Question

One cross-sectional study found that shingles was more common in lower socioeconomic status (SES) groups compared to higher SES groups. A later cohort study which analyzed the incident rate of shingles in both high and low SES groups found no difference among high and low SES groups. The difference in findings could most likely be explained by:

1. Information bias

2. the incidence-prevalence bias

3. the reverse-causality bias

4. 1) Information bias and 3) the reverse-causality bias

Answer 1

Problem statement: Two statistical studies producing conflicting inferences are presented. We have to identify if there are any issue with the research methodology resulting in these findings.

Given: We are provided with the information on the type of studies conducted. One is cohort study while the other is a cross sectional study.

Solution: Answer is option (2).

Explanation: The incidence-prevalence bias speaks about issues of selection bias. In this bias, especially in epidemiology, the cases of mild or severe disease condition are excluded from the study resulting in a biased study.

In the problem shared, the first study is a cross-sectional study implying the data for the study was collected at a particular instance in time. Since data collection is a one time activity for cross-sectional data based research, this study has a high chance of excluding cases of patients of shingles for whom the disease was still in its too early stage or who might have died because of it.Leaving out these extreme cases (early phases of shingles or death caused due to shingles which can't be included) results in selection bias during data collection. While in the case of cohort study it is longitudinal. Data is observed at different instances of time. Also as mentioned in the question ,the study starts measuring or recording cases of shingles right from its incidence in the patients across different groups and follows the disease at different instances of time resulting in reduced selection bias. Here in the cohort study, a group of people exhibiting certain trait/characterstic are studied at different intervals of time. For this case specific, different group of individuals might have been studied at different intervals of time, right from the incidence of shingles resulting in reduced selection bias.