Footwear Style and Risk of Falls in Older Adults (Koepsell et al., 2004, see link under) is described as a nested case-control study.
Case-control studies use two groups, one of which has no relation to the experimental variable (i.e. the "controls"). Nested case-control studies are a type of case-control, where only a subset of the "controls" are used for comparison.
For the Footwear Style/Falls paper, 1371 adults participated. However, only 327 participants were classed as 'qualified for study'. The size of the control group was then matched to the 'qualified' group (through selecting participants who were engaged in a similar activity to those that fell).
To put this another way, of the 1371 participants, 717 people were monitored but not included in the analysis.
Nested controls are often used when the incident in question, or exposure, is very rare.
During the analysis of nested case-control studies, one must make sure that the cases + selected control group are not treated as the original cohort - conditional regression and inverse weighting can often compensate for segments of the original cohort who were not selected for study.