Question

Suppose a random sub sample of 1,000 infants was followed for up to 10 years after their birth. 122 of these infants died within 10 years of their birth date, and the remaining infants either were lost to followup, or survived the full 10 year study period. Suppose the mean of the recorded death and censoring times was computed for this group of infants to estimate mean survival times in the first ten years following birth, and this sample mean is 6.7 years. Is this sample mean a good estimate of the true mean survival time in the 10 years following birth?

a. Yes, it is a good estimate.b. No, the sample mean will tend to overestimate the true mean survival time.c. No, the sample mean will tend to underestimate the true mean survival time.

Answer 1

Ans: Option B is correct.  (No, the sample mean will tend to overestimate the true mean survival time.)

This is beacuse here all the censored organism (who were lost to follow up) have been considered surviving. However, it is very possible that many of them may have died before the study period has completed. Here, the mean survival time calculated here, also contains those censored organisms which may have died, so, mean survival time has been overestimated.

Following assumption have been made in this study, which make the mean survival time overestimated:

• Censored individuals have the same prospect of survival as those who continue to be followed. But this is not the case always.
• Survival prospects are the same for early infants as for late recruits in the study.
• Cause of death is not considered. Factors other than competition should be considered. As there are sudden unexpected factors which should be considered in appropriate in the study of mean survival time.