Sensitivity of the relation between cumulative magnetic field exposure and brain cancer mortality to choice of monitoring data grouping scheme

Self-reported and urinary biomarker-based measures of exposure to glyphosate and mancozeb and sleep problems among smallholder farmers in Uganda

OBJECTIVE

We aim to showcase the impact of applying eight different self-reported and urinary biomarker-based exposure measures for glyphosate and mancozeb on the association with sleep problems in a study among 253 smallholder farmers in Uganda.

METHODS

The questionnaire-based exposure measures included: (1) the number of application days of any pesticide in the last 7 days (never, 1-2; >2 days) and six glyphosate and mancozeb-specific measures: (2) application status over the last 12 months (yes/no), (3) recent application status (never, last 7 days and last 12 months), (4) the number of application days last 12 months, (5) average exposure-intensity scores (EIS) and (6) EIS-weighted number of application days in last 12 months. Based on 384 repeated urinary biomarker concentrations of ethylene thiourea (ETU) and glyphosate from 84 farmers, we also estimated (7) average biomarker concentrations for all 253 farmers. Also in the 84 farmers the measured pre-work and post-work biomarker concentrations were used (8). Multivariable logistic regression models were used to assess the association between the exposure measures and selected Medical Outcomes Study Sleep Scale (MOS-SS) indices (6-item, sleep inadequacy and snoring).

RESULTS

We observed positive associations between (1) any pesticide application in the last 7 days with all three MOS-SS indices. Glyphosate application in the last 7 days (3) and mancozeb application in the last 12 months (3) were associated with the 6-item sleep problem index. The estimated average urinary glyphosate concentrations showed an exposure-response association with the 6-item sleep problem index and sleep inadequacy in the same direction as based on self-reported glyphosate application in the last 7 days. In the analysis with the subset of 84 farmers, both measured and modelled post-work urinary glyphosate concentration showed an association with snoring.

CONCLUSIONS

Self-reported, estimated average biomarker concentrations and measured urinary biomarker exposure measures of glyphosate and mancozeb showed similar exposure-response associations with sleep outcomes.

Developing a company-specific job exposure matrix for the Asbest Chrysotile Cohort Study

OBJECTIVES

Exposure assessment for retrospective industrial cohorts are often hampered by limited availability of historical measurements. This study describes the development of company-specific job-exposure matrices (JEMs) based on measurements collected over five decades for a cohort study of 35 837 workers (Asbest Chrysotile Cohort Study) in the Russian Federation to estimate their cumulative exposure to chrysotile containing dust and fibres.

METHODS

Almost 100 000 recorded stationary dust measurements were available from 1951-2001 (factories) and 1964-2001 (mine). Linear mixed models were used to extrapolate for years where measurements were not available or missing. Fibre concentrations were estimated using conversion factors based on side-by-side comparisons. Dust and fibre JEMs were developed and exposures were allocated by linking them to individual workers' detailed occupational histories.

RESULTS

The cohort covered a total of 515 355 employment-years from 1930 to 2010. Of these individuals, 15% worked in jobs not considered professionally exposed to chrysotile. The median cumulative dust exposure was 26 mg/m3 years for the entire cohort and 37.2 mg/m3 years for those professionally exposed. Median cumulative fibre exposure was 16.4 fibre/cm3 years for the entire cohort and 23.4 fibre/cm3 years for those professionally exposed. Cumulative exposure was highly dependent on birth cohort and gender. Of those professionally exposed, women had higher cumulative exposures than men as they were more often employed in factories with higher exposure concentrations rather than in the mine.

CONCLUSIONS

Unique company-specific JEMs were derived using a rich measurement database that overlapped with most employment-years of cohort members and will enable estimation of quantitative exposure-response.

Grouping schemes of welding fume exposure in shipyard welders

Welding fume exposure can increase the risk of chronic obstructive pulmonary disease (COPD). The aim of this study was to evaluate the optimal grouping schemes of welding fume exposure in shipyard welders for future accurate examination of the association between welding fume exposure and COPD. Industrial hygiene records, including welding fume measurements between 2002 and 2009 were collected from a shipyard. A total of 2,360 personal welding fume measurements was compiled with a geometric mean of 1.66 mg/m³ and a geometric standard deviation of 4.02. Welding jobs were categorized into 8 groups. There were 9 working areas. To obtain the optimal grouping scheme, various grouping schemes were created using job, area, and job^*area combination. To compare various grouping schemes, contrast and precision were calculated for each grouping scheme. For all measurement data, group mean ranking method created by ranking geometric means of the job^*area combination into 3 groups (group mean ranking method) showed the best contrast and precision values among various grouping schemes, followed by grouping based on the job. For a subset of the data excluding job^*area combinations with less than 10 measurements, grouping based on the job showed greater contrast than group mean ranking method, while for other subsets, including only repeated measurement data or further excluding job^*area combinations with less than 10 measurements from the repeated measurement subset, group mean ranking method showed greater contrast than grouping based on the job. Our results showed that group mean ranking or grouping based on the job could be a candidate for the optimal grouping schemes in this shipyard. Our efforts for optimal grouping scheme may aid future cohort study to elucidate the association between welding fume exposure and COPD.

Sensitivity analyses of exposure estimates from a quantitative job-exposure matrix (SYN-JEM) for use in community-based studies

OBJECTIVES

We describe the elaboration and sensitivity analyses of a quantitative job-exposure matrix (SYN-JEM) for respirable crystalline silica (RCS). The aim was to gain insight into the robustness of the SYN-JEM RCS estimates based on critical decisions taken in the elaboration process.

METHODS

SYN-JEM for RCS exposure consists of three axes (job, region, and year) based on estimates derived from a previously developed statistical model. To elaborate SYN-JEM, several decisions were taken: i.e. the application of (i) a single time trend; (ii) region-specific adjustments in RCS exposure; and (iii) a prior job-specific exposure level (by the semi-quantitative DOM-JEM), with an override of 0 mg/m(3) for jobs a priori defined as non-exposed. Furthermore, we assumed that exposure levels reached a ceiling in 1960 and remained constant prior to this date. We applied SYN-JEM to the occupational histories of subjects from a large international pooled community-based case-control study. Cumulative exposure levels derived with SYN-JEM were compared with those from alternative models, described by Pearson correlation ((Rp)) and differences in unit of exposure (mg/m(3)-year). Alternative models concerned changes in application of job- and region-specific estimates and exposure ceiling, and omitting the a priori exposure ranking.

RESULTS

Cumulative exposure levels for the study subjects ranged from 0.01 to 60 mg/m(3)-years, with a median of 1.76 mg/m(3)-years. Exposure levels derived from SYN-JEM and alternative models were overall highly correlated (R(p) > 0.90), although somewhat lower when omitting the region estimate ((Rp) = 0.80) or not taking into account the assigned semi-quantitative exposure level (R(p) = 0.65). Modification of the time trend (i.e. exposure ceiling at 1950 or 1970, or assuming a decline before 1960) caused the largest changes in absolute exposure levels (26-33% difference), but without changing the relative ranking ((Rp) = 0.99).

CONCLUSIONS

Exposure estimates derived from SYN-JEM appeared to be plausible compared with (historical) levels described in the literature. Decisions taken in the development of SYN-JEM did not critically change the cumulative exposure levels. The influence of region-specific estimates needs to be explored in future risk analyses.

Distinguishing the common components of oil- and water-based metalworking fluids for assessment of cancer incidence risk in autoworkers

BACKGROUND

Metalworking fluids (MWF)--straight, soluble, and synthetic--have overlapping components. We derived constituent-based metrics of polycyclic aromatic hydrocarbons (PAHs), water-based MWF, biocides, and nitrosamines to account for this overlap and examined their relations with cancer incidence.

METHODS

An autoworkers cohort of 30,000 was followed for cancer incidence. Hazard ratios were estimated for each cancer and cumulative exposure (lagged) to each new metric; soluble MWF contributed variably to several metrics with weight k = 0-1.

RESULTS

For most cancer sites, the constituent-based metrics resulted in stronger exposure-disease associations than the MWF classes alone. Laryngeal and bladder cancer were most strongly associated with PAH (k = 0). Protective effects for stomach and lung cancer were observed with biocide, a component that may be a surrogate for endotoxin.

CONCLUSIONS

Our findings provide support and clarification of possible etiologies for previous positive associations and provide support for distinguishing exposure from oil- and water-based MWF in epidemiologic studies.

Sensitivity of the association between increased lung cancer risk and bitumen fume exposure to the assumptions in the assessment of exposure

PURPOSE

A multi-centre IARC-coordinated European cohort study provided evidence of an association between lung cancer risk and bitumen fume exposure. Sensitivity analyses were conducted to assess whether these associations were robust to assumptions in the exposure assessment for which support could not be obtained due to lack of either measurements or direct observations.

METHODS

New exposure estimates were generated by changing assumptions on exposure levels, specific tasks, lags, and coal tar use. Subsequently, Poisson regression models estimated relative risks (RRs); change in fit of models was evaluated.

RESULTS

The influence of these assumptions was minimal, with log-likelihood deviations between -5.0 and 4.7% and similar patterns in dose-dependent increases of lung cancer risk. In the highest exposure categories, point estimates of RRs ranged 2.07-2.83 for average, and 1.22-2.23 for cumulative exposure.

CONCLUSIONS

The small increase in lung cancer risk associated with bitumen fume exposure depends only to a limited extent on the subjective judgments made in the exposure assessment for this cohort.

Impact of the Specificity of the Exposure Metric on Exposure???Response Relationships

BACKGROUND

Exposure misclassification may occur when nonspecific exposure indicators are used. Developing estimates of more specific measures may be difficult due to sampling limitations or a paucity of historical measurements and, thus, often requires substantial effort. We examine the impact on exposure-response relationships of moving from 2 measures of exposure mixtures (dust, chlorophenols) to more specific exposure indicators (wood dust, pentachlorophenol, tetrachlorophenol) in a retrospective cohort.

METHODS

The study population consisted of 26,847 male sawmill workers (> or =1 year employment between 1950 and 1995) with linkage to national cancer registries. A subcohort (n = 11,273 employed more than 1 day between 1985 and 1995) was linked to hospital discharge records. We evaluated the shape (log-linear vs log-log models), goodness of fit, precision, and expected versus observed attenuation of the exposure-response relationships.

RESULTS

The correlation between the cumulative exposure indices was moderately high (dust/wood dust, r = 0.68; total chlorophenol/pentachlorophenol, r = 0.88; total chlorophenol/tetrachlorophenol, r = 0.78). An increase in chronic obstructive pulmonary disease hospitalizations was found with wood dust but not with total dust. Stronger associations for non-Hodgkin lymphoma and kidney cancer incidence were observed with pentachlorophenol than with total chlorophenol; no association was observed with tetrachlorophenol. We observed greater attenuation than expected using total dust, but less than expected using total chlorophenol.

CONCLUSIONS

The relationships between health outcomes were substantially attenuated when nonspecific exposure indicators were used. This study demonstrates the importance of developing exposure metrics as specific to the disease-causing agent as possible, particularly when the composition of mixed exposures varies by work areas.

Comparison of two indices of exposure to polycyclic aromatic hydrocarbons in a retrospective aluminium smelter cohort

BACKGROUND

The association between coal tar-derived substances, a complex mixture of polycyclic aromatic hydrocarbons, and cancer is well established. However, the specific aetiological agents are unknown.

OBJECTIVE

To compare the dose-response relationships for two common measures of coal tar-derived substances, benzene-soluble material (BSM) and benzo(a)pyrene (BaP), and to evaluate which among these is more strongly related to the health outcomes.

METHODS

The study population consisted of 6423 men with > or =3 years of work experience at an aluminium smelter (1954-97). Three health outcomes identified from national mortality and cancer databases were evaluated: incidence of bladder cancer (n = 90), incidence of lung cancer (n = 147) and mortality due to acute myocardial infarction (AMI, n = 184). The shape, magnitude and precision of the dose-response relationships and cumulative exposure levels for BSM and BaP were evaluated. Two model structures were assessed, where 1n(relative risk) increased with cumulative exposure (log-linear model) or with log-transformed cumulative exposure (log-log model).

RESULTS

The BaP and BSM cumulative exposure metrics were highly correlated (r = 0.94). The increase in model precision using BaP over BSM was 14% for bladder cancer and 5% for lung cancer; no difference was observed for AMI. The log-linear BaP model provided the best fit for bladder cancer. The log-log dose-response models, where risk of disease plateaus at high exposure levels, were the best-fitting models for lung cancer and AMI.

CONCLUSION

BaP and BSM were both strongly associated with bladder and lung cancer and modestly associated with AMI. Similar conclusions regarding the associations could be made regardless of the exposure metric.

Reducing attenuation in exposure-response relationships by exposure modeling and grouping: the relationship between wood dust exposure and lung function

BACKGROUND

In a study of wood dust exposure and lung function, we tested the effect on the exposure-response relationship of six different exposure metrics using the mean measured exposure of each subject versus the mean exposure based on various methods of grouping subjects, including job-based groups and groups based on an empirical model of the determinants of exposure.

METHODS

Multiple linear regression was used to examine the association between wood dust concentration and forced expiratory volume in 1s (FEV(1)), adjusting for age, sex, height, race, pediatric asthma, and smoking.

RESULTS

Stronger point estimates of the exposure-response relationships were observed when exposures were based on increasing levels of aggregation, allowing the relationships to be found statistically significant in four of the six metrics. The strongest point estimates were found when exposures were based on the determinants of exposure model.

CONCLUSIONS

Determinants of exposure modeling offers the potential for improvement in risk estimation equivalent to or beyond that from job-based exposure grouping.

Exposure variability: concepts and applications in occupational epidemiology

BACKGROUND

Standard approaches to assessing exposures for epidemiologic studies tend to concentrate resources on obtaining detailed data for each study participant at the expense of characterizing within-person variability.

METHODS

This paper presents some basic, generalizeable concepts concerning exposure and its variability, describes methods that can be used to analyze, describe, and understand that variability, and reviews related implications for the design and interpretation of epidemiologic studies.

RESULTS AND CONCLUSIONS

Insufficient attention to the balance of within- and between-person variation in exposure can reduce the efficiency of measurement efforts and attenuate estimates of exposure-disease association. Exposure variability should consequently be considered carefully in the planning, analysis, and interpretation of epidemiologic studies. Greater attention to these matters can lead to more meaningful characterization of exposure itself, and, most importantly, improve the chances that epidemiologic studies can identify and accurately characterize health hazards.

Validity of empirical models of exposure in asphalt paving

AIMS

To investigate the validity of empirical models of exposure to bitumen fume and benzo(a)pyrene, developed for a historical cohort study of asphalt paving in Western Europe.

METHODS

Validity was evaluated using data from the USA, Italy, and Germany not used to develop the original models. Correlation between observed and predicted exposures was examined. Bias and precision were estimated.

RESULTS

Models were imprecise. Furthermore, predicted bitumen fume exposures tended to be lower (-70%) than concentrations found during paving in the USA. This apparent bias might be attributed to differences between Western European and USA paving practices. Evaluation of the validity of the benzo(a)pyrene exposure model revealed a similar to expected effect of re-paving and a larger than expected effect of tar use. Overall, benzo(a)pyrene models underestimated exposures by 51%.

CONCLUSIONS

Possible bias as a result of underestimation of the impact of coal tar on benzo(a)pyrene exposure levels must be explored in sensitivity analysis of the exposure-response relation. Validation of the models, albeit limited, increased our confidence in their applicability to exposure assessment in the historical cohort study of cancer risk among asphalt workers.

Validation of a job-exposure matrix for assessment of utility worker exposure to magnetic fields

The aim of this study was to evaluate a 50-Hz electromagnetic field job-exposure matrix used in epidemiological studies of a nationwide cohort of utility workers in Denmark. We compared a job-exposure matrix that distinguished four categories of exposure to 50-Hz time-weighted average (TWA) magnetic fields: low (< 0.1 microT), medium (0.1-0.29 microT), high (0.3-0.99 microT) and very high (> 1.0 microT) of utility company employees with 196 measurements of 8-h exposure for 129 workers in this industry. The 129 workers were selected from the following five main work environments: generation facilities, transmission lines, distribution lines, substations, and other electrically and non-electrically relates jobs. This study shows that the job-exposure matrix can be expected to introduce misclassification mainly between adjacent categories of exposure. Thus, the distribution of measurements of exposure to 50-Hz magnetic fields was similar for workers in the medium and the high exposure matrix categories. But the two extreme categories satisfactorily separate low and very highly exposed workers. The study shows that epidemiological use of this job-exposure matrix might combine the two intermediate categories of exposure. If the sample size in extreme categories provides enough power, a study in which this job-exposure matrix is used should allow detection of a true association between exposure to 50-Hz magnetic field and disease.

Case-cohort analysis of brain cancer and leukemia in electric utility workers using a refined magnetic field job-exposure matrix

BACKGROUND

The potential association between occupational electric and magnetic field exposure and cancer is well documented in the literature, but there is uncertainty regarding a causal relation.

METHODS

Using data from a completed cohort study, we sought to refine the job-exposure matrix in a case-cohort analysis by regrouping jobs into more homogeneous groups, but without making additional measurements. From the original cohort, we selected the 164 men who died of leukemia, 145 men who died of brain cancer, and a random subcohort of 800 men (0.6% of the cohort). Erroneous job assignments were corrected and job groups were subdivided based on differences in work environments or tasks performed.

RESULTS

Magnetic field exposure remained unrelated to leukemia mortality and positively associated with brain cancer mortality based on both cumulative and average magnetic field indices. Although not monotonic across the middle intervals, increased risk of brain cancer was found in relation to career exposure, with risk ratios of 1.8 (95% CI = 0.7-4.7) and 2.5 (95% CI = 1.0-6.3) in the uppermost categories for cumulative and average exposure, stronger for exposure 2-10 years past.

CONCLUSIONS

Improvements in exposure assignment based only on reassignment of job titles to occupational categories had little impact on the measured associations of magnetic fields with leukemia or brain cancer.

Uncertainty in the relation between exposure to magnetic fields and brain cancer due to assessment and assignment of exposure and analytical methods in dose-response modeling

Incomplete scientific knowledge ensures that, in every study, uncertainty will enter the processes of exposure estimation and exposure-response modeling. In the light of the heated debate about the health effects of magnetic fields resulting from power production and usage, we undertook a sensitivity analysis to evaluate uncertainty related to key decisions in a previous study of brain cancer and occupational exposure to magnetic fields. The findings appeared to be relatively insensitive to most variations in the methods of exposure assessment, exposure assignment, and data analysis. The results can be visualized by defining bands of uncertainty about a best-bet estimate of the association based on our original study. These bands of methodological uncertainties were similar in magnitude to the conventional 95% confidence interval, but they provide a measure of the potential range of systematic bias in the results, rather than reflecting statistical variability alone. The methodology employed here can be applied to other studies, and other researchers are encouraged to conduct sensitivity analysis in order to estimate methodological uncertainty as an alternative to statistical confidence intervals.

Assessment of the sensitivity of the relation between current exposure to carbon black and lung function parameters when using different grouping schemes

BACKGROUND

Equations using variance components in exposure data to predict attenuation and standard error of exposure--response slopes have been published recently. However, to date, no comparisons have been made between results of applying these equations to a real data set with the exposure-response relations estimated directly.

METHODS

Data on lung function parameters and personal inhalable and respirable dust exposure levels from the European carbon black respiratory health study were used. The predicted attenuation and standard error of the relation between current inhalable and respirable dust levels and lung function parameters (FVC and FEV(1)) were calculated for various exposure grouping schemes. These results were compared with the observed exposure-response relations. Workers were grouped by Job Category, the combination of factory and Job Category and five a posteriori created Exposure Groups. In addition, the individual approach was also used, as exposure data were available for each worker.

RESULTS

The rank orders of the coefficients from the regression analyses using the different grouping schemes were similar to those predicted by the equations, although the differences were larger than predicted. When using inhalable dust exposure, the predicted standard errors of the exposure-response slopes were slightly lower than those estimated directly; for respirable dust the predicted standard errors were about a factor two to three smaller than those from the regression analyses. When considering FVC, the predicted exposure-response relations were all statistically significant, whilst the observed relation was only significant when using the five a posteriori Exposure Groups. When reviewing the relations between dust exposure and level of FEV(1), all relations were statistically significant, with the exception of the (observed) relation between respirable dust and FEV(1), when the individual approach was used.

CONCLUSIONS

Using different grouping schemes for estimating exposure can have large effects on the slope and standard error of the exposure-response relation. It is, therefore, important that the effect of the different grouping schemes on the level and precision of the exposure-response slope be estimated. Despite violation of most of the assumptions when applying the equations to predict attenuation and the standard error of the exposure-response slope, the similarities in predicted and observed exposure-response relations and standard errors are indicative of the robustness of these equations. Therefore, the equations appear to be a useful tool in establishing the most efficient way of utilizing exposure measurements.