Health professionals must often analyze disease cluster allegations made by concerned citizens who sense patterns within small structured environments. Examples include households, families, sibships, wards, cellblocks, classrooms, job types, age classes, and locations within a building. It is crucial for the health professional to know what tests are best in these small-environment problems. A few large-scale methods may be modified to handle small-environment problems. However, because sample sizes are small, exact methods may be better. We have extended previous matrix-occupancy methods to yield new, rapid, and exact combinatorial expressions for analysis of patterns of diseased individuals. We propose to investigate the power of both these and other previously-published methods in structured small environments. To our knowledge, no such systematic investigation has been carried out. We expect that many test methods will be optimal in a restricted range of the parameter space, but not ideal elsewhere. Our long-range goal is to use the results of this power simulation to choose methods for incorporation into EXCLUDE, a user-friendly PC program to analyze small structured environment data, and to help the professional choose the right test method.
