All-cause mortality in randomized trials of cancer screening

William C. Black, David A. Haggstrom, H. Gilbert Welch

Research output: Contribution to journalArticle

268 Scopus citations


Background: The most widely accepted end point in randomized cancer screening trials is disease-specific mortality. The validity of this end point, however, rests on the assumption that cause of death can be determined accurately. An alternative end point is all-cause mortality, which depends only on the accurate ascertainment of deaths and when they occur. We compared disease-specific and all-cause mortality in published randomized cancer-screening trials to indirectly assess the validity of the disease-specific mortality end point. Methods: We examined all 12 published randomized trials of cancer screening for which both end points were available (seven of mammography, three of fecal occult blood detection, and two of chest x-ray screening for lung cancer). For each randomized trial, we subtracted disease-specific mortality observed in the screened group from that observed in the control group and all-cause mortality in the screened group from that in the control group. We then compared the differences in these two mortality measures. Results: In five of the 12 trials, differences in the two mortality rates went in opposite directions, suggesting opposite effects of screening. In four of these five trials, disease-specific mortality was lower in the screened group than in the control group, whereas all-cause mortality was the same or higher. In two of the remaining seven trials, the mortality rate differences were in the same direction but their magnitudes were inconsistent; i.e., the difference in all-cause mortality exceeded the disease-specific mortality in the control group. Thus, results of seven of the 12 trials were inconsistent in their direction or magnitude. Conclusion: Major inconsistencies were identified in disease-specific and all-cause mortality end points in randomized cancer screening trials. Because all-cause mortality is not affected by bias in classifying the cause of death, it should be examined when interpreting the results of randomized cancer-screening trials.

Original languageEnglish (US)
Pages (from-to)167-173
Number of pages7
JournalJournal of the National Cancer Institute
Issue number3
StatePublished - Feb 6 2002
Externally publishedYes

ASJC Scopus subject areas

  • Oncology
  • Cancer Research

Fingerprint Dive into the research topics of 'All-cause mortality in randomized trials of cancer screening'. Together they form a unique fingerprint.

Cite this