Approximately 77,000 participants were randomized into the intervention arm. Each was expected to comply with up to four (T0-T3) chest x-rays except for never smokers randomized after 12/7/95 who did not receive a T3 screen. Overall, about 236,000 chest x-rays were conducted among approximately 71,000 participants who complied with at least one screen.
Chest x-ray results are contained in the data as a summarized/qualitative result (i.e., negative, abnormal suspicious, abnormal non-suspicious, inadequate). The description and location of abnormalities were also captured.
In the data, a XRY result in a study year is almost always given as the result of one reading from the radiologist. However, there are some exceptions. For one, about 8,300 XRYs had an accompanying quality assurance (QA) reading by another radiologist. For these screens, if the QA result was more severe than that of the scheduled protocol screen (severity: abnormal suspicious > abnormal non-suspicious > negative > inadequate) then the XRY result for that year is that of the QA reading. This occurred in about 1,000 of the 8,300 XRYs with QA. A much more infrequent exception regards a handful of inadequate XRYs that were reattempted at a later visit. If the second attempt yielded an adequate result, then the XRY result for that year is that adequate result.
For most purposes, screening data can be obtained within the Lung dataset. Additional details are contained within the Lung Screening dataset (one record per screening study year per person) and the Lung Screening Abnormalities dataset (one record per chest x-ray abnormality per person).
Beginning in 2006 an effort was undertaken to create a digital copy of all PLCO screening chest x-ray films. Of the approximately 237,000 chest x-rays screened in the trial, about 189,000 (80%) are digitized. Nearly 71,000 participants (81%) who completed at least one screen have a digitized film. Images are stored in Tagged Image File Format (.tiff) with a low-contrast technique. Digitized X-rays can be obtained through a request.