Using SAS to Perform Individual Matching in Design of Case-Control Studies

Free registration required

Executive Summary

In the design of case-control studies it is often desired to match each case with one or more controls based on a set of variables and maximum differences between case and control in these variables. Case-control studies can arise within cohort studies where it is desired to compare subjects experiencing an event (cases) to subjects not experiencing the event (controls) for some risk factor not yet measured, but available upon further data extraction (e.g., blood determination from stored specimens, extraction of information from medical records, etc.). Typically, all cases are measured, but because of feasibility and cost only a subset of controls are measured, which are often matched to individual cases.

  • Format: PDF
  • Size: 44.7 KB