Session 7

Question 1

Censored Observations

Answer 1

When we don’t know the outcome for an individual by the end of the study period.

Question 2

Differential Loss to Follow-up

Answer 2

When the loss due to exposure status is different for the diseased and non-diseased individuals, or the loss due to disease status is different for the exposed and unexposed. There is an interaction between exposure and disease in causing the loss. This is akin to differential misclassification.

Question 3

Equal Loss to Follow-up

Answer 3

When the probability of loss to follow-up is the same for all exposure and disease categories.

Question 4

Loss to follow-up

Answer 4

When observation of an individual in a cohort ends before they develop the disease or before the end of the study period.

Question 5

Loss to follow-up unequal by exposure

Answer 5

When the loss to follow-up differs by exposure status but not by disease status conditional on exposure. A different proportion of exposed and unexposed are lost to follow-up, but the proportion lost to follow-up is equal for diseased and non-diseased groups within each exposure group.

Question 6

Loss to follow-up unequal by disease

Answer 6

When the loss to follow-up differs by disease status but not by exposure conditional on disease. Exposure has no effect on who is and who is not lost to follow-up but disease status does.

Question 7

Non-differential loss to follow-up

Answer 7

When there is loss to follow-up due to both exposure and disease, but independently. The loss due to exposure is the same for diseased and non-diseased individuals and the loss due to disease is the same for exposed and non-exposed individuals.