Abstract
We previously assessed mortality among U.S. Coast Guard (USCG) marine inspectors (inspectors) and Coast Guard officers who were not marine inspectors (noninspectors). Here, we extended follow-up of the cohort by 14 years, ascertaining vital status 1980–1994, calculating standardized mortality ratios (SMRs) for inspectors and noninspectors, and comparing mortality rates via directly adjusted rate ratios (RRs). Both inspectors and noninspectors had deficits for all causes of death (SMR = 75 and 61, respectively) and all malignant neoplasms (SMR = 86 and 69, respectively). Compared with noninspectors, inspectors had nonstatistically significant excesses of liver cirrhosis (SMR = 124; RR = 2.2) and chronic rheumatic heart disease (SMR = 129; RR = 2.6) and deficits of cancer of the respiratory system (SMR = 59; RR = 0.8). SMRs and RRs rose with increasing probability of exposure to chemicals for cirrhosis of the liver, all accidents and motor vehicle accidents, although they fell for all causes of death, diseases of the nervous system, diseases of the circulatory system, and cancers of the respiratory system. These results suggest that contact with chemicals during inspection of merchant vessels may be involved in the development of these diseases, although other aspects of the job, such as physical activity may account for deficits in respiratory cancers.
INTRODUCTION
The U.S. Coast Guard (USCG) is involved in the inspection of merchant vessels in the United States and throughout the world. Much of this work involves internal inspection of barges and ships where marine inspectors may enter cargo tanks, void spaces, and pump rooms for the purpose of ascertaining the integrity of shipboard hull, machinery, and equipment.1 Marine inspectors, therefore, may be exposed to a variety of chemicals, including acrylonitrile, benzene, carbon tetrachloride, ethylene dibromide, ethylene dichloride, gasoline, styrene, toluene diisocyanate, tetrachloro-ethylene, and trichloroethylene, vinyl chloride, chlorinated hydrocarbons, and others that are known or suspect human carcinogens.1,2
In the 1970s, the National Cancer Institute (NCI) and the USCG collaborated on a study to evaluate the mortality experience of USCG marine inspectors.3 A cohort of marine inspectors who performed inspection duties between 1942 and 1970 and suitable referents were assembled and traced to January 1, 1980, to determine vital status and underlying causes of death. Marine inspectors were found to have higher mortality rates than other USCG officers for several diseases. Of particular interest were excesses for cirrhosis of the liver, motor vehicle accidents, and cancers of the lymphatic and hematopoietic system (i.e., leukemia and lymphoma) because these outcomes have been associated with exposure to solvents in other studies.4–24
We extended the follow-up of these subjects through December 31, 1994 to identify deaths that have occurred since 1980. This additional follow-up allowed for a further evaluation of the mortality experience of the USCG marine inspectors with larger numbers of deaths.
METHODS
Study Population
Details of the study design have been described previously.3 In brief, the cohort included 1,767 marine inspectors who first entered the Coast Guard between 1942 and 1970. For comparison, we selected a group of 1,914 Coast Guard officers who never held marine inspection duties, matched to marine inspectors on year upon entering the Coast Guard and on rank. Marine inspectors and the other Coast Guard personnel included in the study were identified from annual registries of Coast Guard officers and enlisted personnel. Complete work histories for marine inspectors were also accessed.
Vital status for the cohort through January 1, 1980 was ascertained via records from the Coast Guard personnel office, the Veterans Administration, Social Security Administration, credit bureaus, and motor vehicle departments. Vital status from January 1980 through December 31, 1994 was ascertained via the National Death Index (NDI). For those who were deceased up to 1980, death certificates were obtained, and underlying causes of death were determined by an experienced nosologist, who used the rules in effect at the time of death and who assigned a rubric according to the 8th revision of the International Classification of Diseases. For deaths occurring after January 1, 1980, underlying and contributing causes were provided by the NDI and also assigned rubrics of the 8th revision.
Statistical Analyses
The mortality experience of inspectors and noninspectors was compared to that of the general U.S. population using standardized mortality ratios (SMRs).25 Person-year accumulation in the cohort began on January 1, 1942 (the initial year of cohort identification) or when subjects first entered the Coast Guard (if after January 1, 1942). Expected numbers for SMRs were calculated by applying 5 year age and calendar-year morality rates from the appropriate race-sex group (although this cohort included only male subjects) of the U.S. population to the person-year distribution of marine inspectors and referents. Ninety-five percent confidence intervals (95% CIs) were calculated, and a χ2 test was used to evaluate statistical significance of SMR trends.26 We also calculated mortality rates for inspectors and noninspectors, directly adjusted to the age and calendar-time person-year distribution of the combined cohorts to provide directly adjusted rate ratios (mortality rate among inspectors/mortality rate among noninspectors × 100) and 95% CIs.
Exposure Assessment Analysis
Because marine inspectors inspect ships/barges that transport a variety of products, we could not assess exposure to specific chemicals. Estimated level of possible exposure to chemicals in general was assessed using a four-level coding system, based on description of job duties, duty station, and rank. Not all duties of marine inspectors bring them in contact with potentially hazardous exposures. Marine inspectors may rotate through these different duties while assigned to one duty station, but those duties may change when they are transferred to another station. The duty station may also affect the potential for exposure because the proportion and mix of products conveyed by ships and barges varies somewhat by port. For example, ports along the Gulf Coast are major centers for the commercial shipping of petrochemicals, although other ports are not. Rank may also indicate intensity of exposure in that marine inspectors with the most intense inspection schedule and greatest probability of exposure, are usually those of lower rank, while higher ranks are more likely to have noninspection, administrative duties.
The four-point rating scale (no exposure, low, moderate, and high) was assigned as follows. Nonexposed persons generally held administrative positions or were noninspectors. Low exposure was assigned to marine inspectors who were only occasionally involved in vessel inspections. Moderate exposure was assigned to those whose inspection duties did not regularly include hull structures or internal examinations (internals) or regular inspection of hull structures/internals in geographic areas where chemicals/petrochemicals were not major items of cargo. High exposure was assigned to persons who performed hull inspections at ports where vessels transported chemicals. For each marine inspector, this four-point scale was calculated for each duty station, multiplied by number of months at that duty station, and was then summed across all marine inspection jobs held, to yield a cumulative exposure intensity (unitless) value. For each cause of death for which there were at least 5 deaths among the unexposed referent group, we calculated RRs and 95% CIs across exposure intensity levels, stratifying low and high on the basis of the median value and comparing high and low to never-exposed officers.
RESULTS
Of the 3,681 men included in the study, 1,767 (48%) were marine inspectors and 1,914 (52%) were noninspectors (Table I). The cohort was mostly white men (91%); race was unknown for 323 subjects (9%). In the analyses, those of unknown race were considered to be white, on the basis of the ethnic make-up of marine inspectors in the Coast Guard from 1942 to 1970. Through 1994, there were 1,481 deaths (775 among marine inspectors and 711 among nonmarine inspectors). Of the 3,681 Coast Guard officers in this cohort, 198 (5%) were lost to follow-up, distributed fairly evenly between marine inspectors (4%) and nonmarine inspectors (7%). There were no major differences between marine inspectors and noninspectors regarding year commissioned in the Coast Guard, year of birth, or vital status as of end of follow-up.
TABLE I.
Demographic Characteristics of Marine Inspectors and Noninspectors: January 1, 1942 to December 31, 1994
| Marine Inspectors | Noninspectors | |||
|---|---|---|---|---|
| N | (%) | N | (%) | |
| Total | 1,767 | (48) | 1,914 | (52) |
| Year CG Commission | ||||
| <1945 | 630 | (36) | 795 | (42) |
| 1945–1959 | 551 | (31) | 462 | (24) |
| ≥1960 | 493 | (28) | 518 | (27) |
| Unknown | 93 | (5) | 139 | (7) |
| Year of Birth | ||||
| <1910 | 561 | (32) | 441 | (23) |
| 1910–1929 | 712 | (40) | 1,018 | (53) |
| ≥1930 | 494 | (28) | 455 | (24) |
| Vital Status as of December 31,1994 | ||||
| Alive | 922 | (52) | 1,075 | (56) |
| Deceased | 775 | (44) | 711 | (37) |
| Lost to follow-up | 70 | (4) | 128 | (7) |
| Race | ||||
| White | 1,730 | (47) | 1,628 | (44) |
| Missing data | 37 | (1) | 286 | (8) |
Mortality from nonneoplastic causes of death is shown inTable II . There were significant deficits for all causes of death combined among both marine inspectors (SMR = 75) and noninspectors (SM = 61) compared to the general U.S. population. There were deficits for many specific causes of death, and they were usually greater among noninspectors. Significant deficits of mortality for both groups were found for allergic, endocrine, metabolic, and nutritional diseases, diabetes mellitus, all diseases of the circulatory system, arteriosclerotic heart disease, vascular lesions of the CNS, all respiratory diseases, and all diseases of the genito-urinary system. There were no statistically significant elevated SMRs for either group; however, nonstatistically significant elevated SMRs were found among marine inspectors for chronic rheumatic heart disease (SMR = 129) and cirrhosis of the liver (SMR = 123), although correspondingly noninspectors showed deficits of death from these causes (SMRs = 40 and 45, respectively). For most causes of death the directly adjusted rate ratios (RRs) for marine inspectors compared with noninspectors were greater than one, although SMRs were often less than unity for both marine inspectors and noninspectors. Significant elevated directly adjusted RRs for marine inspectors occurred for all causes of death (RR = 1.2; 95% CI, 1.1–1.3), all diseases of the circulatory system (RR = 1.2; 95% CI, 1.0–1.4), arteriosclerotic heart disease (RR = 1.2; 95% CI, 1.0–1.5), and diseases of the digestive system (RR = 1.9; 95% CI, 1.0–2.9).
TABLE II.
Mortality from Nonneoplastic Diseases among Marine Inspectors and Noninspectors
| Noninspectorsa | Marine Inspectorsb | Directly Adjusted RR |
95% CI | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ICD-8 Code | Cause of Death | OBS | EXP | SMR | 95% CI | p-χ2 | OBS | EXP | SMR | 95% CI | p-χ2 | ||||
| 001–998 | All Causes of Death | 706 | 1148.65 | 61 | (57–66) | <0.01 | * | 775 | 1029.11 | 75 | (70–81) | <0.01 | * | 1.2 | (1.1–1.3) |
| 000–139 | Infective and Parasitic Diseases | 2 | 17.39 | 12 | (1–42) | <0.01 | * | 5 | 14.65 | 34 | (11–177) | 0.02 | 2.6 | (0.5–13.4) | |
| 010–019 | Tuberculosis | 0 | 7.00 | 0 | · | · | 4 | 5.78 | 69 | (19–177) | 6.64 | · | · | ||
| 240–279 | Allergic, Endocrine, Metabolic, and Nutritional Diseases | 9 | 21.39 | 42 | (19–80) | 0.01 | * | 10 | 18.50 | 54 | (26–99) | 0.06 | * | 1.2 | (0.5–3.2) |
| 250 | Diabetes Mellitus | 6 | 17.50 | 34 | (13–75) | 0.01 | * | 6 | 15.37 | 39 | (14–85) | 0.02 | * | 1.2 | (0.4–3.8) |
| 280–289 | All Diseases Of Blood and Blood-Forming Organs | 4 | 2.89 | 139 | (37–355) | 0.72 | 1 | 2.52 | 40 | (1–220) | 0.52 | 0.2 | (0.1–2.2) | ||
| 290–317 | Mental, Psychoneurotic, and Personality Disorders | 7 | 7.82 | 90 | (36–184) | 0.92 | 5 | 6.11 | 82 | (26–191) | 0.81 | 0.9 | (0.3–2.9) | ||
| 320–389 | All Diseases of Nervous System | 11 | 14.13 | 78 | (39–139) | 0.48 | 7 | 10.80 | 65 | (26–134) | 0.31 | 0.7 | (0.3–1.9) | ||
| 390–458 | All Diseases of Circulatory System | 307 | 559.47 | 55 | (49–61) | <0.01 | * | 363 | 521.36 | 70 | (63–77) | <0.01 | * | 1.2 | (1.0–1.4) |
| 393–398 | Chronic Rheumatic Heart Disease | 3 | 7.46 | 40 | (8–117) | 0.15 | 8 | 6.20 | 129 | (56–254) | 0.60 | 2.6 | (0.7–10.1) | ||
| 410–414 | Arteriosclerotic Heart Disease | 216 | 383.24 | 56 | (49–64) | <0.01 | * | 261 | 351.08 | 74 | (66–84) | <0.01 | * | 1.2 | (1.0–1.5) |
| 430–438 | Vascular Lesions of the CNS | 40 | 70.16 | 57 | (41–78) | <0.01 | * | 49 | 71.47 | 69 | (51–91) | 0.01 | * | 1.1 | (0.7–1.8) |
| 460–519 | All Respiratory Diseases | 49 | 89.61 | 55 | (40–72) | <0.01 | * | 50 | 77.21 | 65 | (48–85) | <0.01 | * | 1.1 | (0.8–1.8) |
| 480–486 | Pneumonia | 18 | 32.60 | 55 | (33–87) | 0.01 | * | 27 | 29.46 | 92 | (60–133) | 0.72 | 1.7 | (0.9–3.2) | |
| 492 | Emphysema | 7 | 14.32 | 49 | (20–101) | 0.07 | 8 | 13.17 | 61 | (26–120) | 0.20 | 1.1 | (0.4–3.2) | ||
| 493 | Asthma | 2 | 1.90 | 105 | (12–380) | 0.78 | 3 | 1.95 | 154 | (31–451) | 0.69 | 1.9 | (0.3–11.5) | ||
| 520–577 | Diseases of the Digestive System | 26 | 48.05 | 54 | (35–79) | <0.01 | * | 38 | 41.12 | 92 | (65–127) | 0.68 | 1.7 | (1.0–2.9) | |
| 531–533 | Gastric and Duodenal Ulcer | 3 | 6.09 | 49 | (10–144) | 0.29 | 5 | 5.88 | 85 | (27–198) | 0.86 | 1.9 | (0.4–8.2) | ||
| 571 | Cirrhosis of Liver | 10 | 22.07 | 45 | (22–83) | 0.01 | * | 22 | 17.88 | 123 | (77–186) | 0.39 | 2.2 | (0.9–5.2) | |
| 580–629 | All Diseases of the Genito-Urinary System | 8 | 17.52 | 46 | (20–90) | 0.03 | * | 7 | 16.26 | 43 | (17–89) | 0.03 | * | 1.0 | (0.4–2.9) |
| 582 | Chronic Nephritis | 0 | 3.32 | 0 | · | · | 1 | 3.04 | 33 | (0–183) | 0.38 | · | · | ||
| 680–709 | All Diseases of the Skin and Cellular Tissue | 0 | 1.00 | 0 | · | · | 1 | 0.87 | 116 | (2–643) | 0.70 | · | · | ||
| 710–738 | All Diseases of the Bones and Organs of Movement | 1 | 2.03 | 49 | (1–274) | 0.71 | 1 | 1.67 | 60 | (1–334) | 0.89 | 0.8 | (0.1–12.6) | ||
| 780–799 | Symptoms, Senility, and Ill-Defined Conditions | 16 | 10.60 | 151 | (86–245) | 0.13 | 8 | 9.17 | 87 | (38–172) | 0.82 | 0.7 | (0.3–1.7) | ||
| 800–998 | All External Causes of Death | 43 | 81.13 | 53 | (38–71) | <0.01 | * | 45 | 59.70 | 75 | (55–101) | 0.07 | 1.2 | (0.8–1.9) | |
| 800–949 | All Accidents | 25 | 55.54 | 45 | (29–66) | <0.01 | * | 33 | 39.71 | 83 | (57–117) | 0.32 | 1.5 | (0.9–2.6) | |
| 810–827 | Motor Vehicle Accidents | 12 | 24.19 | 50 | (26"87) | 0.02 | * | 15 | 16.47 | 91 | (51–150) | 0.81 | 1.3 | (0.6–2.8) | |
| 950–959 | Suicide | 16 | 19.55 | 82 | (47–133) | 0.49 | 9 | 15.31 | 59 | (27–112) | 0.14 | 0.7 | (0.3–1.6) | ||
Total persons = 1,863 (1,905, including missing cause); total person years = 72,475.54 (73,540.79, including missing cause).
Total persons = 1,721 (1,767, including missing cause); total person years = 50,756.79 (51,782.55, including missing cause).
p < 0.05.
OBS, observed; EXP, expected; SMR, Standardized Mortality Ratio; Directly Adjusted RR, Directly Adjusted Rate Ratio.
Mortality from neoplastic causes of death is shown inTable III . There were significant deficits for death from all malignant neoplasms among both marine inspectors (SMR = 86) and noninspectors (SMR = 69). Directly adjusted RRs for marine inspectors compared to noninspectors revealed elevated RRs for all cancers (RR = 1.2; 95% CI, 1.0–1.5). Statistically significant deficits were found for both marine inspectors and noninspectors for cancers of the respiratory system (SMR = 59 and SMR = 71, respectively; RR = 0.8; 95% CI, 0.5–1.2), particularly for lung cancer (SMR = 61 and SMR =71, respectively; RR = 0.8; 95% CI, 0.6–1.2). Nonstatistically significant deficits were found for both groups for cancers of the digestive system; however, a comparison of the mortality rates between inspectors and noninspectors revealed a slight, but statistically significant excess risk among inspectors for death from cancer of the digestive organs (RR = 1.2; 95% CI, 1.1–2.4). There were no statistically significant elevated SMRs for either group. Cancers that were elevated among marine inspectors but not among noninspectors were cancers of the esophagus (SMRs = 117 and 66, respectively; RR = 1.8; 95% CI, 0.5–6.4), rectum (SMRs = 104 and 33, respectively; RR = 3.5; 95% CI, 0.7–17.8), skin (SMRs = 157 and 65, respectively; RR = 1.9; 95% CI, 0.4–8.0), brain and CNS (SMRs = 161 and 94, respectively; RR = 1.4; 95% CI, 0.5–4.4), lymphatic and hematopoietic system (SMRs = 110 and 57, respectively; RR = 1.6; 95% CI, 0.7–10.8), in particular non-Hodgkin’s lymphoma (SMRs = 104 and 35, respectively; RR = 2.8; 95% CI, 0.7–10.8), and leukemia (SMRs = 109 and 82, respectively; RR = 1.2; 95% CI, 0.8–3.2). Among inspectors, liver cancer was elevated (SMR = 141; ncases = 6 vs. 4.27 expected); however, among noninspectors there were no reported cases, so an SMR for noninspectors and thus a directly adjusted RR could not be calculated.
TABLE III.
Mortality from Cancer among Marine Inspectors and Noninspectors
| Noninspectors | Marine Inspectors | Directly Adjusted RR |
95% CI | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ICD-8 Code | Cancer Cause of Death | OBS | EXP | SMR | 95% CI | p-χ2 | OBS | EXP | SMR | 95% CI | p-χ2 | ||||
| 140–209 | All Malignant Neoplasms | 177 | 256.52 | 69 | (59–80) | <0.01 | * | 185 | 216.04 | 86 | (74–99) | 0.32 | * | 1.2 | (1.0–1.5) |
| 140–149 | Buccal Cavity and Pharynx | 1 | 6.44 | 16 | (0–86) | 0.05 | * | 4 | 5.57 | 72 | (19–184) | 0.32 | 3.6 | (0.4–32.9) | |
| 150–159 | Digestive Organs | 41 | 66.34 | 62 | (44–84) | <0.01 | * | 54 | 59.14 | 91 | (69–119) | 0.32 | 1.2 | (1.1–2.4) | |
| 150 | Esophagus | 4 | 6.06 | 66 | (18–169) | 0.53 | 6 | 5.14 | 117 | (43–254) | 0.86 | 1.8 | (0.5–6.4) | ||
| 151 | Stomach | 6 | 10.38 | 58 | (21–126) | 0.23 | 6 | 10.23 | 59 | (21–128) | 0.24 | 1.3 | (0.4–4.0) | ||
| 153 | Colon | 17 | 23.99 | 71 | (41–113) | 0.18 | 21 | 20.72 | 101 | (63–155) | 1.00 | 1.5 | (0.8–2.9) | ||
| 154 | Rectum | 2 | 6.06 | 33 | (4–119) | 0.15 | 6 | 5.79 | 104 | (38–225) | 0.92 | 3.5 | (0.7–17.8) | ||
| 155–156 | Liver | 0 | 4.67 | 0 | · | · | · | 6 | 4.27 | 141 | (51–306) | 0.55 | · | · | |
| 157 | Pancreas | 11 | 12.87 | 85 | (43–153) | 0.70 | 9 | 11.12 | 81 | (37–154) | 0.63 | 1 | (0.4–2.4) | ||
| 160–163 | Respiratory System | 64 | 89.84 | 71 | (55–91) | <0.01 | * | 43 | 72.40 | 59 | (43–80) | <0.01 | * | 0.8 | (0.5–1.2) |
| 161 | Larynx | 2 | 3.31 | 60 | (7–218) | 0.660 | 1 | 2.86 | 35 | (0–195) | 0.42 | 0.3 | (0.1–2.9) | ||
| 162 | Lung | 61 | 85.72 | 71 | (54–91) | <0.01 | * | 42 | 68.86 | 61 | (44–82) | <0.01 | * | 0.8 | (0.6–1.2) |
| 170 | Bone | 1 | 0.82 | 123 | (2–682) | 0.73 | · | 0 | 0.72 | 0 | · | · | 0 | · | |
| 172–173 | Skin | 3 | 4.64 | 65 | (13–189) | 0.60 | 6 | 3.82 | 157 | (57–342) | 0.39 | 1.9 | (0.4–8.0) | ||
| 174 | Breast | 0 | 0.32 | 0 | · | · | 0 | 0.28 | 0 | · | · | · | · | ||
| 185 | Prostate | 16 | 23.71 | 67 | (39–110) | 0.14 | 20 | 20.98 | 95 | (58–147) | 0.92 | 1.5 | (0.7–2.9) | ||
| 186–187 | Testicular | 0 | 0.95 | 0 | · | · | 0 | 0.65 | 0 | · | · | · | · | ||
| 188 | Bladder | 6 | 7.59 | 79 | (29–172) | 0.69 | 4 | 6.99 | 57 | (15–147) | 0.35 | 0.6 | (0.2–2.1) | ||
| 189 | Kidney | 8 | 6.20 | 129 | (56–254) | 0.60 | 8 | 5.18 | 155 | (67–305) | 0.31 | 1.1 | (0.4–3.1) | ||
| 190 | Eye | 1 | 0.18 | 573 | (7–3,185) | 0.43 | 0 | 0.16 | 0 | · | · | 0 | · | ||
| 191–192 | Brain and CNS | 6 | 6.41 | 94 | (34–204) | 1.00 | 8 | 4.97 | 161 | (69–317) | 0.26 | 1.4 | (0.5–4.4) | ||
| 193 | Thyroid | 1 | 0.43 | 232 | (3–1,290) | 0.92 | 0 | 0.38 | 0 | · | · | 0 | · | ||
| 200–209 | All Lymphatic and Hematopoietic Cancer | 14 | 24.55 | 57 | (31–96) | 0.04 | * | 22 | 20.07 | 110 | (69–166) | 0.75 | 1.6 | (0.7–10.8) | |
| 200, 202 | Non-Hodgkin’s Lymphoma | 3 | 8.33 | 36 | (7–105) | 0.33 | 7 | 6.74 | 104 | (42–214) | 0.34 | 2.8 | (0.7–10.8) | ||
| 201 | Hodgkin’s Disease | 0 | 1.90 | 0 | · | · | 1 | 1.43 | 70 | (1–389) | 1.00 | · | (0.3–9.9) | ||
| 203 | Multiple Myeloma | 2 | 3.71 | 54 | (6–195) | 0.53 | 3 | 3.04 | 99 | (20–288) | 0.79 | 1.6 | (0.5–3.3) | ||
| 204–207 | Leukemia | 8 | 9.74 | 82 | (35–162) | 0.69 | 9 | 8.23 | 109 | (50–208) | 0.92 | 1.2 | (0.8–3.2) | ||
Total persons = 1,863 (1,905, including missing cause); total person years = 72,475.54 (73,540.79, including missing cause).
Total persons = 1,721 (1,767, including missing cause); total person years = 50,756.79 (51,782.55, including missing cause).
p < 0.05.
OBS, observed; EXP, expected; SMR, Standardized Mortality Ratio; Directly Adjusted RR, Directly Adjusted Rate Ratio.
Rate ratios for mortality from selected nonneoplastic causes of death by cumulative level of exposure—the product of a four-point rating scale and months at duty station, summed across all duty stations—are shown inTable IV . Noninspectors and some marine inspectors, i.e., those unexposed in their marine inspector postings (n = 104) comprise the nonexposed group, although marine inspectors with a cumulative exposure score at or below the median (130) comprise the low-exposed group, and those with a cumulative exposure score above the median comprise the high group. The causes of death selected for this table are those for which there were at least 5 deaths in the referent category (nonexposed). For all causes of death there were significantly lower risks among the low-exposed (RR = 0.78; 95% CI, 0.68–0.90) and high-exposed (RR = 0.80; 95% CI, 0.71–0.91) marine inspectors; test for trend was significant (p < 0.01). No statistically significant RRs were found for specific nonneoplastic causes of death. Rate ratios for both the low- and high-exposure categories among marine inspectors were elevated for cirrhosis of the liver (RRlow = 1.95; RRhigh = 2.32; p-trend = 0.06), all accidents (RRlow = 1.49, RRhigh = 1.56; p-trend = 0.17), and motor vehicle accidents (RRlow = 1.55; RRhigh = 2.05; p-trend = 0.15), though none of these estimates was statistically significant. Rate ratios for both the low- and high-exposure categories among marine inspectors showed a nonstatistically significant reduction for suicides (RRlow = 0.53; RRhigh = 0.41; p-trend = 0.09).
TABLE IV.
Rate Ratios for Selected* Nonneoplastic Causes of Death by Level of Exposure (Probabilty of Exposure × Months of Duration)
| Nonexposed a | Low Exposed (<130) b | High Exposed (>130) c | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ICD-8 Code | Cause of Death | n | RR | n | RR | 95% CI | n | RR | 95% CI | p -trend |
| 001–998 | All Causes of Death | 781 | 1.00 | 295 | 0.78 | (0.68–0.90) | 405 | 0.80 | (0.71–0.91) | <0.01 |
| 240–279 | Allergic, Endocrine, Metabolic, Nutritional Diseases | 10 | 1.00 | 6 | 1.03 | (0.37–2.92) | 3 | 0.38 | (0.10–1.42) | 0.18 |
| 250 | Diabetes Mellitus | 6 | 1.00 | 3 | 0.83 | (0.20–3.39) | 3 | 0.61 | (0.15–2.55) | 0.51 |
| 290–317 | Mental, Psychoneurotic, and Personality Disorders | 7 | 1.00 | 0 | . | . | 5 | 1.29 | (0.38–4.38) | 0.80 |
| 320–389 | All Diseases of Nervous System and Sense Organs | 12 | 1.00 | 4 | 0.64 | (0.21–2.01) | 2 | 0.22 | (0.05–0.97) | 0.04 |
| 390–458 | All Diseases of Circulatory System | 349 | 1.00 | 132 | 0.72 | (0.59–0.89) | 189 | 0.76 | (0.63–0.92) | <0.01 |
| 410–414 | Arteriosclerotic Heart Disease | 243 | 1.00 | 95 | 0.77 | (0.61–0.98) | 139 | 0.83 | (0.67–1.03) | 0.07 |
| 430–438 | All Vascular Lesions of CNS | 46 | 1.00 | 22 | 0.82 | (0.49–1.38) | 21 | 0.58 | (0.34–0.99) | 0.05 |
| 460–519 | All Respiratory Diseases | 53 | 1.00 | 18 | 0.66 | (0.38–1.14) | 28 | 0.75 | (0.47–1.22) | 0.21 |
| 480–486 | All Pneumonia | 21 | 1.00 | 8 | 0.68 | (0.29–1.56) | 16 | 0.99 | (0.50–1.95) | 0.94 |
| 492 | Emphysema | 7 | 1.00 | 4 | 0.99 | (0.28–3.43) | 4 | 0.73 | (0.21–2.55) | 0.63 |
| 520–577 | All Diseases of Digestive System | 29 | 1.00 | 17 | 1.23 | (0.67–2.29) | 18 | 1.00 | (0.54–1.86) | 0.94 |
| 571 | Cirrhosis of Liver | 10 | 1.00 | 9 | 1.95 | (0.78–4.91) | 13 | 2.32 | (0.97–5.54) | 0.06 |
| 580–629 | All Diseases of Genito-Urinary System | 9 | 1.00 | 1 | 0.20 | (0.03–1.61) | 5 | 0.69 | (0.23–2.13) | 0.44 |
| 780–799 | Symptoms, Senility, and Ill-Defi ned Conditions | 16 | 1.00 | 3 | 0.51 | (0.15–1.79) | 5 | 0.68 | (0.23–1.95) | 0.38 |
| 800–998 | All External Causes of Death | 46 | 1.00 | 20 | 1.21 | (0.70–2.10) | 22 | 1.10 | (0.63–1.91) | 0.68 |
| 800–949 | All Accidents | 27 | 1.00 | 14 | 1.49 | (0.75–2.95) | 17 | 1.56 | (0.80–3.04) | 0.17 |
| 810–827 | Motor Vehicle Accidents | 12 | 1.00 | 6 | 1.55 | (0.55–4.38) | 9 | 2.05 | (0.77–5.43) | 0.15 |
| 950–959 | Suicide | 17 | 1.00 | 4 | 0.53 | (0.18–1.62) | 4 | 0.41 | (0.13–1.26) | 0.09 |
The causes of death selected for this table are those for which there were at least 5 deaths in the referent category (nonexposed).
Total number of nonexposed = 2,011 (44 of whom had missing cause of death; 104 of whom were marine inspectors with no exposure).
Total number of low exposed = 827 (19 of whom had missing cause of death).
Total number of high exposed = 834 (25 of whom had missing cause of death).
For neoplastic causes of death (Table V), there were statistically significant deficits and downward trends for cancer of the respiratory system (RRhigh = 0.52; 95% CI, 0.31–0.88; p-trend = 0.01), in particular, lung cancer (RRhigh = 0.57; 95% CI, 0.34– 0.96; p-trend = 0.03) with increasing exposure category.
TABLE V.
Rate Ratios for Selected* Neoplastic Causes of Death by Level of Exposure (Probabilty of Exposure × Months of Duration)
| Nonexposed | Low Exposed (<130) | High Exposed (>130) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ICD-8 Code | Disease | n | RR | n | RR | 95% CI | n | RR | 95% CI | p- trend |
| 140–209 | All Malignant Neoplasms | 194 | 1.00 | 68 | 0.77 | (0.58–1.03) | 100 | 0.87 | (0.68–1.13) | 0.23 |
| 150–159 | Digestive Organs | 45 | 1.00 | 21 | 1.04 | (0.1–1.76) | 29 | 1.08 | (0.66–1.77) | 0.75 |
| 151 | Stomach | 6 | 1.00 | 3 | 1.00 | (0.24–4.09) | 3 | 0.72 | (0.17–3.02) | 0.68 |
| 153 | Colon | 19 | 1.00 | 12 | 1.42 | (0.67–2.99) | 7 | 0.64 | (0.26–1.57) | 0.45 |
| 160–163 | Respiratory System | 68 | 1.00 | 19 | 0.65 | (0.39–1.09) | 20 | 0.52 | (0.31–0.88) | 0.01 |
| 162 | Lung | 64 | 1.00 | 19 | 0.71 | (0.42–1.20) | 20 | 0.57 | (0.34–0.96) | 0.03 |
| 185 | Prostate | 18 | 1.00 | 10 | 1.00 | (0.45–2.21) | 8 | 0.60 | (0.25–1.41) | 0.27 |
| 188 | Bladder | 6 | 1.00 | 1 | 0.31 | (0.03–2.69) | 3 | 0.71 | (0.16–3.11) | 0.60 |
| 189 | Kidney | 8 | 1.00 | 2 | 0.53 | (0.11–2.55) | 6 | 1.18 | (0.39–3.59) | 0.82 |
| 191–192 | Brain and CNS | 6 | 1.00 | 5 | 1.85 | (0.54–6.29) | 3 | 0.90 | (0.21–3.82) | 1.00 |
| 200–209 | All Lymphatic and Hematopoietic Cancer | 18 | 1.00 | 6 | 0.62 | (0.24–1.59) | 12 | 0.93 | (0.43–1.99) | 0.80 |
| 204–207 | Leukemia | 9 | 1.00 | 3 | 0.65 | (0.17–2.48) | 5 | 0.84 | (0.27–2.66) | 0.73 |
The causes of death selected for this table are those for which there were at least 5 deaths in the referent category (nonexposed).
We carried out a stratified analysis by geographic region, comparing marine inspectors to noninspectors. We found that those marine inspectors who had ever carried out marine inspector activities in the Gulf Coast (where there is a higher likelihood of carrying out inspections on petrochemical platforms) had a higher risk of cirrhosis of the liver (RR = 2.60; 95% CI, 0.98–6.89) than those who had carried out duties in non-Gulf Coast locations (RR = 1.69; 95% CI, 0.59–4.82), when compared to noninspectors. Similar trends were seen for death from external causes (Gulf: RR = 1.49; non-Gulf: RR = 0.89), accidents (Gulf: RR = 2.08; non-Gulf: RR = 1.12), and motor vehicle accidents (Gulf: RR = 1.75; non-Gulf: RR = 1.35). None of these estimates was statistically significant (data not shown). We also stratified by rank at first job because of the possibility that marine inspectors who started their duties as a lower ranking officer would have had more field inspection duties (thus, more adverse exposures) than would marine inspectors who started their duties as a higher ranking officer (higher likelihood of administrative jobs). We found that those marine inspectors who had started their duties at a rank lower than lieutenant had a slightly higher (nonstatistically significant) risk for cirrhosis of the liver (RR = 2.09; 95% CI, 0.85–5.12) than those who had started their duties at the rank of lieutenant or higher (RR = 1.90; 0.69–5.22) when compared to noninspectors (data not shown).
DISCUSSION
We compared the mortality experience of Coast Guard marine inspectors with the general U.S. population and group of Coast Guard officers not engaged in marine inspections to assess risks from potential exposure to chemicals during performance of their inspection duties. We found that both marine inspectors and noninspectors had a significantly lower mortality from all causes and many individual causes of death than the general U.S. population. This might be expected because of the socio-demographic characteristics of this cohort and their access to excellent health care. Service in the military not only insured access to care during their time as Coast Guard officers, but also qualified them for care and discounted medications through the Department of Veterans Affairs after their retirement or separation from service. The deficit might also be partially because of the healthy worker effect, which is seen in most studies of this design, although this effect typically diminishes with length of follow-up and this cohort has been followed for about 50 years.27
Both marine inspectors and noninspectors had low SMRs for most causes of death. Evaluating our results collectively, i.e., from SMRs, directly adjusted RRs, SMRs and RRs comparing low- and high-exposure groups to the nonexposed group, and RRs for inspectors with a history of jobs in the Gulf Coast region and who started their inspection duties at a lower rank, we found fairly consistently increased risks for death from cirrhosis of the liver, all accidents, and motor vehicle accidents, and decreased risks for death from cancer of the respiratory system, in particular, lung cancer. We found some similarities with the earlier report, as well as some differences.3 In both studies, marine inspectors had larger SMRs for many causes of death than noninspectors. The previous study found elevations among marine inspectors for cirrhosis of the liver, motor vehicle accidents, cancers of the colon, liver, skin, and lymphohematopoitic system (accounted for mainly by the mortality pattern for leukemia). We did not find strong evidence of increased risk for colon cancer or leukemia; however we did find elevated SMRs for cirrhosis of the liver and cancers of the liver, skin, and lymphohematopoietic system.
For cirrhosis of the liver we found an elevated SMR among marine inspectors, a decreased SMR among noninspectors, an elevated directly adjusted RR, and elevated RRs with increasing cumulative exposures that might occur while working as a marine inspector. We also found higher risks among inspectors working in ports where chemicals are more likely encountered, and slightly higher risks among inspectors who started out at a lower rank. None of these estimates was statistically significant. Most liver cirrhosis in the United States is the result of alcohol consumption.28 Information on alcohol consumption was not available for this study, so it was not possible to directly assess the influence alcohol might have had on this outcome. It is well known, however, that chlorinated solvents are metabolized in the liver and may induce liver cirrhosis.13 Cirrhosis of the liver has been induced experimentally in rodents by carbon tetrachloride8,11,21,22 and toluene.20 Cases following exposure have been reported in humans exposed to vinyl chloride monomer,7 carbon tetrachloride,7,10,16,17,23,24 among a historical cohort mortality study in a shipyard in Italy, which followed workers assigned to ship repair, refitting, and construction,17 among workers in tanneries, who are exposed to aromatic organic solvents,24 and among automobile mechanics and gasoline service station workers, who are exposed to benzene and gasoline.23 During vessel inspections, marine inspectors may have been exposed to many solvents, fuels, and other chemicals, such as acrylonitrile, carbon tetrachloride, ethylene dibromide, ethylene dichloride, benzene, gasoline, styrene, toluene, tetrachloroethylene, and tricholoethylene.1,2
Although both marine inspectors and noninspectors had deficits of death from all accidents and from motor vehicle accidents when compared to the general population, both were moderately, though not statistically significantly increased (RRs = 1.5 and 1.3, respectively) among inspectors compared to noninspectors and showed a nonstatistically significant exposure-response gradient with cumulative exposure among marine inspectors. The previous follow-up also found increased mortality among marine inspectors compared to noninspectors from these causes. Exposure to organic solvents has been associated with neurotoxic effects, such as dizziness, lightheadedness, and diminished coordination.4 Although most of the findings for these effects have been related to acute exposures, there is some evidence that chronic conditions occur in individuals with long-term solvent exposure.4 The lag time for reaction would be particularly relevant for motor vehicle accidents and motor vehicle deaths have been reported among workers exposed to methylene chloride.29 A significant excess of deaths by accidental causes was observed among tannery workers who were exposed to a variety of solvents.24 Accidents and motor vehicle accidents, however, are also strongly related to alcohol consumption. There are no strong reasons to believe that marine inspectors as a group would have had significantly higher alcohol intake than noninspectors, but it is not clear whether solvents, alcohol, or both might contribute to some of the excesses observed among marine inspectors.
The excess for liver cancer among inspectors is based on small numbers (6 deaths among inspectors and no deaths among noninspectors); however, the difference is striking, given there are about the same number of inspectors and noninspectors, there are about the same number of total deaths in noninspectors as inspectors, and that liver cancer is a rare disease. Although we are unable to draw any clear statistical conclusions, this is an intriguing finding which may have some grounding in biological plausibility. We compared marine inspectors with high exposure (liver cancer, n = 4) to marine inspectors with low exposure (liver cancer, n = 2) and found a nonstatistically significant elevated RR of 1.85 (95% CI, 0.33–10.49). Liver tumors have been found to develop in animals exposed to chlorinated solvents.30 A review of epidemiologic studies from several countries on the effects of vinyl chloride found a combined SMR of 5.33 (95% CI, 4.23–6.62) for liver cancer. 31 Two other reviews demonstrated slight increases in liver cancer with trichloroethylene.32,33
A factor that may contribute to the statistically significant decreasing overall mortality and mortality from cancers of the respiratory system, in particular lung cancer, among marine inspectors with increasing cumulative exposure is occupational physical activity. It is well established that physical activity is inversely and causally related to the incidence of coronary heart disease.34,35 Daily activities of climbing into and out of numerous deep cargo, ballast, or bilge tanks on a given inspection or transiting via staircase between numerous decks (levels) on a large ship require a considerable amount of exercise and energy expenditure. The level of physical activity experienced daily by marine inspectors, however, would have varied, depending on the duties experienced at different ports. Some inspections would require a lesser amount of physical activity, such as those inspecting small vessels, but over a lifetime of carrying out inspections, the general level of physical activity in marine inspectors might have been higher than that experienced by other officers in the Coast Guard.
The study has limitations. Specific exposures experienced by marine inspectors are complex and are poorly documented here. We lacked detailed information on the multiple exposures which were inevitably experienced by marine inspectors. Finally, as previously mentioned we did not have data on tobacco or alcohol use. A major strength of the study is the availability of a comparison group of Coast Guard officers who were never marine inspectors. These officers should be similar socioeconomically to marine inspectors, which should diminish differences in tobacco and alcohol use and other lifestyle factors. Both groups had the same access to health care, since they both accessed the military health care system. The cohort was followed for many years. All officers were commissioned before 1970 and about 40% of the cohort were deceased at the end of the current follow up.
Exposures that marine inspectors experience today may differ with respect to intensity and duration from what was experienced by marine inspectors in this study. The follow-up of this study ended in 1994. Much of the follow-up period for this cohort was characterized by infrequent use of protective equipment and infrequent medical surveillance. The Coast Guard implemented a personal protective equipment program during the mid 1980s, which required that all inspectors inspecting tanks in which previous cargos contained benzene wear carbon filter masks. In 1985 the Coast Guard charged safety and occupational health coordinators to establish safe work practices for marine inspectors at each local marine inspection office. Regarding medical surveillance, in the early 1980s the Coast Guard launched the Occupational Medical Monitoring Program. Evaluation of occupational exposures was carried out, and the Coast Guard developed recommended surveillance for persons working in high-risk environments with exposures to hazards such as asbestos, benzene, and lead. This program was significantly expanded over the past 2 decades and now includes 10 additional hazards. The declining number of operational U.S. flag deep draft tank/cargo ships over recent decades has reduced the number of U.S. Coast Guard marine inspections of those types of ships. Marine inspectors today may be more involved in inspections of small passenger vessels, coastal barges, and foreign flag port state control inspections, where risk of exposure is diminished.
CONCLUSIONS
Coast Guard officers have reduced mortality compared to the general U.S. population. Comparison of the mortality experience of marine inspectors with other Coast Guard officers uncovered reduced mortality among inspectors for cancers of the respiratory system, although excess mortality among inspectors from cirrhosis of the liver, all accidents, and motor vehicle accidents. These patterns suggest that occupational exposures may have been risk factors, and physical activity may have been protective. These results suggest that contact with chemicals during inspection of merchant vessels may be involved in the development of these diseases among marine inspectors, although other aspects of the job, such as physical activity may account for deficits in respiratory cancers.
ACKNOWLEDGMENTS
We thank the U.S. Coast Guard for assistance in this study, which is, in turn, supported by the Intramural Research Program of the National Institutes of Health (Division of Cancer Epidemiology and Genetics of the National Cancer Institute).
REFERENCES
- 1.Haas TJ. Safety and health hazards in inspection and response. Proc Mar Saf Counc. 1979;36:78–81. [Google Scholar]
- 2.Research Council Committee on Maritime Hazardous Materials N. Principles of Toxicological Interactions Associated with Multiple Chemical Exposuresed. Springfield, VA: National Technical Information Service; 1980. [Google Scholar]
- 3.Blair A, Haas T, Prosser R, et al. Mortality among United States Coast Guard Marine Inspectors. Arch Environ Health. 1989;44:150–156. doi: 10.1080/00039896.1989.9935879. [DOI] [PubMed] [Google Scholar]
- 4.Baker EL, Fine LJ. Solvent neurotoxicity: the current evidence. J Occup Med. 1986;28:126–129. [PubMed] [Google Scholar]
- 5.Chiu BC, Weisenburger DD. An update of the epidemiology of non-Hodgkin’s lymphoma. Clin Lymphoma. 2003;4:161–168. doi: 10.3816/clm.2003.n.025. [DOI] [PubMed] [Google Scholar]
- 6.Dryver E, Brandt L, Kauppinen T, Olsson H. Occupational exposures and non-Hodgkin’s lymphoma in Southern Sweden. Int J Occup Environ Health. 2004;10:13–21. doi: 10.1179/oeh.2004.10.1.13. [DOI] [PubMed] [Google Scholar]
- 7.Du CL, Wang JD. Increased morbidity odds ratio of primary liver cancer and cirrhosis of the liver among vinyl chloride monomer workers. Occup Environ Med. 1998;55:528–532. doi: 10.1136/oem.55.8.528. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Edwards JE, Dalton AJ. Induction of cirrhosis of the liver and hepatomas in mice with carbon tetrachloride. J Natl Cancer Inst. 1941;3:19–41. [Google Scholar]
- 9.Grulich AE, Vajdic CM. The epidemiology of non-Hodgkin lymphoma. Pathology. 2005;37:409–419. doi: 10.1080/00313020500370192. [DOI] [PubMed] [Google Scholar]
- 10.Hardin BLJ. Carbon tetrachloride poisoning: a review. Ind Med Surg. 1954;23:93–97. [PubMed] [Google Scholar]
- 11.Inamura T, Miura S, Tsuzuki Y, et al. Alteration of intestinal intraepithelial lymphocytes and increased bacterial translocation in a murine model of cirrhosis. Immunol Lett. 2003;90:3–11. doi: 10.1016/j.imlet.2003.05.002. [DOI] [PubMed] [Google Scholar]
- 12.Some industrial chemicals and dyestuffed. Vol 29. Lyon, France: IARC; 1982. International Agency for Research on Cancer: IARC monographs on the evaluation of the carcinogenic risk of chemicals to humans. [Google Scholar]
- 13.Leibman KC, Ortez E. Metabolism of halongenated ethylenes. Environ Health Perspect. 1977;21:91–97. doi: 10.1289/ehp.772191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Nilsson RI, Nordlinder R, Horte LG, Jarvholm B. Leukaemia, lymphoma, and multiple myeloma in seamen on tankers. Occup Environ Med. 1998;55:517–521. doi: 10.1136/oem.55.8.517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Persson B, Fredriksson M, Olsen K, Boeryd B, Axelson O. Some occupational exposures as risk factors for malignant lymphomas. Cancer. 1993;72:1773–1778. doi: 10.1002/1097-0142(19930901)72:5<1773::aid-cncr2820720542>3.0.co;2-6. [DOI] [PubMed] [Google Scholar]
- 16.Poindexter CA, Greene CH. Toxic cirrhosis of the liver: report of a case due to long continued exposure to carbon tetrachloride. JAMA. 1943;102:2105. [Google Scholar]
- 17.Puntoni R, Merlo F, Borsa L, Reggiardo G, Garrone E, Ceppi M. A historical cohort mortality study among shipyard workers in Genoa, Italy. Am J Ind Med. 2001;40:363–370. doi: 10.1002/ajim.1110. [DOI] [PubMed] [Google Scholar]
- 18.Rego MA. Non-Hodgkin’s lymphoma risk derived from exposure to organic solvents: a review of epidemiologic studies. Cad Saude Publica. 1998;14:41–66. doi: 10.1590/s0102-311x1998000700006. [DOI] [PubMed] [Google Scholar]
- 19.Rego MA, Sousa CS, Kato M, de Carvalho AB, Loomis D, Carvalho FM. Non-Hodgkin’s lymphomas and organic solvents. J Occup Environ Med. 2002;44:874–881. doi: 10.1097/00043764-200209000-00010. [DOI] [PubMed] [Google Scholar]
- 20.Reuber MD. Carcinomas of the liver in female mice fed toluene-2,4-diamine. Gann. 1979;70:453–457. [PubMed] [Google Scholar]
- 21.Reuber MD, Glover EL. Cirrhosis and carcinoma of the liver in male rats given subcutaneous carbon tetrachloride. J Natl Cancer Inst. 1970;44(2):419–423. [PubMed] [Google Scholar]
- 22.Rhoden EL, Pereira-Lima L, Kalil AN, et al. Effects of ischemia and reperfusion on oxidative stress in hepatic cirrhosis induced by carbon tetrachloride in rats. Kobe J Med Sci. 2000;46:171–180. [PubMed] [Google Scholar]
- 23.Schwartz E. Proportionate mortality ratio analysis of automobile mechanics and gasoline service station workers in New Hampshire. Am J Ind Med. 1987;12:91–99. doi: 10.1002/ajim.4700120110. [DOI] [PubMed] [Google Scholar]
- 24.Stern FB, Beaumont JJ, Halperin WE, Murthy LI, Hills BW, Fajen JM. Mortality of chrome leather tannery workers and chemical exposures in tanneries. Scand J Work Environ Health. 1987;13:108–117. doi: 10.5271/sjweh.2073. [DOI] [PubMed] [Google Scholar]
- 25.Boice J, Pickle L, Thomas T. Observed Versus Expected Events User’s Guide: Version 3.8 National Cancer Institute Epidemiology and Biostatistics Program. Rockville, MD: 1991. [Google Scholar]
- 26.Breslow NE, Lubin JH, Marek P, Langholz B. Multiplicative models and cohort analysis. J Am Stat Assoc. 1983;78:1–12. [Google Scholar]
- 27.McMichael A. Standardized mortality ratios and the “healthy worker effect”: scratching beneath the surface. J Occup Med. 1976;18:165–168. doi: 10.1097/00043764-197603000-00009. [DOI] [PubMed] [Google Scholar]
- 28.American Cancer Society: What Are the Risk Factors for Liver Cancer? Cancer Reference Information. 2006;Vol 2006 [Google Scholar]
- 29.Ott MG, Skory LK, Holder SS, Bronson JM, Williams PR. Health evaluation of employees occupationally exposed to methylene chloride: mortality. Scand J Work Environ Health. 1983;9 Suppl 1:876. [PubMed] [Google Scholar]
- 30.Craft B. W R. Environmental and Occupational Medicine. Boston, Little: Brown, and Company; 1983. Solvents and related compounds; pp. 511–533. [Google Scholar]
- 31.Kielhorn J, Melber C, Wahnschaffe U, Aitio A, Mangelsdorf I. Vinyl chloride: still a cause for concern. Environ Health Perspect. 2000;108:579–588. doi: 10.1289/ehp.00108579. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Weiss NS. Cancer in relation to occupational exposure to trichloroethylene. Occup Environ Med. 1996;53:1–5. doi: 10.1136/oem.53.1.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Wartenberg D, Reyner D, Scott CS. Trichloroethylene and cancer: epidemiologic evidence. Environ Health Perspect. 2000;108:161–176. doi: 10.1289/ehp.00108s2161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Powell KE, Thompson PD, Caspersen CJ, Kendrick JS. Physical activity and the incidence of coronary heart disease. Annu Rev Public Health. 1987;8:253–287. doi: 10.1146/annurev.pu.08.050187.001345. [DOI] [PubMed] [Google Scholar]
- 35.U.S. Department of Health and Human Services: Physical Activity and Health: A Report of the Surgeon General. 1996