Biological heterogeneity and length-biased sampling in asymptomatic neurosurgical patients

Increased numbers of asymptomatic intracranial lesions are being identified because of recent advances in imaging technology. Understanding of the natural history of these diseases, together with length bias is highly important in refining treatment strategy. Two sample models of hypothetical healthy cohorts were constructed, in which diseases showed either dual-type or multiple-type heterogeneity. Relative preclinical interval (PCI) of asymptomatic lesions including length-biased sampling was calculated, confirming that relative PCI increased according to heterogeneity of the disease. Length-biased sampling in asymptomatic lesions results from biological heterogeneity of the disease. Because lesions of longer PCI always are over-represented in a healthy population, conventional frequency analysis will tend to over-estimate risks in these patients.