Spatial analysis of lung, colorectal, and breast cancer on Cape Cod: An application of generalized additive models to case-control data

Table 1 Summary of breast cancer models

Analysis	Latency (yrs)	Span^a	Cases/ Controls	S-Plus p-value^c	Deviance p-value^c	Kelsall/ Diggle p-value^c	Figure #
Adjusted All Residences	0	0.50	1061/ 1371	0.053	0.101	0.198	3a
Adjusted All Residences	15	0.35	528/ 650	0.004	0.010	0.016	3b
Crude All Residences	20	0.35	391/ 509	0.0008	0.003	0.003	---
Crude All Residences	20	0.15^b	391/ 509	---^d	---^d	---^d	3d
Adjusted All Residences	20	0.15	391/ 509	5.6E-6	0.001	0.006	3c,e 5a, 9
Adjusted Longest Duration	20	0.15^b	248/ 341	---^d	---^d	---^d	4a
Adjusted Longest Duration	20	0.45	248/ 341	0.008	0.020	0.029	4b

^a Optimal span obtained by using the Akaike's Information Criterion (AIC) unless otherwise noted.
^b Same span as for adjusted, all residences.
^c Null hypothesis is that the map is flat. S-Plus p-value is only approximate.
^d The p-values were omitted because the maps were created using a non-optimal span.

ISSN: 1476-069X