Informative censoring, also known as non-random censoring, occurs when the censoring of data is not random and is influenced by the characteristics of the study population. This type of censoring can introduce bias into the results and can lead to incorrect conclusions.

One example of informative censoring is when a study is conducted on the effectiveness of a new cancer treatment. The study follows patients for a specific amount of time, but some patients may not complete the study due to the severity of their cancer. In this case, the censoring is not random, as patients with more severe cancer are more likely to be censored from the study. This can lead to an overestimation of the effectiveness of the treatment, as the censored patients may have had worse outcomes.

Another example of informative censoring is when a study is conducted on the relationship between smoking and heart disease. The study follows a group of individuals for a specific amount of time, but some individuals may not complete the study due to death from heart disease. In this case, the censoring is not random, as smokers are more likely to be censored from the study due to their increased risk of heart disease. This can lead to an underestimation of the relationship between smoking and heart disease, as the censored individuals may have had worse outcomes.

In both of these examples, the censoring is not random and is influenced by the characteristics of the study population. This can lead to bias in the results and can cause incorrect conclusions to be drawn. To avoid this type of bias, researchers can use methods such as weighted regression analysis or sensitivity analysis to account for the non-random censoring.

In conclusion, informative censoring occurs when the censoring of data is not random and is influenced by the characteristics of the study population. This type of censoring can introduce bias into the results and can lead to incorrect conclusions. It is important for researchers to account for informative censoring in their analyses to avoid these biases.