Rates (cont.)

Person-years of exposure to risk

When calculating rates we need to think about who is at risk and how long they are at risk for. The time spent during the specified period when the event of interest could happen to them is called person-years at risk. This can be calculated as any unit of time - days, weeks, months etc. - but in demography we most commonly refer to person-years at risk.

The chart below demonstrates how the number of person-years at risk lived by a population in a specified time period are determined.

Interaction Click on the arrows to watch the animation

Person-years of exposure to risk
 dd  dd

 

interaction Based on this information calculate the total person-years at risk for this population.

The correct answer is 4.25. This is calculated as: 1+0.75+0.75+0.50+1+0.25=4.25.

Please attempt the answer.

Well done, 4.25 is correct!

Sorry, that is not correct. Please try again.

This is calculated by summing all the individual person-years at risk.

 

Interaction Now calculate the death rate in this population. Submit your answer (per person-year) rounded to 4 places of decimals.

The correct answer is 0.4706 per person-year. This is calculated as:
2
4.25
It is essential to include the units of measurement.

Please attempt the answer.

Well done, 0.4706 per person-year is correct!

Sorry, that is not correct. Please try again.

Remember that convention gives death rates per 1000, and you must include the units!

 

Although in this example we used death as the outcome of interest, the process of calculation is similar for other events which may be of interest. Calculation is simplest for irreversible events such as death, or HIV infection, as once the event has happened to the individual they will never be at risk of the event occurring again. However demographers often study events which may occur more than once, such as pregnancy or marriage. In these cases we need to be aware that a person’s ‘at risk’ status is more fluid. For example, a single person is ‘at risk’ of marriage. If they then get married, they are no longer at risk of the event (in monogamous populations at least). However, they may later get divorced, becoming ‘at risk’ of marriage again and contributing more person-years to the denominator.