All-cause mortality in randomized trials of cancer screening

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.

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.

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.

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.