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

Does recall bias explain the association of mood disorders with workplace harassment?

Purpose

To determine the contribution of recall bias to the observed excess in mental ill-health in those reporting harassment at work.

Methods

A prospective cohort of 1885 workers in welding and electrical trades was contacted every six months for up to 5 years, asking whether they were currently anxious or depressed and whether this was made worse by work. Only at the end of the study did we ask about any workplace harassment they had experienced at work. We elicited sensitivity and specificity of self-reported bullying from published reliability studies and formulated priors that reflect the possibility of over-reporting of workplace harassment (exposure) by those whose anxiety or depression was reported to be made worse by work (cases). We applied the resulting misclassification models to probabilistic bias analysis (PBA) of relative risks.

Results

We observe that PBA implies that it is unlikely that biased misclassification due to the study subjects' states of mind could have caused the entire observed association. Indeed, the results demonstrated that doubling of risk of anxiety or depression following workplace harassment is plausible, with the unadjusted relative risk attenuated with understated uncertainty.

Conclusions

It seems unlikely that risk of anxiety or depression following workplace harassment can be explained by the form of recall bias that we proposed.

Does adjustment for non-differential misclassification of dichotomous exposure induce positive bias if there is no true association?

This article is a response to an off-the-record discussion that I had at an international meeting of epidemiologists more than decade ago. It centered on a concern, perhaps widely spread, that adjustment for exposure misclassification can induce a false positive result. I trace the possible history of this supposition and test it in a simulated case-control study under the assumption of non-differential misclassification of binary exposure, in which a Bayesian adjustment is applied. Probabilistic bias analysis is also briefly considered. The main conclusion is that adjustment for the presumed non-differential exposure misclassification of dichotomous does not "induce" positive associations, especially if the focus of the interpretation of the result is taken away from the point estimate. The misconception about positive bias induced by adjustment for exposure misclassification, if more clearly explained during the training of epidemiologists, may promote appropriate (and wider) use of the adjustment techniques. The simple message that can be derived from this paper is: "Exposure misclassification as a tractable problem that deserves much more attention than just a typical qualitative throw-away discussion".

Assessing exposure to outdoor advertisement for products high in fat, salt and sugar (HFSS); is self-reported exposure a useful exposure metric?

BACKGROUND

Exposure to advertising of unhealthy commodities such as fast-food and gambling is recognised as a risk factor for developing non-communicable diseases. Assessment of the impact of such advertisement and the evaluation of the impact of any policies to restrict such advertisements on public health are reliant on the quality of the exposure assessment. A straightforward method for assessing exposure is to ask people whether they noticed any such advertisements in their neighbourhoods. However, the validity of this method is unclear. We assessed the associations between measured exposure to outdoor advertising, self-reported exposure, and self-reported consumption.

METHODS

We collected exposure information in January-March 2022 using two methods: (i) through a resident survey investigating advertising and consumption of unhealthy products, distributed across Bristol and neighbouring South Gloucestershire, and (ii) through in-person auditing. Self-reported exposure was obtained from the resident survey (N = 2,560) and measured exposure from photos obtained for all Council owned advertisement sites (N = 973 bus stops). Both data sources were geographically linked at lower-super-output-area level. Reporting ratios (RRs), 95% confidence intervals (CIs), and Cohen's kappas, are presented.

RESULTS

24% of advertisements displayed food and/or drink advertising. Bristol respondents in neighbourhoods displaying food/drink adverts were more likely to also report seeing these adverts compared to those in neighbourhoods without food/drink adverts (59% vs. 51%, RR = 1.15, 95%CI 1.01-1.31). There was no such association in South Gloucestershire (26% vs. 32%, RR = 0.82, 95%CI 0.58-1.14). Respondents in both Bristol and South Gloucestershire who recalled seeing advertising for unhealthy food and drink products were more likely to consume them (e.g. for fast-food: 22% vs. 11%, RR = 2.01, 95%CI 1.68-2.42). There was no such association between measured food and drink adverts in respondents' local areas and self-reported consumption of HFSS product (90.1% vs. 90.7%, RR = 0.99, 95%CI 0.96-1.03).

CONCLUSIONS

Self-reported outdoor advertisement exposure is correlated with measured exposure, making this a useful methodology for population studies. It has the added advantage that it correlates with consumption. However, given that measurement error can be significant and self-reported exposure is known to be susceptible to various biases, inferences from studies using this exposure metric should be made with caution.

Automated Coding of Job Descriptions From a General Population Study: Overview of Existing Tools, Their Application and Comparison

OBJECTIVES

Automatic job coding tools were developed to reduce the laborious task of manually assigning job codes based on free-text job descriptions in census and survey data sources, including large occupational health studies. The objective of this study is to provide a case study of comparative performance of job coding and JEM (Job-Exposure Matrix)-assigned exposures agreement using existing coding tools.

METHODS

We compared three automatic job coding tools [AUTONOC, CASCOT (Computer-Assisted Structured Coding Tool), and LabourR], which were selected based on availability, coding of English free-text into coding systems closely related to the 1988 version of the International Standard Classification of Occupations (ISCO-88), and capability to perform batch coding. We used manually coded job histories from the AsiaLymph case-control study that were translated into English prior to auto-coding to assess their performance. We applied two general population JEMs to assess agreement at exposure level. Percent agreement and PABAK (Prevalence-Adjusted Bias-Adjusted Kappa) were used to compare the agreement of results from manual coders and automatic coding tools.

RESULTS

The coding per cent agreement among the three tools ranged from 17.7 to 26.0% for exact matches at the most detailed 4-digit ISCO-88 level. The agreement was better at a more general level of job coding (e.g. 43.8-58.1% in 1-digit ISCO-88), and in exposure assignments (median values of PABAK coefficient ranging 0.69-0.78 across 12 JEM-assigned exposures). Based on our testing data, CASCOT was found to outperform others in terms of better agreement in both job coding (26% 4-digit agreement) and exposure assignment (median kappa 0.61).

CONCLUSIONS

In this study, we observed that agreement on job coding was generally low for the three tools but noted a higher degree of agreement in assigned exposures. The results indicate the need for study-specific evaluations prior to their automatic use in general population studies, as well as improvements in the evaluated automatic coding tools.

Impact of Variability in Job Coding on Reliability in Exposure Estimates Obtained via a Job-Exposure Matrix

OBJECTIVES

The use of a job-exposure matrix (JEM) to assess exposure to potential health hazards in occupational epidemiological studies requires coding each participant's job history to a standard occupation and/or industry classification system recognized by the JEM. The objectives of this study were to assess the impact of inter-coder variability in job coding on reliability in exposure estimates derived from linking the job codes to the Canadian job-exposure matrix (CANJEM) and to identify influent parameters.

METHOD

Two trained coders independently coded 1000 jobs sampled from a population-based case-control study to the ISCO-1968 occupation classification at the five-digit resolution level, of which 859 could be linked to CANJEM using both assigned codes. Each of the two sets of codes was separately linked to CANJEM and thereby generated, for each of the 258 occupational agents available in CANJEM, two exposure estimates: exposure status (yes/no) and intensity of exposure (low, medium, and high) for exposed jobs only. Then, inter-rater reliability (IRR) was computed (i) after stratifying agents in 4 classes depending, for each, on the proportion of occupation codes in CANJEM defined as 'exposed' and (ii) for two additional scenarios restricted to jobs coded differently: the first one using experts' codes, the other one using codes randomly selected. IRR was computed using Cohen's kappa, PABAK and Gwet's AC1 index for exposure status, and weighted kappa and Gwet's AC2 for exposure intensity.

RESULTS

Across all agents and based on all jobs, median (Q1, Q3; Nagents) values were 0.68 (0.59, 0.75; 220) for kappa, 0.99 (0.95, 1.00; 258) for PABAK, and 0.99 (0.97, 1.00; 258) for AC1. For the additional scenarios, median kappa was 0.28 (0.00, 0.45; 209) and -0.01 (-0.02, 00; 233) restricted to jobs coded differently using experts' and random codes, respectively. A similar decreasing pattern was observed for PABAK and AC1 albeit with higher absolute values. Median kappa remained stable across exposure prevalence classes but was more variable for low prevalent agents. PABAK and AC1 decreased with increasing prevalence. Considering exposure intensity and all exposed jobs, median values were 0.79 (0.68, 0.91; 96) for weighted kappa, and 0.95 (0.89, 0.99; 102) for AC2. For the additional scenarios, median kappa was, respectively, 0.28 (-0.04, 0.42) and -0.05 (-0.18, 0.09) restricted to jobs coded differently using experts' and random codes, with a similar though attenuated pattern for AC2.

CONCLUSION

Despite reassuring overall reliability results, our study clearly demonstrated the loss of information associated with jobs coded differently. Especially, in cases of low exposure prevalence, efforts should be made to reliably code potentially exposed jobs.

Bayesian network-based framework for exposure-response study design and interpretation

Conventional environmental-health risk-assessment methods are often limited in their ability to account for uncertainty in contaminant exposure, chemical toxicity and resulting human health risk. Exposure levels and toxicity are both subject to significant measurement errors, and many predicted risks are well below those distinguishable from background incident rates in target populations. To address these issues methods are needed to characterize uncertainties in observations and inferences, including the ability to interpret the influence of improved measurements and larger datasets. Here we develop a Bayesian network (BN) model to quantify the joint effects of measurement errors and different sample sizes on an illustrative exposure-response system. Categorical variables are included in the network to describe measurement accuracies, actual and measured exposures, actual and measured response, and the true strength of the exposure-response relationship. Network scenarios are developed by fixing combinations of the exposure-response strength of relationship (none, medium or strong) and the accuracy of exposure and response measurements (low, high, perfect). Multiple cases are simulated for each scenario, corresponding to a synthetic exposure response study sampled from the known scenario population. A learn-from-cases algorithm is then used to assimilate the synthetic observations into an uninformed prior network, yielding updated probabilities for the strength of relationship. Ten replicate studies are simulated for each scenario and sample size, and results are presented for individual trials and their mean prediction. The model as parameterized yields little-to-no convergence when low accuracy measurements are used, though progressively faster convergence when employing high accuracy or perfect measurements. The inferences from the model are particularly efficient when the true strength of relationship is none or strong with smaller sample sizes. The tool developed in this study can help in the screening and design of exposure-response studies to better anticipate where such outcomes can occur under different levels of measurement error. It may also serve to inform methods of analysis for other network models that consider multiple streams of evidence from multiple studies of cumulative exposure and effects.