Skip to main content

Table 3 Summary of kidney cancer models showing degree of smoothing and global test statistic

From: Spatial analysis of bladder, kidney, and pancreatic cancer on upper Cape Cod: an application of generalized additive models to case-control data

Analysis Latency (yrs) Spana Cases/Controlsd Deviance p-valuee Figure #
Crude
All Residences
0 0.95 54/1220 <0.001 --
Crude
All Residences
0 0.90b 54/1220 -----f 4a
Adjusted
All Residences
0 0.90 54/1220 <0.001 4b
Adjusted
All Residences
0 0.40c 54/1220 -----f 4c
Adjusted
All Residences
0 0.15c 54/1220 -----f 4d
  1. Span is the percentage of data used for smoothing the model.
  2. a Optimal span obtained by using the Akaike's Information Criterion (AIC) unless otherwise noted.
  3. b Same span as for adjusted, all residences.
  4. c Local minimum span was used.
  5. d Number of residences contributed to the analysis by participant status.
  6. e Null hypothesis is that the map is flat.
  7. f The p-values were omitted because the maps were created using a non-optimal span.