Fig 1.
Flow chart of the selection of meta-analyses.
Table 1.
Characteristics of cohort and case–control studies.
Fig 2.
Difference in treatment effect estimates between 133 case–control and 138 cohort studies.
Difference in treatment effect estimates is expressed as ratio of estimates (RE). A RE < 1 indicates that case–control studies yielded larger estimates of the intervention effect or adverse events than cohort studies.
Fig 3.
Sensitivity analysis of differences in effect estimates between case–control and cohort studies with available data on adjusted estimates.
Difference in treatment effect estimates is expressed as ratio of estimates (RE). A RE < 1 indicates that case–control studies yielded larger estimates of the intervention effect or adverse events than cohort studies.
Fig 4.
Secondary analysis of differences in effect estimates between case–control and prospective cohort studies.
Difference in treatment effect estimates is expressed as ratio of estimates (RE). A RE < 1 indicates that case–control studies yielded larger estimates of the intervention effect or adverse events than prospective cohort studies.
Fig 5.
Secondary analysis of differences in effect estimates between case–control and retrospective cohort studies.
Difference in treatment effect estimates is expressed as ratio of estimates (RE). A RE < 1 indicates that case–control studies yielded larger estimates of the intervention effect or adverse events than retrospective cohort studies.