Prostate

Data was collected on procedures that were part of a diagnostic work-up for prostate cancer and for staging procedures following a diagnosis. Two different types of events triggered the collection of diagnostic procedure information:

1. Screens suspicious for prostate cancer: Among approximately 36,000* men screened once a year for up to 6 years, there were about 23,000 screening PSA exams or Digital Rectal Exams (DREs) suspicious for prostate cancer. These "positive" exams triggered further diagnostic follow-up, which was systematically captured, in most cases, for up to one year following the screen. Any positive screen among the PSA, DRE and, less frequently a QA exam would refer a participant to further follow-up. This information was captured regardless of the eventual diagnosis.
2. Confirmed diagnoses of prostate cancer: When a participant was diagnosed with prostate cancer within the trial period, procedures that were part of a diagnostic work-up that occurred in the year leading up to that diagnosis were captured in the data.

Therefore, if a participant had a diagnostic procedure during the trial such that it was neither one year following a positive screen nor one year prior to a prostate cancer diagnosis, that information was not captured. As such, the only diagnostic procedures captured in the control arm are those that eventually lead to a prostate cancer diagnosis.

For most purposes, diagnostic procedure data can be obtained within the Prostate dataset. Additional details can be found in the Prostate Diagnostic Procedures dataset (one record per procedure per person) and the Prostate Medical Complications dataset (one record per medical complication per person).

*38,000 men were randomized for screening, but 2,000 never completed screening.

Lung

Data was collected on procedures that were part of a diagnostic work-up for lung cancer and for staging procedures following a diagnosis. Two different types of events triggered the collection of diagnostic procedure information:

1. Screens suspicious for lung cancer: Among approximately 71,000 participants screened once a year for up to 4 years, there were about 18,000 screening chest x-rays suspicious for lung cancer. These "positive" exams triggered further diagnostic follow-up, which was systematically captured, in most cases, for up to one year following the screen. Any positive chest x-ray screen and, less frequently a QA exam would refer a participant to further follow-up. This information was captured regardless of the eventual diagnosis.
2. Confirmed diagnoses of lung cancer: When a participant is diagnosed with lung cancer within the trial period, procedures that were part of a diagnostic work-up that occurred in the year leading up to that diagnosis were captured in the data.

Therefore, if a participant had a diagnostic procedure during the trial such that it was neither one year following a positive screen nor one year prior to a lung cancer diagnosis, that information was not captured. As such, the only diagnostic procedures captured in the control arm are those that eventually lead to a lung cancer diagnosis.

For most purposes, diagnostic procedure data can be obtained within the Lung dataset. Additional details can be found in the Lung Diagnostic Procedures dataset (one record per procedure per person) and the Lung Medical Complications dataset (one record per medical complication per person).

Colorectal

Among 22,000 follow-up endoscopies, data was collected on at least 30,000 polyps and 15,000 adenomas seen on endoscopy. Polyp counts, sizes, locations and histologies are included in the data when available.

For most purposes, adenoma and polyp data can be obtained within the Colorectal dataset. Additional details can be found in the Colorectal Endoscopies dataset (one record per endoscopy per person) and the Colorectal Polyps dataset (one record per polyp per endoscopy per person).

*77,000 participants were randomized for screening, but 10,000 never completed screening.

Ovarian

Data was collected on procedures that were part of a diagnostic work-up for ovarian cancer and for staging procedures following a diagnosis. Two different types of events triggered the collection of diagnostic procedure information:

1. Screens suspicious for ovarian cancer: Among approximately 31,000 women screened once a year for up to 6 years, there were about 6,000 screening CA-125 exams or Transvaginal Ultrasounds (TVUs) suspicious for ovarian cancer. These "positive" exams triggered further diagnostic follow-up, which was systematically captured, in most cases, for up to one year following the screen. Any positive screen among the CA-125, TVU and, less frequently a QA exam would refer a participant to further follow-up. This information was captured regardless of the eventual diagnosis.
2. Confirmed diagnoses of ovarian cancer: When a participant is diagnosed with ovarian cancer within the trial period, procedures that were part of a diagnostic work-up that occurred in the year leading up to that diagnosis were captured in the data.

Therefore, if a participant had a diagnostic procedure during the trial such that it was neither one year following a positive screen nor one year prior to a ovarian cancer diagnosis, that information was not captured. As such, the only diagnostic procedures captured in the control arm are those that eventually lead to an ovarian cancer diagnosis.

For most purposes, diagnostic procedure data can be obtained within the Ovarian dataset. Additional details can be found in the Ovarian Diagnostic Procedures dataset (one record per procedure per person) and the Ovarian Medical Complications dataset (one record per medical complication per person).

Annual Study Update

The Annual Study Update (ASU) was used to ascertain cancer diagnosis. The ASU was mailed annually to participants who were alive and remained in the study. Participants were asked if they were diagnosed with cancer, the type of cancer, date of diagnosis, hospital or clinic of diagnosis and physician contact information. In addition, men were asked if they were taking Proscar or Propecia (Finasteride).

Over 1.7 million ASUs were expected for participants who were alive and remained in the study through 2009. Of those expected ASUs, 93% were completed.

Annual Study Update data and documentation are incorporated into each of the cancer datasets.

Data was collected on the following form:

• ASU - Scanned (PDF - 83 KB)
• ASU - Accessible (PDF - 19 KB)