Skip to main content

Table 1 Summary statistics for county-level and point-level estimates of lung and cervix cancer mortality.

From: Geostatistical analysis of disease data: accounting for spatial support and population density in the isopleth mapping of cancer mortality risk using area-to-point Poisson kriging

Estimators

Lung cancer

Cervix cancer

County-level estimates

Mean

Variance

Min-max

Mean

Variance

Min-Max

Observed rates

21.19

18.48

9.071–31.79

2.851

2.446

0.000–8.138

Global EBS

21.95

10.57

13.91–31.68

3.052

0.092

2.281–4.051

Local EBS

22.10

10.79

13.36–31.71

2.838

0.224

2.041–4.014

ATA Poisson kriging

21.58

10.96

13.22–31.56

2.851

0.264

1.921–4.081

Point estimates

Kriged rates

21.15

4.970

15.46–26.85

2.972

1.003

0.470–6.702

Kriged global EBS

21.90

3.703

17.09–27.92

3.085

0.010

2.731–3.444

Kriged local EBS

22.02

4.520

16.25–28.24

2.883

0.154

2.061–3.519

ATP Poisson kriging

21.22

9.510

12.13–33.90

2.985

0.266

1.906–4.308

Aggregates of point estimates

Kriged rates

21.22

4.832

16.14–26.29

2.849

0.641

0.955–5.084

Kriged global EBS

21.99

3.529

17.74–27.11

3.052

0.011

2.765–3.314

Kriged local EBS

22.12

4.390

16.94–27.43

2.842

0.157

2.064–3.406

ATP Poisson kriging

21.58

10.96

13.22–31.56

2.851

0.264

1.921–4.081

  1. Mean, variance and range of cancer mortality rates per 100,000 person-years estimated at the county level or at the nodes of a 5 km grid using four alternative methods: point kriging of observed rates, point kriging of global and local empirical Bayesian smoothed (EBS) rates, area-to-area (ATA) and area-to-point (ATP) Poisson kriging. Statistics for the county-level aggregates of point estimates are listed at the bottom of the table.