Example1 We want to estimate the effect of age (x) on all-cause mortality (y) among a group of individuals ages 65 and older within a ten-year follow-up period. The failure event is death (0=No event, 1=Event). Age is measured in years, with values ranging from 65 to 100. The HR for age in years is 1.13, which suggests that the expected hazard is 1.13 times higher for an individual who is one year older than another individual.
Example2 In this example, we estimate the association between weight (x) and hospitalization attributable to cardiovascular disease (y) on a population of 50-year-old women, who are followed for five years. The failure event is hospitalization for cardiovascular disease (0=No event, 1=Event). Weight at age fifty is measured in kilograms, ranging from 48 to 114. We find that the HR is 1.03. This suggests that a one-kilogram increase in weight is associated with a 3% increase in the expected hazard for hospitalization.
Practical example
Dataset
StataData1.dta
Variable name
cvd
Variable label
Out-patient care due to CVD (Ages 41-50, Year 2011-2020)
Value labels
0=No 1=Yes
Variable name
gpa
Variable label
Grade point average (Age 15, Year 1985)
Value labels
N/A
sum cvd gpa if pop_cox==1
stcox gpa if pop_cox==1, noshow
When we look at the results for gpa, we see that the hazard ratio (HR) is 0.48. Thus, for each unit increase in grade point average, the hazard of out-patient care due to CVD decreases.
The association is statistically significant, as reflected in the p-value (0.000) and the 95% confidence intervals (0.42-0.56).
Summary The higher the grade point average at age 15, the lower the risk of having experienced out-patient care due to CVD in ages 41-50 (HR=0.48). The association is statistically significant (95% CI=0.42-0.56).
