What is the purpose of inverse probability weighting in cohort studies?

• A. To reweight observed data to represent the original population given the probability of remaining in the study.
• B. To increase sample size.
• C. To impute missing outcomes.
• D. To test for interaction.

Answer: (A)

Inverse probability weighting is a way to adjust for bias from participants who are lost to follow-up or have missing data by reweighting the observed set so it better represents the original target population. It assigns each observed individual a weight equal to the inverse of the probability that they would remain in the study (given their covariates). Those who are less likely to stay in the study get larger weights, compensating for their underrepresentation and aiming to restore the distribution of covariates seen in the full cohort. This helps produce unbiased estimates under the missing-at-random assumption, assuming the model for the probability of remaining is correctly specified and there is sufficient variability in that probability across individuals. It does not increase the actual sample size, does not impute missing outcomes, and does not directly test interactions.