Epiville

Glossary

Matching

In this study, it is necessary to perform a statistical analysis appropriate for a matched case-control study since the authors matched on age (in decades), sex, hospital, and hospital- room status (private, semiprivate, or ward). Since cases and controls are matched on these factors, you can no longer elucidate the effects of these variables. Cases can be matched individually or frequency matched on particular variable. In the first scenario, one or more controls are selected to match a particular case on a set of variables, while in the second the controls are selected in such a way that their distribution on a set of variables resembles that of the cases.