Lead time bias
Lead time bias is a bias that may be encountered in radiology literature on imaging detection of disease.
Lead time is the time between detection of a disease with imaging and its usual clinical presentation. An imaging technique or modality may claim to lengthen survival time by earlier detection of the disease, but if there is no actual survival time difference between patients who have the disease detected clinically and those who have it detected on screening, this represents a bias.
- patient 1: disease detected clinically
- survival time = (death - clinical detection time)
- patient 2: disease detected on imaging
- survival time = (death - imaging detection time)
The lead time is the (clinical detection time - imaging detection time). If both patient 1 and patient 2 actually die at the same time, and earlier detection and treatment does not result in actual lengthened survival time, then the perceived longer survival time is a misrepresentation and a bias.
This bias is important to account for in screening studies because if research data shows a lengthened survival time with an imaging study, but this is due to lead time bias, then the patient may be subjected to expensive and invasive procedures that are unnecessary.
Lead time biases are concerns in many areas, including:
- breast cancer imaging/screening
- prostate cancer imaging/screening
- renal cancer imaging
- aortic aneurysm imaging
- clinical trials
- descriptive studies
- sensitivity and specificity
- positive predictive value (PPV)
- negative predictive value (NPV)
- likelihood ratio (LR)
- normal distribution
- type I error
- type II error
- confidence interval
- ROC curve
- retrospective studies
- prospective studies
- analyses of variance
- nonparametric statistics