Class of 1996

At 22, I’m now a little over a quarter of the way through my allotted standard-issue UK male lifetime. Inspired by FlowingData‘s article Causes of Death, the suicide statistics for my age group cited in Amber Baldet’s 2013 DEF CON talk Suicide Risk Assessment and Intervention Tactics, the narrative statistics approach of Hans Rosling’s famous 2007 TED talk The best stats you’ve ever seen and Charles Joseph Minard’s 1869 Carte figurative des pertes successives en hommes de l’Armée Française dans la campagne de Russie 1812–1813, I decided to look into how many of my birth cohort were still chugging along with me.

Click on any graph to view it full-screen.

Methodology

All figures are for England and Wales and come from ONS statistics. The numbers of births and stillbirths is from the 1996 instalment of Births in England and Wales. Death figures and causes for 1996–2005 are from Mortality statistics: Cause (series DH2), 2006–2014 from Deaths registered in England and Wales (series DR). The 2015–2017 instalments of series DR lacked the Deaths: underlying cause, sex and age-group table, and so for these years the Death registrations summary tables – England and Wales datasets were used instead.

After one year of age, ONS deaths statistics are grouped into age groups (e.g., 1–4 yrs, 5–9 yrs). As this use of ranges mean that each year’s figures only partially represents the mortality rate of the 1996 cohort, the average of the figures for all years in which a child born on January 1st 1996 would have fallen into a given age group were taken. Where the ONS later changed their age groupings, I have recalculated the figures in order to continue the original classification, and from 2016 individual years are shown.

A handful of ICD cause of death classifications changed with the ONS’ adoption of the Tenth Revision in 2001—e.g., ICD-9’s Diseases of the nervous system and sense organs was split into Diseases of the nervous system, Diseases of the eye and adnexa and Diseases of the ear and mastoid process in ICD-10—and disjoint classifications have been merged to allow the maximum possible level of precision through all years.

As a caveat, and as will become clear shortly, my n for a lot of this analysis is often quite small.

Finally, you can download my data here.

The Journey So Far

We’ll start prenatally. Picture the opening of Saving Private Ryan, the landing crafts packed with 653,024 foetuses angling for their one shot at personhood. It’s a roughly equal split, sex-wise—335,298 males to 317,726 females—and tensions are running high. Suddenly, the grinding of metal on gravel. Whistles are blown, rifles are clutched tighter and the landing ramps drop as the machine guns open up. 3,539 don’t even make it off of the uterine landing crafts. The remaining 649,485 emerge into the world kicking and screaming, but the danger is not yet passed. The first 28 days on the beachhead of life claims 2,645 neonate lives, and the long trek out of the shores of infancy sees the loss of 1,314 more. 645,526 see their first birthday, and all but 679 of those their fourth.

Fig. 1: Surviving population of 1996 cohort (zoomed and un-zoomed)

It’s a relatively uneventful few years now for the class of 1996, with a loss of only 333 between our fourth and ninth birthdays, and 372 between that and our fourteenth. The recklessness of teenhood takes a greater toll, and 813 never reach nineteen candles. Another trend also alters. So far, males and female have died at an average ratio of 1.265 males to every 1 female. Between fourteen and nineteen years this rate almost doubles to 2.02 to 1. The disparity increases over the next year, then drops slightly (fig. 2). After nineteen, we can get more granular with our ages: 252 call it quits before twenty, 237 before twenty-one.

That brings us to today: from the initial pool of 653,024, the vast majority—642,840—remain (fig. 1). Over 21 years the class of 1996 has lost 10,184 members, or 1.56% of its starting population, with males making up the slight majority (56.5%).

Fig. 2: Ratio of male deaths to female deaths

Not bad going so far, guys and girls, although considering we’re all over a quarter of the way through the race it’s probably safe to assume that the rate of decline will increase precipitiously in future decades.

Watch Out for…?

Going back to the DEF CON talk on suicide that I mentioned earlier, I was interested to find out just what it was that had killed 10,000 of my peers, both in terms of overall body counts (fig. 3) and proportionally (fig. 4), how this has changed over the years and how this differs (if at all) between males and females.

Fig. 3: Number of deaths by cause (all)

Three types of death start off common but drop sharply within the first four years: Perinatal issues (as we might expect of babies), infectious and parasitic diseases and causes of death not elsewhere classified. Without being an expert on infant medical care, I would intuit that the decline in Not elsewhere classifieds suggests that the ultimate cause of a baby’s death might be harder to identify than that of a more grown child or adult.

Deaths due to diseases of the respiratory system and congenital issues also decline as time goes on, all but disappearing as causes by the twenties, whilst neoplasms are the only internal cause of death that increases drastically during the first decade before plateauing for the next and then declining. As a point of interest, whilst deaths classified as due to Mental and behavioural disorders remain very low throughout, they begin far earlier than I would have expected with the first such death appearing between 1–4 yrs.

Fig. 4: Proportion of deaths by cause

Comparing the proportion of each cause of death for males and females presents a couple of interesting points. One is the sharp disparity in deaths caused by external causes, particularly in later years—21-year-old males died this way at just over 4.3 times the rate of females. For reasons I don’t feel I am medically qualified to investigate further, the relative lack of external causes of death for teenaged-and-above females is made up primarily by a resurgence in deaths due to congenital issues (which seem to continue to pose a significant threat to females longer than it does for males, with only one 21-year-old male dying for six such females) and an almost-doubling of neoplasm-related deaths.

External Causes of Death

It’s under the blanket of external causes that we discover some of the most interesting trends. Consider the sharp, almost-quintupling of such deaths that we see during the teen years. To finish, I wanted to dig into the figures a little more in order to find out whether the greatest risk is posed by oneself, by another or by chance (i.e., by accidental death), and how that changes across time and between males and females. Here we have comparisons of the overall numbers of deaths (fig. 5) and their proportional weight (fig. 6).

Fig. 5: External causes of death

The first and most obvious takeaway is that random chance—accidents—is responsible for between 70–90% of all such deaths. Watch out for slipping in the bath.

As might also be expected, the risk posed by oneself—of self-inflicted death—is nil during the early years. It begins to appear across both sexes during the teen years, at proportionally the same rate amongst males than females (though with the sharp rise in male deaths during this period, they represent far more overall deaths than females).

On the other hand, deaths due to another’s action—assault or homicide—are a far more substantial risk during those early years before becoming almost a non-issue by 10–14, although females are still marginally more at risk from others than males after this point. Whether this high early level of infanticide is due to people being more prone to murdering babies and young children, or just that they are far easier to kill, whether intentionally or by accident, I do not know.

Fig 6: External causes of death by proportion

Replies

No comments yet.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.