|
|
|
|
|
Indications/Recommendations for use:
Diggle's method allows one to
fit an increased incidence function to a focus.
|
|
Description:
Diggle's method is appropriate in the presence of focus of disease:
that is, we use it if we wish to ascertain whether disease risk is higher
about a point in space. We fit a model using maximum likelihood estimation;
the significance of the model parameters indicates potential increased risk.
(More).
|
|
Test statistic: Not relevant. Diggle's method involves fitting a function; the relevant tests are for whether parameter values in the model are zero.
|
Stats: p-values and standard errors are calculated for the parameter
values in the Local Risk function.
|
|
Null Hypothesis:
That the parameter values are zero in the Local Risk function.
|
Ho:
There is no increased incidence.
|
|
Alternative Hypothesis:
Incidence is increased in the neighborhood of the focus. This raised
incidence is modelled by the function at right.
|
Ha:
|
|
GeoMed Inputs:
Diggle's method requires a data file of cases
per region and controls per region; a purported focus; and initial estimates
of the parameter values for the model.
|
|
GeoMed Outputs:
parameters of the risk model, log-likelihoods,
standard errors, and chi-square statistics; a plot of the fitted model.
|
|
Example Analysis
|
Reference:
Diggle, P.J. and B.S. Rowlingson. A Conditional
Approach to Point Process Modelling of Elevated Risk. JRSSA (1994),
157, Part 3, pp. 433-40.
|
