Bayesian Approach to "Healthy Worker Hire Effect" in Standardized Mortality Ratio Analysis

Comparison of internal and external reference populations for occupational cancer surveillance in a cohort drawn from a diverse workforce

OBJECTIVES

Prior analyses of the Occupational Disease Surveillance System (ODSS) have compared cancer rates using internal referent groups. As an exploratory analysis, we sought to estimate cancer risk using general population reference rates to evaluate the impact that the comparison population has on findings from our surveillance program.

METHODS

A cohort of approximately 2.3 million workers in Ontario, Canada with an accepted lost-time workers' compensation claim were followed for all cancer diagnoses between 1983 and 2018. Standardized incidence ratios (SIRs) and 95% confidence intervals were calculated for workers in specific occupational groups using (1) all other workers in the ODSS cohort, and (2) the general population of Ontario.

RESULTS

SIRs using the general population reference group were generally equal to or modestly lower compared to SIRs using the internal reference group. Within occupation groups, SIRs had a discordant direction of association (increased rate in the internal comparison and decreased in the external comparison) for some cancer sites including urinary, prostate, and colorectal.

CONCLUSIONS

Findings emphasize the importance of the choice of reference group when evaluating cancer risks in large occupational surveillance cohorts. Importantly, the magnitude of confounding and the healthy worker hire bias may depend on the occupation group and cancer site of interest.

Mortality due to respiratory system disease and lung cancer among female workers exposed to chrysotile in Eastern China: A cross-sectional study

Female workers in the asbestos processing industry of Eastern China are at high risk of developing multiple types of cancer, and more data are urgently needed to better understand and address this issue. Death certificate data were selected from an asbestos processing city in China from 2005 to 2006. Information was investigated using the relatives of those individuals who had died as sources of information. Individuals were classified into one of three asbestos exposure levels. Standardized mortality ratio and 95% confidence interval were calculated. A total of 2,964 individual deaths were identified from 2005 to 2006; of these, 21.4% were occupationally exposed to asbestos. The main cause of death was circulatory system diseases (21.2%). The proportion of individuals with respiratory system diseases increased by age among each exposure subgroup (P_trend < 0.01). Among females, a significant trend was observed between increased asbestos exposure and mortality due to respiratory system diseases and lung cancer. Our study indicated that asbestos exposure was associated with excess mortality from lung cancer and respiratory diseases, particularly among female workers in an asbestos processing area in Eastern China.

Correction of odds ratios in case-control studies for exposure misclassification with partial knowledge of the degree of agreement among experts who assessed exposures

OBJECTIVES

Estimates of association between exposures and diseases are often distorted by error in exposure classification. When the validity of exposure assessment is known, this can be used to adjust these estimates. When exposure is assessed by experts, even if validity is not known, we sometimes have information about interrater reliability. We present a Bayesian method for translating the knowledge of interrater reliability, which is often available, into knowledge about validity, which is often needed but not directly available, and applying this to correct odds ratios (OR).

METHODS

The method allows for inclusion of observed potential confounders in the analysis, as is common in regression-based control for confounding. Our method uses a novel type of prior on sensitivity and specificity. The approach is illustrated with data from a case-control study of lung cancer risk and occupational exposure to diesel engine emissions, in which exposure assessment was made by detailed job history interviews with study subjects followed by expert judgement.

RESULTS

Using interrater agreement measured by kappas (κ), we estimate sensitivity and specificity of exposure assessment and derive misclassification-corrected confounder-adjusted OR. Misclassification-corrected and confounder-adjusted OR obtained with the most defensible prior had a posterior distribution centre of 1.6 with 95% credible interval (Crl) 1.1 to 2.6. This was on average greater in magnitude than frequentist point estimate of 1.3 (95% Crl 1.0 to 1.7).

CONCLUSIONS

The method yields insights into the degree of exposure misclassification and appears to reduce attenuation bias due to misclassification of exposure while the estimated uncertainty increased.