James A. Deddens

Approaches for estimating prevalence ratios.

Recently there has been much interest in estimating the prevalence (risk, proportion or probability) ratio instead of the odds ratio, especially in occupational health studies involving common outcomes (for example, with prevalence rates above 10%). For example, if 80 out of 100 exposed subjects have a particular disease and 50 out of 100 non-exposed subjects have the disease, then the odds ratio (OR) is (80/20)/(50/50) = 4. However, the prevalence ratio (PR) is (80/100)/(50/100) = 1.6. The latter indicates that the exposed subjects are only 1.6 times as likely to have the disease as the non-exposed subjects, and this is the number in which most people would be interested. There is considerable literature on the advantages and disadvantages of OR versus PR (see Greenland,1 Stromberg,2 Axelson et al 3 and others). In this article we will review the existing methods and give examples and recommendations on how to estimate the PR. The most common method of modelling binomial (no/yes or 0/1) health outcomes today is logistic regression. In logistic regression one models the probability of the binomial outcome (Y = 1) of interest as: P(Y = 1| X1, X2, …, Xk) = eXβ/(1+eXβ) where Xβ = β+β1X1+β2X2+…+βkXk. Then exp(β1) = OR for a 1 unit increase in X1 adjusted for all other variables in the model. Logistic regression yields maximum likelihood estimates (MLEs) of the OR (adjusted for other covariates). If the adjusted OR is the parameter of interest, then these MLEs are generally considered the best estimators available. The adjusted OR can also be used to estimate the adjusted PR, but this should only be done for a rare disease (eg, one with a prevalence of 10% or less). This, together with the fact that …

Maternal and Paternal Risk Factors for Cryptorchidism and Hypospadias: A Case–Control Study in Newborn Boys

Little is known on environmental risk factors for cryptorchidism and hypospadias, which are among the most frequent congenital abnormalities. The aim of our study was to identify risk factors for cryptorchidism and hypospadias, with a focus on potential endocrine disruptors in parental diet and occupation. In a case–control study nested within a cohort of 8,698 male births, we compared 78 cryptorchidism cases and 56 hypospadias cases with 313 controls. The participation rate was 85% for cases and 68% for controls. Through interviews, information was collected on pregnancy aspects and personal characteristics, lifestyle, occupation, and dietary phytoestrogen intake of both parents. Occupational exposure to potential endocrine disruptors was classified based on self-reported exposure and ratings of occupational hygienists based on job descriptions. Our findings indicate that paternal pesticide exposure was associated with cryptorchidism [odds ratio (OR) = 3.8; 95% confidence interval (95% CI), 1.1–13.4]. Smoking of the father was associated with hypospadias (OR = 3.8; 95% CI, 1.8–8.2). Maternal occupational, dietary, and lifestyle exposures were not associated with either abnormality. Both abnormalities were associated with suboptimal maternal health, a lower maternal education, and a Turkish origin of the parents. Being small for gestational age was a risk factor for hypospadias, and preterm birth was a risk factor for cryptorchidism. Because paternal pesticide exposure was significantly associated with cryptorchidism and paternal smoking was associated with hypospadias in male offspring, paternal exposure should be included in further studies on cryptorchidism and hypospadias risk factors.

Cancer Risk Among Firefighters: A Review and Meta-analysis of 32 Studies

Objective: The objective of this study was to review 32 studies on firefighters and to quantitatively and qualitatively determine the cancer risk using a meta-analysis. Methods: A comprehensive search of computerized databases and bibliographies from identified articles was performed. Three criteria used to assess the probable, possible, or unlikely risk for 21 cancers included pattern of meta-relative risks, study type, and heterogeneity testing. Results: The findings indicated that firefighters had a probable cancer risk for multiple myeloma with a summary risk estimate (SRE) of 1.53 and 95% confidence interval (CI) of 1.21–1.94, non-Hodgkin lymphoma (SRE = 1.51, 95% CI = 1.31–1.73), and prostate (SRE = 1.28; 95% CI = 1.15–1.43). Testicular cancer was upgraded to probable because it had the highest summary risk estimate (SRE = 2.02; 95% CI = 1.30–3.13). Eight additional cancers were listed as having a “possible” association with firefighting. Conclusions: Our results confirm previous findings of an elevated metarelative risk for multiple myeloma among firefighters. In addition, a probable association with non-Hodgkin lymphoma, prostate, and testicular cancer was demonstrated.

A comparison of two methods for estimating prevalence ratios

BackgroundIt is usually preferable to model and estimate prevalence ratios instead of odds ratios in cross-sectional studies when diseases or injuries are not rare. Problems with existing methods of modeling prevalence ratios include lack of convergence, overestimated standard errors, and extrapolation of simple univariate formulas to multivariable models. We compare two of the newer methods using simulated data and real data from SAS online examples.MethodsThe Robust Poisson method, which uses the Poisson distribution and a sandwich variance estimator, is compared to the log-binomial method, which uses the binomial distribution to obtain maximum likelihood estimates, using computer simulations and real data.ResultsFor very high prevalences and moderate sample size, the Robust Poisson method yields less biased estimates of the prevalence ratios than the log-binomial method. However, for moderate prevalences and moderate sample size, the log-binomial method yields slightly less biased estimates than the Robust Poisson method. In nearly all cases, the log-binomial method yielded slightly higher power and smaller standard errors than the Robust Poisson method.ConclusionAlthough the Robust Poisson often gives reasonable estimates of the prevalence ratio and is very easy to use, the log-binomial method results in less bias in most common situations, and because it fits the correct model and obtains maximum likelihood estimates, it generally results in slightly higher power, smaller standard errors, and, unlike the Robust Poisson, it always yields estimated prevalences between zero and one.

Pak-1 Expression Increases with Progression of Colorectal Carcinomas to Metastasis

Purpose: The p21-activated kinase-1 (Pak-1) promotes cell motility and invasiveness. Pak-1 is activated by the Rac, Rho, and Cdc42 small GTPases in response to a variety of stimuli including ras and phosphatidylinositol 3′-kinase/AKT pathway activation. Because Pak-1 plays a central role in regulating cell motility and invasiveness, we sought to determine whether Pak-1 may be involved in the malignant progression of colorectal carcinoma. Experimental Design: Pak-1 expression was examined by immunohistochemistry in archived tissues from normal human colons, tubular and tubulovillous adenomas, invasive adenocarcinomas (stages I-III/IV), and lymph node metastases (184 total specimens from 38 patients). Specific cytoplasmic immunostaining was evaluated for overall intensity and uniformity to derive a combined histoscore (stain intensity × percentage of epithelium stained). Results: Pak-1 expression was increased significantly with colorectal cancer progression from normal tissue to lymph node metastases (P < 0.0001). Furthermore, Pak-1 expression was increased significantly in adenomas, invasive carcinomas, and lymph node metastases compared with normal colon (P < 0.0001). Strikingly, Pak-1 expression was significantly higher in lymph node metastases than in invasive cancers, adenomas, or normal colon (P < 0.0001). Moreover, in patients with multiple lesions representing different stages of disease, Pak-1 expression was increased specifically in the most advanced lesions. Conclusions: This study demonstrates that Pak-1 expression is increased significantly with malignant progression of human colorectal carcinoma. These data, along with numerous functional studies demonstrating a central role for Pak-1 activity in tumor invasiveness and motility, implicate Pak-1 as an exciting target for therapy of colorectal carcinoma.

eIF4E Activation Is Commonly Elevated in Advanced Human Prostate Cancers and Significantly Related to Reduced Patient Survival

Elevated eukaryotic translation initiation factor 4E (eIF4E) function induces malignancy in experimental models by selectively enhancing translation of key malignancy-related mRNAs (c-myc and BCL-2). eIF4E activation may reflect increased eIF4E expression or phosphorylation of its inhibitory binding proteins (4E-BP). By immunohistochemical analyses of 148 tissues from 89 prostate cancer patients, we now show that both eIF4E expression and 4E-BP1 phosphorylation (p4E-BP1) are increased significantly, particularly in advanced prostate cancer versus benign prostatic hyperplasia tissues. Further, increased eIF4E and p4E-BP1 levels are significantly related to reduced patient survival, whereas uniform 4E-BP1 expression is significantly related to better patient survival. Both immunohistochemistry and Western blotting reveal that elevated eIF4E and p4E-BP1 are evident in the same prostate cancer tissues. In two distinct prostate cancer cell models, the progression to androgen independence also involves increased eIF4E activation. In these prostate cancer cells, reducing eIF4E expression with an eIF4E-specific antisense oligonucleotide currently in phase I clinical trials robustly induces apoptosis, regardless of cell cycle phase, and reduces expression of the eIF4E-regulated proteins BCL-2 and c-myc. Collectively, these data implicate eIF4E activation in prostate cancer and suggest that targeting eIF4E may be attractive for prostate cancer therapy.

A Practical Guide to Dose-Response Analyses and Risk Assessment in Occupational Epidemiology

Dose-response modeling in occupational epidemiology is usually motivated by questions of causal inference (eg, is there a monotonic increase of risk with increasing exposure?) or risk assessment (eg, how much excess risk exists at any given level of exposure?). We focus on several approaches to dose-response in occupational cohort studies. Categorical analyses are useful for detecting the shape of dose-response. However, they depend on the number and location of cutpoints and result in step functions rather than smooth curves. Restricted cubic splines and penalized splines are useful parametric techniques that provide smooth curves. Although splines can complement categorical analyses, they do not provide interpretable parameters. The shapes of these curves will depend on the degree of “smoothing” chosen by the analyst. We recommend combining categorical analyses and some type of smoother, with the goal of developing a reasonably simple parametric model. A simple parametric model should serve as the goal of dose-response analyses because (1) most “true” exposure response curves in nature may be reasonably simple, (2) a simple parametric model is easily communicated and used by others, and (3) a simple parametric model is the best tool for risk assessors and regulators seeking to estimate individual excess risks per unit of exposure. We discuss these issues and others, including whether the best model is always the one that fits the best, reasons to prefer a linear model for risk in the low-exposure region when conducting risk assessment, and common methods of calculating excess lifetime risk at a given exposure from epidemiologic results (eg, from rate ratios). Points are illustrated using data from a study of dioxin and cancer.

Evaluation of diabetes mellitus, serum glucose, and thyroid function among United States workers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin.

OBJECTIVE: Some studies suggest that exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) may affect glucose metabolism and thyroid function. To further assess the relation between exposure to TCDD and endocrine function, data from the largest morbidity study of industrial workers exposed to TCDD were examined. METHODS: A cross sectional study of workers employed > 15 years earlier in the manufacture of 2,4,5-trichlorophenol or one of its derivatives at two United States chemical plants was conducted. The referent group consisted of people with no occupational exposure to phenoxy herbicides and were recruited from the neighbourhoods where the workers lived. RESULTS: A total of 281 workers and 260 unexposed referents participated. The mean current serum lipid adjusted TCDD concentration among workers was 220 pg/g lipid, and among referents was 7 pg/g lipid (p < 0.05). The half life extrapolated TCDD concentrations (the estimated TCDD concentration when occupational exposure to TCDD stopped) among workers averaged 1900 pg/g lipid (range: not detected--30,000 pg/g lipid). Overall, the prevalence of diabetes mellitus was not significantly different between the workers and referents. Also, there was not a significant positive trend between prevalence of diabetes and increasing serum TCDD concentration. However, diabetes was found in six of 10 (60%) workers with current serum TCDD concentrations > 1500 pg/g lipid. After excluding subjects being treated for diabetes, workers in the group with the highest half life extrapolated TCDD concentrations had a significantly increased adjusted mean serum glucose concentration compared with referents (p = 0.03). Workers were also found to have a significantly higher adjusted mean free thyroxine index compared with referents (p = 0.02), especially among workers in the group with the highest half life extrapolated TCDD concentrations. However, no evidence was found that workers exposed to TCDD were at increased risk of thyroid disease. CONCLUSIONS: These findings provide modest evidence that exposure to TCDD may affect thyroid function and glucose metabolism.

Occupational Exposure Assessment in Carbon Nanotube and Nanofiber Primary and Secondary Manufacturers: Mobile Direct-Reading Sampling

UNLABELLED RESEARCH SIGNIFICANCE: Toxicological evidence suggests the potential for a wide range of health effects from exposure to carbon nanotubes (CNTs) and carbon nanofibers (CNFs). To date, there has been much focus on the use of direct-reading instruments (DRIs) to assess multiple airborne exposure metrics for potential exposures to CNTs and CNFs due to their ease of use and ability to provide instantaneous results. Still, uncertainty exists in the usefulness and interpretation of the data. To address this gap, air-monitoring was conducted at six sites identified as CNT and CNF manufacturers or users and results were compared with filter-based metrics. METHODS Particle number, respirable mass, and active surface area concentrations were monitored with a condensation particle counter, a photometer, and a diffusion charger, respectively. The instruments were placed on a mobile cart and used as area monitors in parallel with filter-based elemental carbon (EC) and electron microscopy samples. Repeat samples were collected on consecutive days, when possible, during the same processes. All instruments in this study are portable and routinely used for industrial hygiene sampling. RESULTS Differences were not observed among the various sampled processes compared with concurrent indoor or outdoor background samples while examining the different DRI exposure metrics. Such data were also inconsistent with results for filter-based samples collected concurrently at the same sites [Dahm MM, Evans DE, Schubauer-Berigan MK et al. (2012) Occupational exposure assessment in CNT and nanofiber primary and secondary manufacturers. Ann Occup Hyg; 56: 542-56]. Significant variability was seen between these processes as well as the indoor and outdoor backgrounds. However, no clear pattern emerged linking the DRI results to the EC or the microscopy data (CNT and CNF structure counts). CONCLUSIONS Overall, no consistent trends were seen among similar processes at the various sites. The DRI instruments employed were limited in their usefulness in assessing and quantifying potential exposures at the sampled sites but were helpful for hypothesis generation, control technology evaluations, and other air quality issues. The DRIs employed are nonspecific, aerosol monitors, and, therefore, subject to interferences. As such, it is necessary to collect samples for analysis by more selective, time-integrated, laboratory-based methods to confirm and quantify exposures.

Indoor particles and symptoms among office workers: results from a double-blind cross-over study

Background. We studied the effects of removing small airborne particles in an office building without unusual contaminant sources or occupant complaints. Methods. We conducted a double-blind crossover study of enhanced particle filtration in an office building in the Midwest United States in 1993. We replaced standard particle filters, in separate ventilation systems on two floors, with highly efficient filters on alternate floors weekly over 4 weeks. Repeated-measures models were used to analyze data from weekly worker questionnaires and multiple environmental measurements. Results. Bioaerosol concentrations were low. Enhanced filtration reduced concentrations of the smallest airborne particles by 94%. This reduction was not associated with reduced symptoms among the 396 respondents, but three performance-related mental states improved; for example, the confusion scale decreased (−3.7%; 95% confidence limits (CL) = −6.5, −0.9). Most environmental dissatisfaction variables also improved;eg, “stuffy” air, −5.3% (95% CL = −10.3, −0.4). Cooler temperatures within the recommended comfort range were associated with remarkably large improvement in most outcomes; for example, chest tightness decreased −23.4% (95% CL = −38.1, −8.7) for every 1°C decrease. Conclusions. Benefits of enhanced filtration require assessment in buildings with higher particulate contaminant levels in studies controlling for temperature effects. Benefits from lower indoor temperatures need confirmation.