Hi,

you can insert the variable "Age" into the model as a covariate, and since it is a continuous measure you will get a single Hazard Ratio indicating the increase (or decrease) in risk of event for every unit of the variable (i.e. for every additional year of age, if you measure age in years).

Since you have a continuous measurement, it may be advisable to insert it as it is in the model, in order not to lose data or statistical power. However, if you prefer to calculate different Hazard Ratios for various age categories, you can create a categorical variable with the various age groups and insert it in the model as a covariate. In this case, remember to specify the categorical nature of that variable in the "Categorical ..." submenu, where you can also specify whether to use the first or last category as a reference for the calculation of the Hazard Ratios of the other groups.

Read this guide, http://www.sthda.com/english/wiki/cox-proportional-hazards-model

It depends on your model.

If you think about time, age is the time up to the entry into the model and also is the 'survival' time! So you can think of age as a single time variable. The person becomes 'at risk' at a particular age (perhaps birth, perhaps later), the enter the calculation at age of enrolment and they exit at age of exit.

This approach allows you to model the hazard function for all ages from the youngest entrant to the oldest exit! See this brilliant and much-cited paper by Ed Korn:

Korn, E., Graubard, B., Midthune, D. (1997). Time-to-event analysis of longitudinal follow-up of a survey: choice of the time-scale. American Journal of Epidemiology 145(1), 72 - 80.