Invited Commentary: Cross-Sectional Studies and Causal Inference-It's Complicated.

Area of Science:

• Epidemiology
• Causal Inference

Background:

• Cross-sectional studies are frequently used in etiological research.
• Their utility for causal inference, particularly in addressing etiological questions, has been debated.
• Savitz and Wellenius (Am J Epidemiol. 2023;192(4):514-516) recently discussed this contribution.

Purpose of the Study:

• To elaborate on the conditions required for cross-sectional studies to effectively contribute to causal inference in etiological research.
• To apply a modern causal inference lens to the use of cross-sectional data for etiological questions.

Main Methods:

• Conceptual discussion and elaboration on existing literature.
• Application of modern causal inference frameworks.
• Analysis of the strengths and limitations of cross-sectional designs for etiological research.

Main Results:

• Cross-sectional studies can inform causal inference for etiological questions under specific, well-defined conditions.
• Identifying these conditions is crucial for maximizing the value of cross-sectional data.
• A modern causal inference perspective highlights potential biases and necessary assumptions.

Conclusions:

• Cross-sectional studies, when thoughtfully designed and analyzed within a causal inference framework, can provide valuable insights into etiology.
• Researchers must carefully consider the assumptions and limitations inherent in the cross-sectional design when addressing etiological questions.
• Further methodological development is needed to enhance the causal inference capabilities of cross-sectional studies.