Case-control Study Of Exposure Research Articles

Night-shift work and breast cancer – a systematic review and meta-analysis

The aim of this review was to synthesize the evidence on the potential relationship between nightshift work and breast cancer. We searched multiple databases for studies comparing women in shift work to those with no-shift work reporting incidence of breast cancer. We calculated incremental risk ratios (RR) per five years of night-shift work and per 300 night shift increases in exposure and combined these in a random effects dose-response meta-analysis. We assessed study quality in ten domains of bias. We identified 16 studies: 12 case-control and 4 cohort studies. There was a 9% risk increase per five years of night-shift work exposure in case-control studies [RR 1.09, 95% confidence interval (95% CI) 1.02-1.20; I (2) = 37%, 9 studies], but not in cohort studies (RR 1.01, 95% CI 0.97-1.05; I (2) = 53%, 3 studies). Heterogeneity was significant overall (I (2) = 55%, 12 studies). Results for 300 night shifts were similar (RR 1.04, 95% CI 1.00-1.10; I (2) = 58%, 8 studies). Sensitivity analysis using exposure transformations such as cubic splines, a fixed-effect model, or including only better quality studies did not change the results. None of the 16 studies had a low risk of bias, and 6 studies had a moderate risk. Based on the low quality of exposure data and the difference in effect by study design, our findings indicate insufficient evidence for a link between night-shift work and breast cancer. Objective prospective exposure measurement is needed in future studies.

Inside the black box: starting to uncover the underlying decision rules used in a one-by-one expert assessment of occupational exposure in case-control studies

ObjectivesEvaluating occupational exposures in population-based case-control studies often requires exposure assessors to review each study participant's reported occupational information job-by-job to derive exposure estimates. Although such assessments likely have underlying...

A Bayesian mixture modeling approach for assessing the effects of correlated exposures in case-control studies

Predisposition to a disease is usually caused by cumulative effects of a multitude of exposures and lifestyle factors in combination with individual susceptibility. Failure to include all relevant variables may result in biased risk estimates and decreased power, whereas inclusion of all variables may lead to computational difficulties, especially when variables are correlated. We describe a Bayesian Mixture Model (BMM) incorporating a variable-selection prior and compared its performance with logistic multiple regression model (LM) in simulated case-control data with up to twenty exposures with varying prevalences and correlations. In addition, as a practical example we re analyzed data on male infertility and occupational exposures (Chaps-UK). BMM mean-squared errors (MSE) were smaller than of the LM, and were independent of the number of model parameters. BMM type I errors were minimal (≤1), whereas for the LM this increased with the number of parameters and correlation between exposures. The numbers of type II errors were comparable. Re analysis of Chaps-UK data demonstrated more convincingly than by using a LM that occupational exposure to glycol ethers and VOCs are likely risk factors for male infertility. This BMM proves an appealing alternative to standard logistic regression when dealing with the analysis of (correlated) exposures in case-control studies.

WILL WOMEN DIAGNOSED WITH BREAST CANCER PROVIDE BIOLOGICAL SAMPLES FOR FUTURE RESEARCH PURPOSES?

Background and Aims: Biological samples are often used to assess exposures in case control studies of cancer risk. Little is known about the response rates for biological sample provision, especially in younger age groups, and how these are affected by other demographic or lifestyle variables. A population based study of breast cancer patients was conducted to estimate these rates and to obtain information on the recruitment of friend/colleague controls. Methods: Breast cancer cases diagnosed April-May 2010 were recruited to determine the proportion willing to provide biological samples (morning or 24 hour urine, saliva, blood) and contact information for similarly aged women for control recruitment. Cases (n=417) were women 25-74 years of age identified through the Ontario Cancer Registry and 67% (n=278) of those completed a mailed questionnaire. Results: Women indicated they were willing to provide blood samples, by visiting a clinic (62%) or having a nurse visit the home (61%). They also were willing to provide saliva samples (73%) and morning (66%) or 24-hr urine samples (52%). Compared to older women, younger women (•45) were consistently more likely to agree to provide any biological samples and were 3.1 times (OR) more likely to report they would collect a morning urine sample (95% CL: 1.15-8.35). Only 26% of cases indicated they would be willing to give contact information for similarly aged women (controls) to provide samples in future studies. Educated women were more likely to agree to provide biological samples, and women who consumed alcohol were more likely to agree to provide contact information than non-drinkers. Ethnicity, income, body mass index, and smoking status did not influence these results. Conclusions: Reasonable response rates for biological sample collection can be expected in future case controls studies of breast cancer (especially among younger women), however, other methods of control selection must be devised.

The scope for biased recall of risk-factor exposure in case-control studies: Evidence from a cohort study of Scottish men

Case-control studies are prone to recall bias, a participant's case-control status influencing their recall of exposure to risk factors. We aimed to demonstrate empirically the scope for this bias. Two thousand five hundred and fifty men without coronary heart disease at enrollment to a prospective cohort study underwent two health assessments, about 5 years apart. The association between the development of coronary heart disease in the intervening period and changes in reported stress and cigarette smoking were investigated. Men admitted to hospital with coronary heart disease reported a greater increase in psychological stress (p=0.032) and greater cessation of smoking (22% vs. 10%; p=0.007) than men not admitted. Consequently, when exposure data are collected at the end rather than at the start of the follow-up period, coronary heart disease is observed to be more strongly associated with psychological stress, and more weakly associated with smoking. At the time when a case-control study is conducted, levels of exposure to risk factors will have been influenced by disease development. When participants are asked about their level of exposure for a previous time period, recall is likely to be influenced by present outcome and exposure status, especially when psychological states are being investigated.

Estimation of time‐dependent rate ratios in case‐control studies: comparison of two approaches for exposure assessment

In pharmacoepidemiology, it is well recognized that the rate of adverse events may vary as a function of the cumulative duration of the drug exposure and/or the time since the end of the exposure. In case-control studies, two different approaches have been used to estimate temporal effects of drug exposure: the time-windows (T-Ws) approach and the duration-specific (D-S) approach. We decided to conduct a simulation study to compare the two approaches when the rate ratios (RRs) vary as a function of the cumulative duration of exposure and/or the time since the end of exposure. We generated three cohorts of 500,000 individuals in which the rate of the event was varying as a function of the cumulative duration of exposure and the time since the end of exposure. For each cohort, a nested case-control analysis was performed using both the D-S and the T-Ws approaches. In the T-Ws approach, a RR is estimated within specific periods of time prior to the outcome, while a RR is estimated within periods of cumulative duration of exposure and time since the end of exposure in the D-S approach. We found that the RRs obtained from the D-S approach exactly corresponded to the RRs obtained from the cohort analyses, while the RRs obtained from the T-Ws approach generally not. RRs obtained from the T-Ws approach were difficult to interpret in terms of the effect of the duration and timing of the exposure. The D-S approach should be used to investigate the duration-related effects of exposure in case-control studies.

Comparison of asbestos exposure assessments by next-of-kin respondents, by an occupational hygienist, and by a job-exposure matrix from the National Occupational Hazard Survey

Assessments of occupational exposures in case-control studies of rapidly fatal illnesses often rely on data from next-of-kin respondents, which may be inaccurate. Three methods for assessing exposure to asbestos from case-control data on mesothelioma, including next-of-kin assessment, expert assessment, and use of a generic job-exposure matrix (JEM). Interview data [Spirtas et al. (1994): Occup Environ Med 51:804-811] were reviewed to determine exposure status by an occupational hygienist (C.R.) who was unaware of disease status. Exposure odds ratios were calculated using standard methods, and measures of agreement included the kappa statistic and conditional and marginal odds ratios. Expert assessment detected higher proportions of exposed subjects than the next-of-kin respondents or JEM methods. The disease-exposure odds ratios were highest for respondents, perhaps because of recall bias, and lowest for the JEM method. The agreement was highest between the respondent and expert assessments. A combination of respondent's assessment and JEM assessment led to the best prediction of the expert's assessment. Results for spouse respondents were similar to those for other "next-of-kin" respondents. Expert assessments were the most plausible, but the data indicate that disease associations could also be detected with the other exposure assessment methods. Using some combination of the proxy respondent's assessment and the JEM assessment, one can predict the expert's assessment. A strategy that relied on the respondent's assessment when it was positive and otherwise obtained an expert assessment could reduce costs with little error, compared to expert assessment on all subjects.

Agreement of job-exposure matrix (JEM) assessed exposure and self-reported exposure among adult leukemia patients and controls in Shanghai.

Estimating a person's history of occupational exposure in case-control studies is difficult. Percent agreement between selected self-reported occupational exposures and job-exposure matrix (JEM) exposure assessment for all participants and various subgroups of a population-based case-control interview study of 486 leukemia subjects and 502 healthy controls in Shanghai was evaluated. With JEM as the "gold standard," the sensitivities for self-reported exposures ranged from 0.75 to 0.98. However, that for pesticide exposure was 0.44 in subjects >51 years old. Self-reported exposures specificities ranged from 0.87 to 0.99. Agreement between self-reported exposures and JEM assessment was good (kappa coefficients [kappa]: 0.48-0.84). Variations in agreement for benzene exposure between males and females as well as between the direct interview and surrogate interview subgroups were observed. The levels of agreement between self-report and JEM in this study suggest that self-reported exposures are a suitable method for assessing occupational exposures in this population.

An Empirical Investigation of Exposure Measurement Bias and Its Components in Case-Control Studies

This study investigated bias due to misclassification of exposure in case control studies. Using a metaanalysis of data from a number of case control studies, in which a possibly misclassified exposure was validated against a more reliable one, estimates of misclassification indices and bias were obtained. The estimates were used to investigate whether misclassification indices, in particular, sensitivity and specificity and so-called quality indices, are typically nondifferential with respect to cases and controls. It is concluded that quality indices do not show any tendency to be either lower or higher in cases than controls. On the other hand, sensitivity tends to be higher in cases than controls and specificity lower. Estimates of misclassification bias may be either positive or negative and are consistent with random variation; there is little to suggest that bias is present in the studies analyzed.

RESPONSE: Re: Multicenter Case-Control Study of Exposure to Environmental Tobacco Smoke and Lung Cancer in Europe

RESPONSE: Re: Multicenter Case-Control Study of Exposure to Environmental Tobacco Smoke and Lung Cancer in Europe

A graphical display useful for meta-analysis

Graphical methods are frequently used in meta-analysis to summarize their results and to explore potential sources of heterogeneity across studies. In this paper, we illustrate a graphical method for meta-analysis of studies with dichotomous exposures and outcomes that complements other graphical and analytical approaches to meta-analysis. In prospective studies, the proportion of cases among the unexposed is plotted on the horizontal axis versus the proportion of cases among the exposed on the vertical axis. Contour lines for equal values of relative risk, odds ratio or risk difference and for the combined estimate of effect and its confidence interval are then superimposed on the graph. In case-control studies, the proportion of exposed controls is plotted on the horizontal axis versus the proportion of exposed cases on the vertical axis, although only the contour lines of equal odds ratios yield direct epidemiological interpretation. In these graphs, the distribution of the individual estimates of effect with respect to the contour lines offers a due as to the adequacy of the scale of measurement used (additive or multiplicative). This graphical method also permits direct inspection of the range of disease frequency in follow-up studies and of the range of exposure in case-control studies. Its use is illustrated with the aid of 3 examples derived from the literature.

Assessment of exposure misclassification bias in case-control studies using validation data

It has recently been shown that the bias due to misclassification of a binary exposure in case-control studies can be simply expressed in terms of so-called quality indices. Quality indices are measures of the degree of misclassification that are symmetric with respect to the true and measured exposure. They can be estimated from sub-study validation data and used to estimate the bias. Bias is not present if quality indices are nondifferential with respect to cases and controls. In this article the sampling distribution of these indices is investigated by developing formulae for their standard errors and these formulae are used to derive the standard error of the estimated bias. Simulation studies are used to confirm the validity of the formulae. The formulae are used to assess the bias in two studies of risk factors for Sudden Infant Death Syndrome (SIDS) and can be applied to the assessment of diagnostic tests.

Imprecise exposure assessment and the sample size requirements of case‐control studies of residential magnetic field exposure and cancer in adults

A computer program simulating case-control studies is described. It is used to estimate the minimum sample size required and to assess how this is affected by imprecise exposure assessment. In particular, the consequences of neglecting measurements of nonresidential exposure in case-control studies of residentially exposed adults are investigated. According to this model, while the consequent loss of power is not as large as was predicted by algebraic methods, it would be unwise to neglect it when planning a study.

Misclassification of exposure in case-control studies: assessment by quality indices.

Misclassification of a binary exposure in case-control studies is usually assessed by sensitivity and specificity of the measured exposure. Sensitivity and specificity do not adequately measure the degree of misclassification because they assess only the proportion of truly exposed (or unexposed) subjects who are misclassified by the defective measurement. They do not account for the proportion of subjects, categorized by the defective measurement as either exposed or unexposed, who are misclassified, in other words, the predictive value of the measurement. A more appropriate way of measuring misclassification is by "quality indices" that take both of these criteria into account and that are essentially rescaled sensitivity and specificity, or predictive value, measures. I present the relation between measured and actual odds ratios in terms of quality indices. If quality indices are nondifferential or proportional, there is no misclassification bias to the odds ratio. The relation offers a new approach to correcting measured odds ratios.

Hypertensive End-Stage Renal Disease in Blacks: The Role of End-Stage Renal Disease Surveillance

The end-stage renal disease (ESRD) networks and the United States Renal Data System recently have described the epidemiology of ESRD and broadened our appreciation of the impact of ESRD in the United States. This surveillance system also can be used to study the occurrence and control of ESRD. Among the epidemiologic study designs available to conduct research on the etiology of renal failure, case-control studies are uniquely suited for this purpose. The case of hypertensive ESRD illustrates this concept; risk factors for hypertensive ESRD, as a hypothetical exposure in case-control studies, are briefly described. Case-control studies are an efficient and readily used means to study causes of renal failure.

Efficiency Loss from Categorizing Quantitative Exposures into Qualitative Exposures in Case-Control Studies

In the analysis of data from case-control studies, quantitative exposure variables are frequently categorized into qualitative exposure variables, such as quarters. The qualitative exposure variables may be scalar variables that take the median values of each quantile interval, or they may be vectors of indicator variables that represent each quantile interval. In a qualitative analysis, the scalar variables may be used to test the dose-response relation, while the indicator variables may be used to estimate odds ratios for each higher quantile interval versus the lowest. Qualitative analysis, implicitly and explicitly documented by many epidemiologists and biostatisticians, has several desirable advantages (including simple interpretation and robustness in the presence of a misspecified model or outlier values). In a quantitative analysis, the quantitative exposure variables may be directly regressed to test the dose-response relation, as well as to estimate odds ratios of interest. As this paper demonstrates, quantitative analysis is generally more efficient than qualitative analysis. Through a Monte Carlo simulation study, the authors estimated the loss of efficiency that results from categorizing a quantitative exposure variable by quartiles in case-control studies with a total of 200 cases and 200 controls. In the analysis of the dose-response relation, this loss is about 30% or more; the percentage may reach about 50% when the odds ratio for the fourth quartile interval versus the lowest is around 4. In estimating odds ratios, the loss of efficiency for the second, third, and fourth quartile intervals versus the lowest is around 90%, 75%, and 40%, respectively. The authors consider the pros and cons of each analytic approach, and they recommend that 1) qualitative analysis be used initially to estimate the odds ratios for each higher quantile interval versus the lowest to examine the dose-response relation and determine the appropriateness of the assumed underlying model; and 2) quantitative analysis be used to test the dose-response relation under a plausible log odds ratio model.

Defining Exposure in Case-Control Studies: a New Approach

Many epidemiologists conducting case-control studies choose to dichotomize their exposure data to make the analysis of the data easier and its presentation more straightforward. The choice of a specific rule for dichotomization can have a large effect on the outcome measure, the odds ratio, although this effect is rarely studied. The authors present a graphic approach for exploring this effect, the quantile-quantile (Q-Q) plot. By examining a Q-Q plot, an investigator simultaneously gains information about the distribution of exposures among cases, the distribution of exposures among controls, and odds ratios at all possible cutpoints and their standard errors. In addition, by finding the slope at each point along the Q-Q curve, it is possible to estimate the rate ratios for each possible value of exposure. The authors view the use of the Q-Q plot as an exploratory tool. It enables the investigator to become more familiar with the data and check for irregularities such as outliers, nonlinearities, or nonmonotonic dose-response curves, and idiosyncratic variations of the odds ratio. The authors present an example evaluating the risk of low birth weight as a function of mother's age for Boston births in 1984.

Methodologic issues in exposure assessment for case‐control studies of cancer and herbicides

Epidemiologic studies of cancer and exposure to herbicides have shown puzzling inconsistencies. Exposure-response gradients have been reported for non-Hodgkin's lymphoma in Sweden and Kansas, but no significant associations were seen in New Zealand or Washington State. Subjects in these studies were categorized by exposure using information obtained primarily by interview. A number of questions can be raised regarding the reliability and validity of such an exposure assessment. We examined procedures used to assess pesticide exposures in case-control studies of cancer to evaluate their limitations and their probable effects on risk estimates. Except for case recall bias, problems of misclassification in these studies would tend to bias risk estimates toward the null and dilute exposure-response gradients. These problems are, therefore, unlikely explanations for the positive associations between cancer and herbicide use noted in some investigations. A tendency for false-negative findings, however, is not reassuring, and improvements in exposure assessment are needed if epidemiologic investigations are to continue to provide reliable information on the relationships of cancer and pesticide exposure.

Evaluating Occupation and Industry Separately to Assess Exposures in Case-Control Studies

Abstract In most occupational case-control studies, assessment of exposures is based on occupation and industry due to the lack of any other exposure information. The traditional approach to assessing exposures to specific substances is to estimate exposure for each occupation/industry combination. A modified procedure that substantially reduces the effort necessary to complete such assessments is described. In this procedure, exposure levels (none, low, medium, and high) are estimated for each occupation and industry separately. Numeric weights are assigned to these exposures levels. Final estimates for each occupation/industry combination are derived using exposure weights for occupations and industries. This method was applied to a case-control data set to estimate levels of silica exposure for the work histories of subject. The traditional method required 13,445 individual evaluations while the new approach required only 2,000 estimates. Results from the new approach were compared with those from the ...