the difference in SHRs between industrial sectors is higher than expected, it could be assumed that there is a causal association between industrial sector and injury category. Thus, for each injury category Pearson s chi-square sum [Pearson, 1900] was calculated to test the null hypothesis that the SHRs were independent of industrial sector. The level of significance was set to 5%. SHRs are reported rather than incidence rates as they are independent of admission criteria. Comparing incidence rates temporally or geographically would be influenced by differences in the level of public health services. SHRs are relative rates between comparison groups and only influenced by time and place to the extent that the relative level of public health services between individual groups differ over time and place. Thus, reporting SHRs instead of incidence rates facilitates comparisons across time and place. This approach has one drawback, though, as the injury level cannot be compared with other sources. Finally, standardized hospitalization (inpatient) ratios are also reported for comparative purposes, as well as the excess fraction for each body region. This fraction is the proportion of women s hospital treated injuries that would not have occurred if the hospital treatment rate in each of the

4 16 Kines et al. TABLE I. Industrial Standardized Hospital Treatment Ratios (SHR) for Injury Among Women in Danish Industrial Sectors1999^2003 (Ranked by Number of Significantly High or Low SHR) No. Industrial sector Persons at risk Head Thorax Back Upper extremities Lower extremities 1 Cleaning,laundries, and dry cleaners *,a 141*,a 141*,a 117*,a 110*,a 2 Transport ofpassengers * 129* 130*,a 110* 121* 3 Hotels and restaurants *,a 131* 127* 128* 114* 4 Hospitals c * 109* 120* 144*,a 107* 5 Transport of goods * 117* 122* 107* 110* 6 Slaughterhouse industry * 160* 186* 115* 7 Finishing * * 124* 120* 8 Nursinghomes,home care,etc a 107*,a 111* 105* 105*,a 9 Fire service,lighthouse and salvage corps *,a * 137* 10 Beverage industry * * 113* 11 Unstated *,a 115*,a 117* * 12 Surveillance, armedforces,police etc * *,a 13 Entertainment,culture and sport *,a * Manufacture ofbread,chocolate,tobacco etc *,a 108* 15 Manufacture of dairy products * 116* 16 Metal and steelworks, and foundries * 84* 17 Manufacture of transport equipment * 44* 120* Paper, cardboard and bookbinding industries * Pharmaceutical industry * Poultry slaughtering and fish products *,a Supermarkets, department stores etc * Mineral,oil,rubber andplastic products a * Garage * 24 Car industry * 148* Iron and metal industries * 92*,b 26 Child care etc *,a b 98* 109* 27 Stone-works, pottery, and glass industry Shipyards Manufacture of industrial chemicals Car dealers Fishing Photographers/film and videoproduction * Horticulture and forestry * 34 Printing works and publishing * 95 b 35 Personal care and other services * Navy and road contractors * Medical equipment/toys/cameras/etc * 38 Telecommunication * Office and adm. (transport and wholesale) b * 96 b 40 Bricklayer, joiner, and carpentry work * Insulation and installation businesses * Electricity and heat supply * Heavy raw material and semi-manufacture * Sewers,water- and gas supply * 76* 45 Electricity and electronics industry * * 46 Wholesale trade * 96*,b 47 Office and adm. (service) * 89* 48 Education and research * 95 92* 98*

5 Women s Injuries and Industry 17 TABLE I. (Continued) No. Industrial sector Persons at risk Head Thorax Back Upper extremities Lower extremities 49 Shops * 91 87* 93* Manufacture of wood and woodproducts * 91 68* 106 a 87* 51 Libraries and archives * 87 74* 81*,b 52 Textile, clothing, and leather industry * * 70* 53 Private office and adm b 88*,b 87*,b 84*,b 91* 54 Health care not elsewhere classified * 85* 85* 87* 85* 55 Engineering industry * 76* 78* * 56 Agriculture *,a 79* 92 84* 81* 57 Finance/ Public office and adm *,b 84*,b 78*,b 80*,b 93* 58 General practitioners,dentists etc * 79* 76* 79* 78*,b Total *P < a Significantly high for hospitalized injury. b Significantly low for hospitalized injury. c Possible referral bias (see text). industrial sectors had been as low as that in the sector with the lowest hospital treatment risk. The excess amount of cases is then calculated to show, theoretically, how many hospital treated injuries could be spared if the risk level in all industrial sectors is as low as that in the sector with the lowest hospital treatment risk [Feveile et al., 2006]. RESULTS The study population consisted of 1,165,898 women in the 58 industrial sectors. The SHRs for each industry and injured body region are given in Table I, ordered after the number of body regions with significantly high or low SHRs in each industrial sector. Fifteen industrial sectors (no. 1 15), accounting for 35% of the study population, had significantly high SHRs for at least two body regions. In five of these sectors (no. 1 5), for example cleaning, laundries and dry cleaners, the SHRs were significantly high for all five body regions. Fifteen industrial sectors (no ), accounting for 42% of the study population, had significantly low SHRs for at least two body regions. In two of these sectors (no ), for example General practitioners, dentists, etc., the SHRs were significantly low for all five body regions. In terms of hospitalized (inpatient) injury the cleaning, laundries and dry cleaners and nursing home, home care, etc. have significantly high SHRs for five and three body regions, respectively (see letters a in Table I). Figures 1 and 2 are provided for further illustration of these results. Figure 1 shows all the industrial sectors with significantly high (P ¼ 0.05) SHRs for upper extremity injuries with 95% confidence intervals, ordered by their SHR, highest to lowest. Figure 2 shows the SHR profile for injured slaughterhouse workers, with 95% confidence intervals (CI). The excess fraction for upper extremity injuries was (95% CI: ), resulting in an average of 7,481 excess cases per year (Table II). In other words, if the risk level for upper extremity injuries for all industrial sectors was as low as that in the sector with the lowest hospital treatment risk, then 27.4% of all upper extremity injuries (or 7,481 per year) could have been avoided. DISCUSSION A very strong association was found between women s risk of hospital treated injury and industrial sector. This is in spite of there being no differentiation between whether the injuries were due to occupational or leisure time activities. Industrial sectors such as Cleaning, laundries and dry cleaning, Transport of passengers, Hotels and restaurants, Hospitals and Transport of goods are all high risk sectors for all five body regions. The results also reveal that back and upper extremity injuries, as well as lower extremity and thorax SHRs are significantly high in adult health care such as in the hospital, nursing home and home care sectors, but not for personal care and other services. These results can be put in perspective by comparison with a Swedish study [Lindqvist et al., 1999] which reported that women in the health sector had high levels of injuries to upper extremities (55%), spine, chest and pelvis (20%). An alternative explanation is that health care workers in hospitals are more likely to be treated in the hospital than by a general practitioner [Tüchsen et al., 1996].

6 18 Kines et al. FIGURE 1. Significantly high industrial standardizedhospital treatmentratios (SHRs) forupperextremity injuries amongwomen in Danishindustrialsectors1999^2003and95% confidenceintervals.

7 Women s Injuries and Industry 19 FIGURE 2. Industrialstandardizedhospitaltreatmentratios(SHRs) forinjuredfemale slaughterhouseworkersindenmark1999^2003and95% confidenceintervals. Studies have estimated that 31% 34% of nonfatal occupational injuries are treated at emergency departments and hospitals [Geidenberger et al., 1997; McCaig et al., 1998]. No comparable data exist for Denmark, however, one study found that 28% of self-reported occupational hand injuries were treated at an emergency department [Skov et al., 1999]. Other studies show that the most serious nonfatal occupational injuries will likely present to hospital emergency departments [Fingar et al., 1992; McCaig et al., 1998]. Hospital treated cases are, however, often overrepresented by fall injuries, chemical injuries, leg fractures, transportation and nursing home injuries, and underrepresented by overexertion injuries [Waller et al., 1995]. Women have been shown to have lower incidence rates than men for emergency department and hospital contacts [Forst et al., 1999], and it is likely that they are underrepresented to an even greater degree in hospitalizations. As the proportion of hospital treated injuries may vary according to industrial sector, the rate of hospital treated injuries, as a proxy measure of the underlying total rates of injuries, is liable to differential industrial sector referral or treatment seeking bias. Furthermore, studies based on hospital admissions are often associated with referral bias arising from social and geographical differences in the tendency to consult hospitals for medical care [Tüchsen et al., 1996]. All acute hospital care in Denmark is free of charge and emergency hospitals are geographically distributed in a way that hospital care can be promptly reached [Soll-Johanning et al., 2004]. Compared to other studies, the advantage of the present study is that the OHR represents all admissions to all public hospitals in Denmark, and since the study also deals with hospitalized injuries, it also has the advantage of these specific injuries being the more severe cases. One of the strengths of the study was the availability and use of information about person-years at risk, however, no information was available about the number of working hours at risk. Today women are as occupationally active as men in Denmark, but in some industries women may still work fewer hours and/or to a greater degree be employed only part-time [Messing et al., 1994]. If so, the results in this study may underestimate occupational risks in specific industries. In comparison to the results of a similar study of men s hospitalized injuries (inpatient only) using OHR data [Baarts et al., 2000], the results give some credibility to industrial and occupational gender segregation [Lindqvist et al., 1999; Chen and Hendricks, 2001; Islam et al., 2001; Smith and Mustard, 2004]. It is the job and not the gender that is the problem [McDiarmid et al., 2000]. A clear case for industrial gender segregation is the Cleaners, laundries and dry cleaning sector, where the women in this study have high SHRs for all body regions, both in terms of hospital treatment in general and to a greater degree for hospital admissions, whereas men in the Baarts et al. study had no significantly high SHRs for this sector. This provides evidence for gender segregation that favors men in this industry. Other sectors provide similar examples in which women have significantly high SHRs for at least two body regions, whereas no significantly high SHRs are found for men, for example, Beverage industry and manufacturing industries for dairy products, transport equipment, bread, chocolate, tobacco, etc. However, previous studies have provided evidence that women carrying out physically demanding work are more likely to perform light, repetitive tasks, whereas men are more likely to perform heavy lifts [McDiarmid et al., 2000; Messing et al., 2003]. Women involved in the Transport of passengers sector had significantly high SHRs for all body regions, whereas men TABLE II. Excess Fraction and Excess cases among Hospital Treated Women in the DanishWorkforce1999^2003 and 95% Confidence Intervals (CI) Injured body part Cases Excess fraction 95% CI Excess cases 95% CI Lower extremity Head Thorax Back Upper extremity

8 20 Kines et al. in the sector had low SHRs for upper and lower extremity SHRs. In the Entertainment, culture and sport sector women had high head/neck and back SHRs, whereas men had low head/neck and lower extremity SHRs. The agriculture sector is another example of job task gender segregation, as men have significantly high SHRs for thorax, upper and lower extremities, whereas women have significantly low SHRs for all these three body regions. The engineering sector is also an example of job task gender segregation where men have significantly high head/neck and thorax SHRs compared to women s significantly low SHRs for these two body regions. In the General practitioners, dentists, etc. sector both men and women have significantly low head/neck and upper extremity SHRs, and in Education and research they both have low thorax, upper and lower extremity SHRs. On the other hand the Transport of goods sector had significantly high SHRs for all body regions for both men and women. In general, the results provide an initial basis for justifying the need for gender-sensitive analyses to orient injury prevention programs. The evidence shows that gender segregation can favor either men or women depending on the industrial sector. Studies have pointed to wage level [Islam et al., 2001], job control, and women s longer tenure in jobs with fewer opportunities for advancement [Martin, 1993] in predicting the risk of women s occupational injury. In another study, the USA s National Electronic Injury Surveillance System was used to analyze nonfatal occupational injuries among African-American women, and they found that injury patterns varied by industry in terms of source, event, diagnosis and body part [Chen and Hendricks, 2001]. There is an increased focus on occupational health and safety including improvements to equipment standards, procedures, safety culture and the working environment, as well as the eradication or outsourcing of hazardous work. However, many of the current physical, emotional, and psychological work environment exposure limits are based on the average man s capacity, and as a result, women may be at greater risk than men of incurring specific types of injuries in specific industries and job tasks [McDiarmid et al., 2000; Chen and Hendricks, 2001]. Hospital registries provide more specific details on the types of injuries that have occurred. The value of narrative data in an injury surveillance system was demonstrated in one study that allows for more precise identification and coding of etiologic factors of the injury incident [Layne and Landen, 1997]. Refinement of narrative data collection will also increase the effectiveness of injury surveillance for the evaluation of injury prevention efforts. Hospital registries can also be used to supplement national authority surveillance data that provides more information on the injury process such as where the event occurred, what the person was doing at the time, what products or materials were in use, what the person was injured by, etc. Combination of the two registries could contribute to solving the need for injury classifications that include specific tasks within an occupation [Islam et al., 2001]. Future studies could include analyses of the person s age, as employment demographics and injury patterns have been shown to differ for elder workers (>54) compared to a younger work force (<55) [Layne and Landen, 1997]. In conclusion, the results in this study provide strong evidence for an association between women s risk of hospital treated injury and industrial sector. Given equal access to treatment, occupational surveillance systems based on hospital treated injuries can be used inexpensively to identify high-risk industries for occupational injury prevention efforts. There is also evidence for positive and negative gender segregation within and between industrial sectors, whereby women can carry a greater or lesser burden of hospital treated injuries. These results provide an initial basis for justifying the need for gender-sensitive analyses to orient injury prevention programs. These programs could deal with re-engineering jobs, tasks and equipment, as well as with eradicating discriminatory practices in the assignment of workers to hazardous jobs and tasks. REFERENCES Ashbury FD Occupational repetitive strain injuries and gender in Ontario, 1986 to J Occup Env Med 37(4): Baarts C, Mikkelsen KL, Hannerz H, Tüchsen F Use of a national hospitalization register to identify industrial sectors carrying high risk of severe injuries: A three-year cohort study of more then 900,000 Danish men. Am J Ind Med 38: Bautier P A statistical view of the life of men and women in the EU25. Eurostat news release, Eurostat, Luxembourg. Burr H Occupational accidents (In Danish). In: Borg V, Burr H, editors. Work environment and health among Danish employees , Chapter 5: Copenhagen, Denmark: National Institute of Occupational Health Burr H, Bjorner JB, Kristensen TS, Tüchsen F, Bach E Trends in the Danish work environment in and their associations with labor-force changes. Scand J Work Environ Health 29: Chen GX, Hendricks KJ Nonfatal occupational injuries among African American women by industrial group. J Saf Res 32(1): Feveile H, Mikkelsen KL, Hannerz H, Olsen O Quantifying inequality in health in the absence of a natural reference group. Sci Total Env 367: Fingar AR, Hopkins RS, Nelson M Work-related injuries in Athens County 1982 to 1986: A comparison of emergency department and worker s compensation data. J Occup Med 34(8): Forst LS, Hryhorczuk D, Jaros M A state trauma registry as a tool for occupational injury surveillance. J Occup Env Med 41(6): Fullerton H Labor force projections to 2008: Steady growth and changing compositions. Mon Labor Rev 122(11): Geidenberger C, Jackson LL, Walker FJ National estimates of occupational injury from the National Health Interview Survey.

9 Women s Injuries and Industry 21 National Occupational Injury Research Symposium, Morgantown, West Virginia, USA October Gluck JV, Oleinick A Claim rates of compensable back injuries by age, gender, occupation, and industry. Do they relate to return-towork experience? Spine 23(14): Hannerz H, Tüchsen F Hospitalizations among female homehelpers in Denmark, Am J Ind Med 41:1 10. Husberg BJ, Conway GA, Moore MA, Johnson MS Surveillance for non-fatal work-related injuries in Alaska, Am J Ind Med 34: Islam SS, Velilla AM, Doyle EJ, Ducatman AM Gender differences in work-related injury/illness: Analysis of workers compensation claims. Am J Ind Med 39(1): Jackson LL Surveillance for nonfatal occupational injuries and illnesses treated in hospital emergency departments: United States, National Occupational Research Symposium 2000, Pittsburgh, Pennsylvania, USA, October 17 19, Kelsh MA, Sahl JD Gender differences in work-related injury rates among electric utility workers. Am J Epidem 143: Kines P Occupational injury risk assessment using injury severity ratios: Male falls from heights in the Danish construction industry, Human Ecol Risk Assess 7(7): Layne LA, Landen DD A descriptive analysis of nonfatal occupational injuries to older workers, using a national probability sample of hospital emergency departments. J Occup Environ Med 39(9): Lindqvist K, Schelp L, Timpka T Gender aspects of work-related injuries in a Swedish municipality. Saf Sci 31(3): Martin P Multiple gender contexts and employee rewards. Work Occup 20: McCaig LF, Burt CW, Stussman BJ A comparison of workrelated injury visits and other injury visits to emergency departments in the United States, J Occup Env Med 40(10): McDiarmid M, Oliver M, Ruser J, Gucer P Male and female rate differences in carpal tunnel syndrome injuries: Personal attributes or job tasks? Env Res 83: Messing K, Courville J, Boucher M, Dumais L, Seifert A Can safety risks of blue-collar jobs be compared by gender. Saf Sci 18: Messing K, Punnett L, Bond M, Alexanderson K, Pyle J, Zahm S, Wegman D, Stock SR, degrosbois S Be the fairest of them all: Challenges and recommendations for the treatment of gender in occupational health research. Am J Ind Med 43(6): Mustard C, Cole DC, Shannon H, Pole J, Sullivan T, Allingham R Declining trends in work-related morbidity and disability, : A comparison of survey estimates and compensation insurance claims. Am J Public Health 93(8): National Work Environment Authority Underreporting of occupational accidents (In Danish). Copenhagen, Denmark: National Work Environment Authority. National Working Environment Authority Reported occupational accidents Annual report 2004 (In Danish) National Work Environment Authority, Copenhagen, Denmark. Nickelsen T Data validity and coverage in the Danish National Health register (in Danish with a summary in English). Ugeskrift for Læger 164: Nielsen JW, Frimodt-Møller B Extrapolation of data on occupational injuries to national level. 16th Nordic Research Conference on Safety. National Institute of Public Health, Copenhagen, Denmark. Pearson K On the criterion that a given system of deviations from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling. Lond Edinb Dublin Phil Mag J Sci 50(5): Skov O, Jeune B, Lauritzen J, Barfed T Work-related hand injuries (in Danish). Ugeskrift for Læger 16: Smith PM, Mustard CA Examining the associations between physical work demands and work injury rates between men and women in Ontario, Occup Environ Med 61(9): Soll-Johanning H, Hannerz H, Tüchsen F Referral bias in hospital register studies of geographical and industrial differences in health. Dan Med Bull 51: Spangenberg S, Mikkelsen KL, Kines P Efficiency in reducing lost-time injuries of a nurse-based and a first-aid-based on-site medical facility. Scand J Work Env Health 31 Suppl 2: Statistics Denmark Danish Industrial Classification of All Economic Activities Second Edition. Copenhagen Statistics Denmark. Statistics Denmark Occupational accidents: Workforce study (In Danish). Labour Market News 43:1 7. Statistics Denmark Workforce for age by area, age and gender (In Danish). RAS33 Statistics Denmark, Denmark. Tüchsen F, Andersen O, Olsen J Referral bias in health personnel in studies using hospitalization as a proxy of the underlying incidence rate. J Clin Epid 49: Tyler KL, Jackson LL. 2003a. Hospitalized occupational injuries and illnesses treated in the United States emergency departments. The 131st Annual Meeting (November 15 19, 2003) of APHA, Abstract Tyler KL, Jackson LL. 2003b. Occupational injury events leading to hospitalization. NOIRS 2003: National Occupational Research Symposium 2000, Pittsburgh, Pennsylvania, USA. Waller JA, Skelly JM, Davis JH Treated injuries in northwestern Vermont. Acc Anal Prevent 27(6): Wilén H Ageing workforce how old are Europe s human resources in science and technology? Eurostat, Luxembourg. Wohl AR, Morgenstern H, Kraus JF Occupational injury in female aerospace workers. Epidem 6(2): World Health Organization International Classification of Diseases, 10 th Revision. Geneva, Switzerland: World Health Organization. Zwerling C, Sprince NL, Ryan J, Jones MP Occupational injuries Comparing the rates of male and female postal workers. Am J Epidem 138(1):46 55.