A SAS Macro to Analyze Data From a Matched or Finely Stratified Case-Control Design

A matched case-control design is a common approach used to assess diseaseexposure relationships, and is often a more efficient method than an unmatched design. However, for the valid analysis of such an approach, a modeling technique that incorporates the matched nature of the data is needed. This prohibits the use of a standard unconditional logistic regression analysis generally available in PROC LOGISTIC. A stratified conditional logistic model has the same flexibility as an unconditional model, yet can still take into account the correlation structure attributable to matching. This paper presents a SAS macro that fits a conditional logistic regression model to matched or finely stratified data using the PHREG procedure. The macro enhances standard PHREG output by producing summary tables and statistics used to describe the matched sets. It also calculates several regression diagnostics, some not available in PHREG, that can be used to assess model fit. This paper is intended for an audience with a working knowledge of statistical modeling.