A weighted Cox model for modelling time‐dependent exposures in the analysis of case–control studies

Leffondré, Karen; Wynant, Willy; Cao, Zhirong; Abrahamowicz, Michał; Heinze, Georg; Siemiatycki, Jack

doi:10.1002/sim.3764

Cited by 21 publications

(24 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, for subjects who alternated periods of high and low levels of exposure, the MIE value was the same whether high levels of exposure occurred either early or later in life. This issue will be further investigated using some recent approaches [35,36].…”

mentioning

confidence: 99%

Temporal patterns of occupational asbestos exposure and risk of pleural mesothelioma

Lacourt¹,

Leffondré²,

Gramond³

et al. 2011

Eur Respir J

View full text Add to dashboard Cite

Asbestos is the primary cause of pleural mesothelioma (PM). The objective of this study was to elucidate the importance of different temporal patterns of occupational asbestos exposure on the risk of PM using case-control data in male subjects.Cases were selected from a French case-control study conducted in 1987-1993 and the French National Mesothelioma Surveillance Program in 1998-2006. Population controls were frequency matched to cases by year of birth. Occupational asbestos exposure was evaluated with a jobexposure matrix. The dose-response relationships were estimated using restricted cubic spline functions in logistic regression models.A total of 2,466 ever-asbestos-exposed males (1,041 cases and 1,425 controls) were used. After adjustment for intensity and total duration of occupational asbestos exposure, the risk of PM was lower for subjects first exposed after the age of 20 yrs and continued to increase until 30 yrs after cessation of exposure. The effect of total duration of exposure decreased when age at first exposure and time since last exposure increased.These results, based on a large population-based case-control study, underline the need to take into account the temporal pattern of exposure on risk assessment.

show abstract

mentioning

confidence: 99%

Temporal patterns of occupational asbestos exposure and risk of pleural mesothelioma

Lacourt¹,

Leffondré²,

Gramond³

et al. 2011

Eur Respir J

View full text Add to dashboard Cite

show abstract

“…A time-dependent weighted Cox (WC) model has recently been proposed to incorporate this dynamic information on exposure, in order to more accurately estimate the effect of time-dependent exposures in population-based case-control studies [4]. The WC model consists in using age as the time axis and weighting cases and controls according to their case-control status and the age conditional probabilities of developing the disease in the source population.…”

Section: Introductionmentioning

confidence: 99%

“…The weights proposed in the WC model are therefore time-dependent and estimated from data of the source population. A simulation study showed that the WC model improved the accuracy of the regression parameters estimates of time-dependent exposure variables as compared with standard logistic regression with fixed-in-time covariates [4]. However, the average robust sandwich variance estimates based on dfbetas residuals [5] were systematically lower than the empirical variance of the parameter estimates, which resulted in too narrow confidence intervals (CI) and low coverage probabilities [4].…”

Section: Introductionmentioning

confidence: 99%

“…However, the simulation study considered only fixed-in-time covariates and simple selection probabilities that did not reflect the more complex sampling scheme of population-based case-control studies. It is therefore unclear how the superpopulation variance estimator would perform for the estimation of the effects of time-dependent covariates using the specific estimated time-dependent weights proposed in the WC model [4]. In addition, for further applications to population-based case-control data, it would be important to clarify the performance of the WC model, as compared with standard logistic regression analyses, for estimating the effects of several correlated temporal patterns of protracted exposures.…”

Section: Introductionmentioning

confidence: 99%

“…The first objective of the present study is to investigate through extensive simulations the accuracy of the Lin variance estimator for estimating the effects of time-varying covariates in case-control data, using the weights proposed in the WC model [4]. The second objective is to compare the estimates from the WC model and standard logistic regression for estimating the effects of selected correlated temporal aspects of exposure with detailed information on exposure history.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On the proportional hazards model for occupational and environmental case-control analyses

Gauvin

Lacourt

Leffondré

2013

BMC Med Res Methodol

View full text Add to dashboard Cite

BackgroundCase-control studies are generally designed to investigate the effect of exposures on the risk of a disease. Detailed information on past exposures is collected at the time of study. However, only the cumulated value of the exposure at the index date is usually used in logistic regression. A weighted Cox (WC) model has been proposed to estimate the effects of time-dependent exposures. The weights depend on the age conditional probabilities to develop the disease in the source population. While the WC model provided more accurate estimates of the effect of time-dependent covariates than standard logistic regression, the robust sandwich variance estimates were lower than the empirical variance, resulting in a low coverage probability of confidence intervals. The objectives of the present study were to investigate through simulations a new variance estimator and to compare the estimates from the WC model and standard logistic regression for estimating the effects of correlated temporal aspects of exposure with detailed information on exposure history.MethodWe proposed a new variance estimator using a superpopulation approach, and compared its accuracy to the robust sandwich variance estimator. The full exposure histories of source populations were generated and case-control studies were simulated within each source population. Different models with selected time-dependent aspects of exposure such as intensity, duration, and time since cessation were considered. The performances of the WC model using the two variance estimators were compared to standard logistic regression. The results of the different models were finally compared for estimating the effects of correlated aspects of occupational exposure to asbestos on the risk of mesothelioma, using population-based case-control data.ResultsThe superpopulation variance estimator provided better estimates than the robust sandwich variance estimator and the WC model provided accurate estimates of the effects of correlated aspects of temporal patterns of exposure.ConclusionThe WC model with the superpopulation variance estimator provides an alternative analytical approach for estimating the effects of time-varying exposures with detailed history exposure information in case-control studies, especially if many subjects have time-varying exposure intensity over lifetime, and if only one control is available for each case.

show abstract

Evaluation of Association Methods for Analysing Modifiers of Disease Risk in Carriers of High‐Risk Mutations

Barnes

Lee

Investigators

et al. 2012

Genetic Epidemiology

View full text Add to dashboard Cite

There is considerable evidence indicating that disease risk in carriers of high-risk mutations (e.g. BRCA1 and BRCA2) varies by other genetic factors. Such mutations tend to be rare in the population and studies of genetic modifiers of risk have focused on sampling mutation carriers through clinical genetics centres. Genetic testing targets affected individuals from high-risk families, making ascertainment of mutation carriers non-random with respect to disease phenotype. Standard analytical methods can lead to biased estimates of associations. Methods proposed to address this problem include a weighted-cohort (WC) and retrospective likelihood (RL) approach. Their performance has not been evaluated systematically. We evaluate these methods by simulation and extend the RL to analysing associations of two diseases simultaneously (competing risks RL-CRRL). The standard cohort approach (Cox regression) yielded the most biased risk ratio (RR) estimates (relative bias-RB: -25% to -17%) and had the lowest power. The WC and RL approaches provided similar RR estimates, were least biased (RB: -2.6% to 2.5%), and had the lowest mean-squared errors. The RL method generally had more power than WC. When analysing associations with two diseases, ignoring a potential association with one disease leads to inflated type I errors for inferences with respect to the second disease and biased RR estimates. The CRRL generally gave unbiased RR estimates for both disease risks and had correct nominal type I errors. These methods are illustrated by analyses of genetic modifiers of breast and ovarian cancer risk for BRCA1 and BRCA2 mutation carriers.

show abstract

A weighted Cox model for modelling time‐dependent exposures in the analysis of case–control studies

Cited by 21 publications

References 29 publications

Temporal patterns of occupational asbestos exposure and risk of pleural mesothelioma

Temporal patterns of occupational asbestos exposure and risk of pleural mesothelioma

On the proportional hazards model for occupational and environmental case-control analyses

Evaluation of Association Methods for Analysing Modifiers of Disease Risk in Carriers of High‐Risk Mutations

Contact Info

Product

Resources

About