Investing in prospective cohorts for etiologic study of occupational exposures.

Prospective cohorts have played a major role in understanding the contribution of diet, physical activity, medical conditions, and genes to the development of many diseases, but have not been widely used for occupational exposures. Studies in agriculture are an exception. We draw upon our experience using this design to study agricultural workers to identify conditions that might foster use of prospective cohorts to study other occupational settings. Prospective cohort studies are perceived by many as the strongest epidemiologic design. It allows updating of information on exposure and other factors, collection of biologic samples before disease diagnosis for biomarker studies, assessment of effect modification by genes, lifestyle, and other occupational exposures, and evaluation of a wide range of health outcomes. Increased use of prospective cohorts would be beneficial in identifying hazardous exposures in the workplace. Occupational epidemiologists should seek opportunities to initiate prospective cohorts to investigate high priority, occupational exposures.

GSTM1 null and NAT2 slow acetylation genotypes, smoking intensity and bladder cancer risk: results from the New England bladder cancer study and NAT2 meta-analysis.

Associations between bladder cancer risk and NAT2 and GSTM1 polymorphisms have emerged as some of the most consistent findings in the genetic epidemiology of common metabolic polymorphisms and cancer, but their interaction with tobacco use, intensity and duration remain unclear. In a New England population-based case-control study of urothelial carcinoma, we collected mouthwash samples from 1088 of 1171 cases (92.9%) and 1282 of 1418 controls (91.2%) for genotype analysis of GSTM1, GSTT1 and NAT2 polymorphisms. Odds ratios and 95% confidence intervals of bladder cancer among New England Bladder Cancer Study subjects with one or two inactive GSTM1 alleles (i.e. the 'null' genotype) were 1.26 (0.85-1.88) and 1.54 (1.05-2.25), respectively (P-trend = 0.008), compared with those with two active copies. GSTT1 inactive alleles were not associated with risk. NAT2 slow acetylation status was not associated with risk among never (1.04; 0.71-1.51), former (0.95; 0.75-1.20) or current smokers (1.33; 0.91-1.95); however, a relationship emerged when smoking intensity was evaluated. Among slow acetylators who ever smoked at least 40 cigarettes/day, risk was elevated among ever (1.82; 1.14-2.91, P-interaction = 0.07) and current heavy smokers (3.16; 1.22-8.19, P-interaction = 0.03) compared with rapid acetylators in each category; but was not observed at lower intensities. In contrast, the effect of GSTM1-null genotype was not greater among smokers, regardless of intensity. Meta-analysis of the NAT2 associations with bladder cancer showed a highly significant relationship. Findings from this large USA population-based study provided evidence that the NAT2 slow acetylation genotype interacts with tobacco smoking as a function of exposure intensity.

The statistical power of epidemiological studies analyzing the relationship between exposure to ionizing radiation and cancer, with special reference to childhood leukemia and natural background radiation.

The etiology of childhood leukemia remains generally unknown, although risk models based on the Japanese A-bomb survivors imply that the dose accumulated from protracted exposure to low-level natural background ionizing radiation materially raises the risk of leukemia in children. In this paper a novel Monte Carlo score-test methodology is used to assess the statistical power of cohort, ecological and case-control study designs, using the linear low-dose part of the BEIR V model derived from the Japanese data. With 10 (or 20) years of follow-up of childhood leukemias in Great Britain, giving about 4600 (or 9200) cases, under an individual-based cohort design there is 67.9% (or 90.9%) chance of detecting an excess (at 5% significance level, one-sided test); little difference is made by extreme heterogeneity in risk. For an ecological design these figures reduce to 57.9% (or 83.2%). Case-control studies with five controls per case achieve much of the power of a cohort design, 61.1% (or 86.0%). However, participation bias may seriously affect studies that require individual consent, and area-based studies are subject to severe interpretational problems. For this reason register-based studies, in particular those that make use of predicted doses that avoid the need for interviews, have considerable advantages. We argue that previous studies have been underpowered (all have power <80%), and some are also subject to unquantifiable biases and confounding. Sufficiently large studies should be capable of detecting the predicted risk attributable to natural background radiation.

Differential effects of smoking on lung cancer mortality before and after household stove improvement in Xuanwei, China.

BACKGROUND: In Xuanwei County, Yunnan Province, China, lung cancer mortality rates in both males and females are among the highest in China. METHODS: We evaluated differential effects of smoking on lung cancer mortality before and after household stove improvement with chimney to reduce exposure to smoky coal emissions in the unique cohort in Xuanwei, China. Effects of independent variables on lung cancer mortality were measured as hazard ratios and 95% confidence intervals using a multivariable Cox regression model that included separate time-dependent variables for smoking duration (years) before and after stove improvement. RESULTS AND CONCLUSION: We found that the effect of smoking on lung cancer risk becomes considerably stronger after chimney installation and consequent reduction of indoor coal smoke exposure.

Autoimmune thyroiditis and exposure to iodine 131 in the Ukrainian cohort study of thyroid cancer and other thyroid diseases after the Chornobyl accident: results from the first screening cycle (1998-2000).

CONTEXT: Due to the Chornobyl accident, millions were exposed to radioactive isotopes of iodine and some received appreciable iodine 131 (131I) doses. A subsequent increase in thyroid cancer has been largely attributed to this exposure, but evidence concerning autoimmune thyroiditis (AIT) remains inconclusive. OBJECTIVE: The objective of the study was to quantify risk of AIT after 131I exposure. DESIGN/SETTING/PARTICIPANTS: Baseline data were collected from the first screening cycle (1998-2000) of a large cohort of radiation-exposed individuals (n = 12,240), residents of contaminated, iodine-deficient territories of Ukraine. Study individuals were under the age of 18 yr on April 26, 1986, and had thyroid radioactivity measurements made shortly after the accident. OUTCOMES: AIT was defined a priori based on various combinations of elevated antibodies to thyroid peroxidase (ATPO), TSH, and clinical findings; elevated ATPO were considered to be an indicator of thyroid autoimmunity. RESULTS: No significant association was found between 131I thyroid dose estimates and AIT, but prevalence of elevated ATPO demonstrated a modest, significant association with 131I that was well described by several concave models. This relationship was apparent in individuals with moderately elevated ATPO and euthyroid, thyroid disease-free individuals. CONCLUSIONS: Twelve to 14 yr after the Chornobyl accident, no radiation-related increase in prevalence of AIT was found in a large cohort study, the first in which 131I thyroid doses were estimated using individual radioactivity measurements. However, a dose-response relationship with ATPO prevalence raises the possibility that clinically important changes may occur over time. Thus, further follow-up and analysis of prospective data in this cohort are necessary.

Adjusting lung cancer risks for temporal and spatial variations in radon concentration in dwellings in Gansu Province, China.

Our recent study in Gansu Province, China reported an increasing risk of lung cancer with increasing residential radon concentration that was consistent with previous pooled analyses and with meta-analyses of other residential studies (Wang et al., Am. J. Epidemiol. 155, 554-564, 2002). Dosimetry used current radon measurements (1-year track-etch detectors) in homes to characterize concentrations for the previous 30 years, resulting in uncertainties in exposure and possibly reduced estimates of disease risk. We conducted a 3-year substudy in 55 houses to model the temporal and spatial variability in radon levels and to adjust estimates of radon risk. Temporal variation represented the single largest source of uncertainty, suggesting the usefulness of multi-year measurements to assess this variation; however, substantial residual variation remained unexplained. The uncertainty adjustment increased estimates of the excess odds ratio by 50-100%, suggesting that residential radon studies using similar dosimetry may also underestimate radon effects. These results have important implications for risk assessment.

Studies of radon and lung cancer in North America and China.

Studies of radon-exposed underground miners indicate that residential radon is the second leading cause of lung cancer. Seven case-control studies of residential radon have been conducted in North America and two in China, and represent all studies in these areas which included 200 or more lung cancer cases and used long-term radon detectors. North American studies enrolled 4081 cases and 5281 controls, and Chinese studies enrolled 1076 cases and 2015 controls. Based on analyses of pooled data, odds ratios (ORs) and 95% confidence limits at 100 Bq m(-3) were 1.106 (1.00,1.28) for the North American studies and 1.139 (1.01,1.37) for the Chinese studies. Tests of homogeneity of ORs within populations were not significant. Among subjects with complete dosimetry for the 5-30 y exposure period prior to interview, ORs at 100 Bq m(-3) were 1.205 (1.03,1.50) for the North American studies and 1.279 (1.07,0.75) for the Chinese studies. Results are consistent with extrapolations from miners and indicate an excess lung cancer risk from residential radon.

Thyroid cancer following scalp irradiation: a reanalysis accounting for uncertainty in dosimetry.

In the 1940s and 1950s, over 20,000 children in Israel were treated for tinea capitis (scalp ringworm) by irradiation to induce epilation. Follow-up studies showed that the radiation exposure was associated with the development of malignant thyroid neoplasms. Despite this clear evidence of an effect, the magnitude of the dose-response relationship is much less clear because of probable errors in individual estimates of dose to the thyroid gland. Such errors have the potential to bias dose-response estimation, a potential that was not widely appreciated at the time of the original analyses. We revisit this issue, describing in detail how errors in dosimetry might occur, and we develop a new dose-response model that takes the uncertainties of the dosimetry into account. Our model for the uncertainty in dosimetry is a complex and new variant of the classical multiplicative Berkson error model, having components of classical multiplicative measurement error as well as missing data. Analysis of the tinea capitis data suggests that measurement error in the dosimetry has only a negligible effect on dose-response estimation and inference as well as on the modifying effect of age at exposure.

Mortality among augmentation mammoplasty patients.

Much attention has focused on disease risks among women receiving silicone breast implants, but there has been little evaluation of their mortality experience. We undertook a retrospective cohort study of 13,488 women receiving cosmetic implants and 3,936 patients with other types of plastic surgery at 18 plastic surgery practices. After an average of 13 years of follow-up, deficits in overall mortality were found as compared with the general population (U.S. rates) for both implant [255 deaths; standardized mortality ratio (SMR) = 0.69, 95% confidence interval (CI) = 0.6-0.8] and comparison subjects (125 deaths; SMR = 0.58, 95% CI = 0.5-0.7). These findings indicate that patients seeking plastic surgery are in general healthier than their peers. Implant patients, however, experienced excess risks of death compared with the general population for brain cancer (SMR = 2.45) and suicide (SMR = 1.54). Internal analyses showed a higher overall mortality among the implant than among the comparison patients (relative risk = 1.27, 95% CI = 1.0-1.6). This overall excess reflected increases for respiratory tract (SMR = 3.03) and brain (SMR = 2.25) cancers and for suicide (SMR = 4.24).

Cancer risk at sites other than the breast following augmentation mammoplasty.

PURPOSE: There has been limited investigation of cancer risk other than breast cancer among patients with breast implants, despite some clinical and laboratory evidence suggesting links with certain cancer sites, including hematopoietic and connective tissue malignancies. METHODS: A retrospective cohort study of 13,488 patients who received cosmetic breast implants at 18 plastic surgery practices in six geographic areas was conducted to assess long-term health effects. After an average of 12 years of follow-up, questionnaires were administered to subjects located and alive (78% of eligible population). Attempts were made to obtain death certificates for deceased subjects and medical verification for all reported cancers. Expected numbers of cancers were derived using general population cancer incidence rates and an internal comparison series of 3936 patients who received other types of plastic surgery at the same practices as the implant patients. RESULTS: A total of 359 malignancies was observed versus 295.95 expected based on general population rates, resulting in a standardized incidence ratio (SIR) of 1.21 [95% confidence interval (CI) 1.1-1.4]. Individual malignancies for which incidence was significantly elevated included cancers of the stomach (SIR = 2.65), cervix (SIR = 3.18), vulva (SIR = 2.51), brain (SIR = 2.16), and leukemia (SIR = 2.19). No excess risks were observed for other hematopoietic malignancies, including multiple myeloma. The internal analyses, however, based on cancer rates derived among the comparison patients, showed no increased cancer risk among the implant patients [relative risk (RR) = 1.00, 95% CI 0.8-1.2], as well as no statistically significant elevations for most individual sites. Cervical cancer continued to be elevated (RR = 1.78), although to a lesser extent than in the external analyses, while the risk for respiratory cancers was higher (RR = 2.40). Non-significant elevations in risk persisted in this analysis for liver cancer (RR = 2.65), brain cancer (RR = 2.83), and leukemia (RR = 1.83). Many of the cancers showing excesses were defined on the basis of death certificates, requiring caution in interpretation. The histologies of the leukemias were quite varied, which makes a biologic relationship appear unlikely. However, respiratory cancers showed some evidence of increasing risk with follow-up time and both respiratory and brain cancers were elevated in the mortality analyses. CONCLUSIONS: Although excesses of cervical and vulvar cancer among implant patients might be attributable to lifestyle factors, reasons for excesses of respiratory and brain cancers were less apparent.

Previous pulmonary diseases and risk of lung cancer in Gansu Province, China.

BACKGROUND: Although active smoking is well established as the main cause of lung cancer, there is accumulating evidence that history of prior lung diseases may be an independent risk factor for lung cancer. METHODS: A population-based case-control study in Gansu Province, China identified 886 lung cancer cases (656 male, 230 female) diagnosed between January 1994 and April 1998. A standardized interview collected information on a variety of potential risk factors including a history of physician-diagnosed non-malignant lung diseases (pulmonary tuberculosis, chronic bronchitis/emphysema, asthma, pneumonia), age and year in which each condition was first diagnosed, and any therapy or hospitalization received. RESULTS: Pulmonary tuberculosis (odds ratio [OR] = 2.1, 95% CI : 1.4-3.1) and chronic bronchitis/emphysema (OR = 1.4, 95% CI : 1.1-1.8) were associated with increased risk of lung cancer, after adjustment for active smoking and socioeconomic status. The OR for asthma (OR = 1.4, 95% CI : 0.9-2.1) and pneumonia (OR = 1.5, 95% CI : 1.0-2.3) were also elevated. The risk of lung cancer remained significant for pulmonary tuberculosis and chronic bronchitis/emphysema when analysis was limited to the pathologically confirmed cases and self-responders. CONCLUSIONS: This study provides additional evidence that previous pulmonary tuberculosis and chronic bronchitis/emphysema are causally related to lung cancer, although the precise mechanism is still unclear. The results for asthma and pneumonia, while suggestive of a positive association, did not reach the traditional level of statistical significance and should be interpreted with caution.

Analysis of exposure-time-response relationships using a spline weight function.

To examine the time-dependent effects of exposure histories on disease, we estimate a weight function within a generalized linear model. The shape of the weight function, which is modeled as a cubic B-spline, gives information about the impact of exposure increments at different times on disease risk. The method is evaluated in a simulation study and is applied to data on smoking histories and lung cancer from a recent case-control study in Germany.

Breast cancer following augmentation mammoplasty (United States).

OBJECTIVE: Although clinical reports have raised concern that breast implants may either increase the risk of breast cancer or delay its diagnosis, epidemiologic studies have generally shown implant recipients to be at a reduced risk of subsequent breast cancer. A large retrospective cohort study was undertaken to clarify effects of cosmetic breast implantation. METHODS: Medical records of 13,488 women receiving cosmetic implants at 18 plastic surgery practices and a group of 3936 patients who received other types of plastic surgery at the same practices were reviewed and information abstracted. Questionnaires were sent to all subjects located as alive, with 71% being completed. Attempts were made to obtain medical verification for all reported cancers and to obtain death certificates for deceased subjects. RESULTS: A total of 136 breast cancers were observed among the breast implant patients. External analyses, using general population rates from the Surveillance, Epidemiology and End Results (SEER) program, resulted in 152.2 cases expected and a standardized incidence ratio (SIR) of 0.9 (95% CI 0.8-1.1). A comparable SIR was found for the other plastic surgery patients (SIR =1.0, 95% CI 0.7-1.2). Internal analyses, directly comparing the implant patients with the other plastic surgery patients, showed a RR of 0.8 (95% CI 0.6-1.1). In neither the external nor internal analyses was there any systematic variation in risk by age or calendar year of initial implant. Risk also did not vary by years of follow-up or by type of implant. Risk was not affected by exclusion of patients who received their implants following surgery for benign breast disease. Although breast tumors tended to be detected at a somewhat later stage among the breast implant than the comparison patients, the difference was not statistically significant, nor was there any significant difference in breast cancer mortality between the two groups. CONCLUSIONS: Breast implants do not appear to alter the risk of subsequent breast cancer.

The use of sliding time windows for the exploratory analysis of temporal effects of smoking histories on lung cancer risk.

To examine the time-dependent effects of exposure histories on disease we use sliding time windows as an exploratory alternative to the analysis of variables like time since last exposure and duration of exposure. The method fits a series of risk models which contain total cumulative exposure and an additional covariate for exposures received during fixed time intervals. Characteristics of the fitted models provide insight into the influence of exposure increments at different times on disease risk. A simulation study is performed to check the validity of the approach. We apply the method to data from a recent German case-control study on smoking and lung cancer risk with about 4300 lung cancer cases and a similiar number of controls. The sliding time window approach indicates that the amount of cigarettes smoked from two to 11 years before disease incidence is most predicitive of lung cancer incidence. Among different smoking profiles that result in the same lifelong cumulative number of cigarettes smoked, those with a concentration of smoked cigarettes within 20 years before interview bear substantially larger risk than others.

Lung cancer and environmental tobacco smoke in a non-industrial area of China.

We report results from a population-based case-control study of lung cancer and environmental tobacco smoke (ETS) among never-smokers conducted in 2 rural prefectures of China, including 200 female and 33 male lung cancer cases, and 407 female and 114 male controls, matched on age, sex and prefecture of current residence. The odds ratio (OR) for ever-exposed to ETS was 1.19 (95% CI 0.7-2.0), with a significant trend (p<0.05) with increasing exposure. ORs were 1.00, 1.04, 1.13 and 1.51 for non-exposed, <10, 10-19 and >/=20 pack-years of ETS exposure, respectively. Excess risks were limited to ETS exposures in childhood (

Respiratory cancer in a cohort of copper smelter workers: results from more than 50 years of follow-up.

Several studies have linked inhalation of airborne arsenic with increased risk of respiratory cancer, but few have analyzed the shape of the exposure-response curve. In addition, since inhaled airborne arsenic affects systemic levels of inhaled arsenic, there is concern that inhaled arsenic may be associated with cancers of the skin, bladder, kidney, and liver, which have been linked to ingested arsenic. The authors followed 8,014 white male workers who were employed for 12 months or more prior to 1957 at a Montana copper smelter from January 1, 1938 through December 31, 1989. A total of 4,930 (62%) were deceased, including 446 from respiratory cancer. Significantly increased standardized mortality ratios (SMRs) were found for all causes (SMR = 1.14), all cancers (SMR = 1.13), respiratory cancer (SMR = 1.55), diseases of the nervous system and sense organs (SMR = 1.31), nonmalignant respiratory diseases (SMR = 1.56), emphysema (SMR = 1.73), ill-defined conditions (SMR = 2.26), and external causes (SMR = 1.35). Internal analyses revealed a significant, linear increase in the excess relative risk of respiratory cancer with increasing exposure to inhaled airborne arsenic. The estimate of the excess relative risk per mg/m3-year was 0.21/(mg/m3-year) (95% confidence interval: 0.10, 0.46). No other cause of death was related to inhaled arsenic exposure.