Spatial clustering of physical activity and obesity in relation to built environment factors among older women in three U.S. states

Table 1 Characteristics of spatial clusters of physical activity in California and Massachusetts and obesity in Pennsylvania

	Area: Counties	Radius (km)	Participants	Cases^a	Relative risk	P-value
Physical activity clusters in California
Cluster 1	Coastal area: San Luis Obispo, Santa Barbara	96.74	232	88	1.51	0.0024
Cluster 2	Bay Area: San Francisco, Santa Clara, Santa Cruz, Alameda, San Mateo, Marin, Contra Costa	73.19	1837	527	1.17	0.035
Cluster 3	South inland: Tulare, Kern Kings	121.09	129	14	0.42	0.0027
Cluster 4	North inland: Lassen, Shasta, Tehama, Plumas, Butte, Glenn, Sierra, Yuba, Nevada, Placer, Sutter, El Dorado	139.21	385	71	0.71	0.047
Physical activity clusters in Massachusetts
Cluster 5	Cape Cod: Barnstable, Dukes, Nantucket	50.67	427	138	1.39	0.0003
Cluster 6	Boston: Suffolk	0^b	122	43	1.48	0.053
Cluster 7	Central/Western Massachusetts: Berkshire, Franklin, Hampshire, Hampden Worcester	117.08	1432	306	0.86	0.06
Obesity clusters in Pennsylvania
Cluster 8	Western Pennsylvania: Allegheny, Armstrong, Beaver, Butler, Cambria, Clarion, Forest, Indiana, Jefferson, Lawrence, Venango, Washington, Westmoreland	82.93	2424	657	1.17	0.029
Cluster 9	Near Philadelphia: Montgomery, Chester, Delaware	36.54	1335	268	0.8	0.01

^aCases are defined as participants meeting physical activity recommendations and as obese participants.
^bSince Suffolk County was the only county identified as cluster 5, the radius was 0.

ISSN: 1471-2458