What interesting stats have you come across recently that have piqued your curiosity? I was recently listening to a podcast that asked listeners to consider their date of death!

It was an uncomfortable feeling, albeit an insightful podcast*.

This got me thinking, what information is available on how much longer we can expect to live? And what factors influence this, beyond expected health factors?

I’m going to focus on *my stats*. I’m a 38 year old female, living in NSW, Australia, who does not identify as Aboriginal or Torres Strait Islander. There are links throughout the article and to calculators at the bottom that you can use to calculate your own. Some even give you “the” specific date!

Checking the Australian Bureau of Statistics website:

• The key statistic reported is life expectancy at birth. This tells us the mean number of years a child born today can expect to live.
• A girl born today in Australia has a life expectancy of 85.3 years
• An extra month is gained by living in NSW (it’s 85.4 years here)
• A female in NSW who has lived as long as I have, has a life expectancy of 86 years. (This is shown under Table 1.1 column ‘ex’, expressed as expecting a 38 year old female to live another 48.086 years.)

Why is my life expectancy higher than the baby girl’s of 85.4 years? I have already survived into my late 30s (go me!), so this updated life expectancy is conditional on not dying before the age of 38 and therefore higher.

How were these Australian life expectancies calculated?

First, we need to know current mortality rates (i.e. what proportion of people die at each specific age). The most recent life tables released (4/11/2021) use the mortality data from the years 2018-2020 and the estimated population size for each age in 2019 (the midpoint of 2018-2020). These mortality rates can be seen under the column ‘qx’.

Table 1 shows us that in NSW:

• 0.287% of girls under 12 months of age die
• This drops to 0.021% for girls aged 13 to 24 months
• It’s 0.057% for 38 year old females (phew)
• While 11.2% of 90 year old females in NSW die

How can we compare groups (say by sex or region)?

We start with the same sized hypothetical group of newborn babies for each group that we want to compare. The ABS uses 100,000 (you’ll see this at the top of the ‘lx’ column). Then the group-specific mortality rates are applied to calculate the life expectancy (more details here.)

Let’s compare sub-regions across Australia in Table 2.

• I lose over half a year from the NSW life expectancy at birth because of where I live in Greater Sydney (it’s only 84.7 years at birth in my current area of residence)
• But I would have gained over 2 years if I hadn’t moved house within Sydney 6 months ago (it’s 87.6 years at birth there)

This has me ogling, why wasn’t that written in the real estate agent’s listing?!

While I don’t believe I just cut 3 years off my estimated life expectancy by moving house, I am now curious about the differences in where I live now as compared to previously. My first guess is that the increased distance to a major hospital contributes to the lower life expectancy, but there will be other factors too. I cannot determine these from the life tables released to the public.

Not that I can really complain about my numbers, sadly those who identify as Aboriginal or Torres Strait Islander have a life expectancy 10 years lower (75.9 years for a female in NSW) than those who do not! I was very disappointed this information was not listed or nested on the main ABS life tables webpage. I had to specifically search for this information and found it here.

The mean vs the median

There is a little more to see in Table 1. It’s an interesting table, albeit not the easiest to read if you didn’t study this topic at uni! The ‘lx’ column tells me that only 40% of girls born in NSW today are expected to die by the age of 85 years, their life expectancy. (That is, 62,379 of the hypothetical 100,000 baby girls survive to 85 years of age, which is just over 60%.) The reason for this is that most people tend to die at an old age in Australia (we do not see a bell shape in the chart below – it is skewed).

Moving down the ‘lx’ column of the life table, I can see that the median age at death for females in NSW is 88 years. (This is where over half the females survive to 88 years but it dips below 50,000 (i.e. less than half) surviving to 89 years.)

I like this median statistic – it’s the highest of all the ones I’ve looked at so far 😉

Some superannuation companies like the median too. Perhaps to encourage us to put more in our super accounts for retirement?! In my quest to find an online calculator to share with you, I found this one that told me my life expectancy is 88.34 years. I had to read the ‘Notes’ section to read that it was my median life expectancy. This is where data literacy is so important (and knowing to read the fine print!)

This super calculator was the most accurate that I found, although they haven’t updated it with the latest ABS data release. And this one uses your date of birth to tell you the actual date of your life expectancy … but it’s currently only working for males. Do you know a good life expectancy calculator to share?

Finally, the ABS life tables are ‘period’ life tables, which means that they use current data on mortality but do not consider future changes in mortality. Life expectancy is increasing over time in Australia, so the true life expectancy of a child born today is likely to be higher than reported.

That’s about as much as I can see in this release. Now, with this knowledge to hand, what do I want to achieve with my remaining time?!

*For those interested in the podcast, it was This Working Life featuring Michael Bungay Stanier discussing goal setting on Jan 13, 2022.