The authors of a study state that careful autopsies show that 60 percent of all persons who die have evidence of recent or previous pulmonary embolism and conclude that pulmonary embolism is the leading cause of death in India possible reasons for disagreeing with the conclusion include all the following except:
a. The exact proportion of deaths due to pulmonary embolism is not known and could only be determined by a study of a random sample of all persons who died. This would avoid the problem of selection bias (only certain persons undergo autopsies).
b. The study should be large enough to avoid random error (perhaps the cited study found emboli in three of five autopsies), and the investigators should have careful and precise definitions of pulmonary embolism to avoid over- diagnosis of the condition. The association between pulmonary embolism and death may be due to confounding if some other factor causes both.
c. Thus, pulmonary embolism, though present, may not be the cause of death. Lead-time bias refers to an apparent increase in survival among persons whose disease is detected by screening. For example, 5-year survival of cancer patients identified on screening might appear to be prolonged simply as a result of starting to count the survival time earlier in the course of disease.