Any remaining stratification - after careful design of a case-control study - can be investigated and to some extent controlled for by analytical methods 28;29. When a covariate has a marked influence on disease risk but is unlikely to be associated with allele frequency differences (e.g. age), matching may still improve the power of a study by ensuring that controls had the same opportunity as cases to develop - and be diagnosed with - the disease. For example, when controls are selected that are substantially younger than cases, they may include individuals who would have developed the disease of interest given time and thus reduce power.
