Skip to main content
. 2017 Nov 17;7:15821. doi: 10.1038/s41598-017-16118-6

Figure 2.

Figure 2

Association of dog ownership with cardiovascular disease and all-cause mortality in the Swedish Twin Register following adjustments and exclusions in sensitivity analysis using Cox proportional hazards regression with attained age as time-scale. The “National cohort model” was adjusted for sex, marital status, type of family, area of residence, education level, population density and occupation level.