Cancer case-control studies with other cancers as controls

Area of Science:

• Epidemiology
• Cancer Research

Background:

• Case-control studies are crucial for cancer research.
• Selection bias, recall bias, and interviewer bias are common challenges in selecting control groups.
• General population controls present unique difficulties, especially with deceased subjects.

Purpose of the Study:

• To review theoretical considerations for using other cancer patients as controls in cancer case-control studies.
• To evaluate the potential biases and advantages of this control selection strategy.
• To assess the future utility of this design in cancer epidemiology.

Main Methods:

• Theoretical review of control selection in case-control studies.
• Analysis of potential selection bias when using other cancers as controls.
• Comparison of other cancer controls with general population controls regarding recall and interviewer bias.

Main Results:

• Selection bias can occur if the exposure under study influences the risk of other cancers, potentially biasing the odds ratio towards unity.
• This bias is most pronounced with low-prevalence exposures and high attributable risks for other cancers.
• Other cancer controls offer advantages in reducing recall and interviewer bias compared to general population controls.
• Using other cancer patients as controls simplifies control selection for both live and deceased cases.
• Logistical and cost efficiencies are associated with employing other cancer patients as controls.

Conclusions:

• Using other cancer patients as controls is a viable strategy that can mitigate significant biases in case-control studies.
• Potential selection bias can be identified and managed using population data.
• The advantages regarding recall bias, interviewer bias, and logistical efficiency suggest increased adoption of this method.
• This approach holds particular promise for cancer registries and future epidemiological research.