Researchers are interested in the association between colorectal carcinoma and nonsteroidal anti-inflammatory drug use. They first interview a group of patients with biopsy-proven colorectal carcinoma and then interview a group consisting of the patients' neighbors who are of similar age and race. The analysis is based on comparing the results of pairs of individuals (one from each of the 2 groups) who have similar characteristics. This design technique best helps address which of the following potential problems with this study?

Question

Researchers are interested in the association between colorectal carcinoma and nonsteroidal anti-inflammatory drug use.  They first interview a group of patients with biopsy-proven colorectal carcinoma and then interview a group consisting of the patients' neighbors who are of similar age and race.  The analysis is based on comparing the results of pairs of individuals (one from each of the 2 groups) who have similar characteristics.  This design technique best helps address which of the following potential problems with this study?

Quizplus · Accepted Answer

Matching is a method generally used in the design stage of case-control studies to control confounding (ie, when a perceived association between an exposure and an outcome is actually explained by a confounding variable associated with both the exposure and the outcome). The initial step in matching involves selecting variables that could be confounders (eg, age, race). Cases and controls are then selected based on the matching variables so that both groups have a similar distribution in accordance with the variables. In this scenario, the "cases" (patients with colorectal cancer) were matched with neighborhood "controls" of similar age and race. Selecting neighbors as controls has another advantage of matching the cases to controls by variables that are difficult to measure (eg, socioeconomic status, environmental factors). Gender and smoking status are other common confounders. (Choices A, C, D, and E) Observer bias and ascertainment bias result from mislabeling exposed/unexposed or cases/controls. Recall bias could be a limitation of this study as the interviewed participants with colorectal cancer may be more likely to recall certain exposures. Selection bias is a potential problem in this study because the controls selected may not reflect the exposure experience of the general population. However, although these biases may be present, matching best addresses confounding rather than any of these biases. Educational objective: Matching is used in case-control studies in order to control confounding. Matching variables should always be the potential confounders of the study (eg, age, race). Cases and controls are then selected based on the matching variables so that both groups have a similar distribution in accordance with the variables. __________ References: Matching. (http://www.ncbi.nlm.nih.gov/pubmed/7987107)

Researchers Are Interested in the Association Between Colorectal Carcinoma and Nonsteroidal