Turnbull's Test
|
|
|
|
Indications/Recommendations for use: Scan for clusters of cases arising from a population size R. This method is most powerful when the population size at elevated risk is known a priori.
|
|
Description:
A circular window is centered on each region in turn and expanded to include nearest neighboring regions until the total aggregated population equals R.
|
|
Test statistic: The maximum number of cases observed among all areas with fixed population sizes.
|
Stat:
|
|
Null Hypothesis:
The number of cases in the constant-population areas follow a Poisson distribution with a common rate, but they are not statistically independent as the areas overlap
|
Ho:
|
|
Alternative Hypothesis:
Not the Null Hypothesis
|
Ha:
|
|
GeoMed Inputs:
Region-level data with a centroid coordinate, case counts, and population at risk per region; Population radius R.
|
|
GeoMed Outputs:
Tabled information includes test statistic and p-value for the 3 most significant clusters; mapped information provides cluster locations.
|
|
Example Analysis
|
Reference:
Turnbull, B.W., et. al., 1990. Monitoring for clusters of disease: Application to leukemia incidence in upstate New York, Am J Epidemiology, 132:S136-S143.
|