Length time bias

Length time bias is a form of selection bias, a statistical distortion of results which can lead to incorrect conclusions about the data. Length time bias can occur when the lengths of intervals are analysed by selecting intervals that occupy randomly chosen points in time or space. This process favors longer intervals, thus skewing the data.

Length time bias is often discussed in the context of the benefits of cancer screening, where it can lead to the perception that screening leads to better outcomes when in reality it has no effect. Fast-growing tumors generally have a shorter asymptomatic phase than slower-growing tumors. This means that there is a shorter period of time when the cancer is present in the body (and therefore might be detected by screening) but not yet large enough to cause symptoms, which would cause the patient to seek medical care and be diagnosed without screening. As a result, if the same number of slow-growing and fast-growing tumors appear in a year, the screening test will detect more slow-growers than fast-growers. If these slow growing tumors are less likely to be fatal than the fast growers are, the people whose cancer is detected by screening will do better, on average, than the people whose tumors are detected from symptoms (or at autopsy), even if there is no real benefit to catching the cancer earlier. This can give the impression that detecting cancers through screening causes cancers to be less dangerous, when the reality is that less dangerous cancers are simply more likely to be detected by screening.^[1]

Length time bias in cancer screening. Screening appears to lead to better survival, even when no effective treatment is given.

References

1. Length bias

See also

• Lead time bias