ycbm
Einstein would be proud of my Insanity...
Aaaargh! After all that I forgot the unforgettable traumatic one!
6 dropped dead on a hack under me, ruptured aorta.
6 dropped dead on a hack under me, ruptured aorta.
OK, figures are up to date as of eggs post #185.
Sorry if these come out huge for everyone!
Edit: please ignore the god awful colours, I forgot to change the colour scheme.
Histogram of age at death
View attachment 65266
It looks like the most common age for a horse to pass away is 18, however there are definite peaks at 3, 6-8, 12/13, and then later in the 20s.
Histogram cause of death
View attachment 65267
The most common COD is lameness -- this includes horses PTS due to arthritis, navicular, tendons, etc. When I have more time I'll go back through and break this down properly. 'Uk' means 'unknown'. I did try to group together similar reasons which is why these are so broad. So, for example, if someone said 'laminitis caused by cushings' I was putting that down as the COD is cushings, as without that health problem the horse wouldn't have had lami etc.
Relationship between age at death and cause of death
View attachment 65268
Horses that die of old age are likely to be older! Surprise . If anyone is unsure how to read this just google 'box plots'. We can see that grass sickness and worm damage are more likely to take younger horses, as are riding accidents. 'Sudden' deaths (i.e. horses that drop dead with no warning) happen across age groups, as do deaths due to neurological causes. Horses are more likely to be PTS for behavioural reasons around 7yo, and the median age for fatal colic to strike is ~18yo.
View attachment 65269
This is just a nicer way of reading the box plot above. We can see a definite pattern to deaths due to lameness (turn it sideways if that helps), with a gap at around 16/17yo. Distributions of colic deaths are roughly even over time.
I'm hoping I'll find time to do an interactive notebook of these results, plus some actual stats, and will maybe add a front end for wider data collection (so it can be posted to Facebook, for example) that is easier than doing it by hand.
Edit 2: a lot of these (n=506) are missing breed and height info so I’ve not included that yet.
OK, figures are up to date as of eggs post #185.
Sorry if these come out huge for everyone!
Edit: please ignore the god awful colours, I forgot to change the colour scheme.
Histogram of age at death
View attachment 65266
It looks like the most common age for a horse to pass away is 18, however there are definite peaks at 3, 6-8, 12/13, and then later in the 20s.
Histogram cause of death
View attachment 65267
The most common COD is lameness -- this includes horses PTS due to arthritis, navicular, tendons, etc. When I have more time I'll go back through and break this down properly. 'Uk' means 'unknown'. I did try to group together similar reasons which is why these are so broad. So, for example, if someone said 'laminitis caused by cushings' I was putting that down as the COD is cushings, as without that health problem the horse wouldn't have had lami etc.
Relationship between age at death and cause of death
View attachment 65268
Horses that die of old age are likely to be older! Surprise . If anyone is unsure how to read this just google 'box plots'. We can see that grass sickness and worm damage are more likely to take younger horses, as are riding accidents. 'Sudden' deaths (i.e. horses that drop dead with no warning) happen across age groups, as do deaths due to neurological causes. Horses are more likely to be PTS for behavioural reasons around 7yo, and the median age for fatal colic to strike is ~18yo.
View attachment 65269
This is just a nicer way of reading the box plot above. We can see a definite pattern to deaths due to lameness (turn it sideways if that helps), with a gap at around 16/17yo. Distributions of colic deaths are roughly even over time.
I'm hoping I'll find time to do an interactive notebook of these results, plus some actual stats, and will maybe add a front end for wider data collection (so it can be posted to Facebook, for example) that is easier than doing it by hand.
Edit 2: a lot of these (n=506) are missing breed and height info so I’ve not included that yet.
DirectorFury - amazeballs!!!!!!
I am wondering whether the data we have are skewing the result a bit to show an earlier average age of death. As well as those deceased, we all know lots that are alive so they don't figure in the calculations. Some will be a good old age, and if we are basing this on those that have died within our experience, won't this slant it to include more young horses? Or am I just wrong??
We've had the full range. This includes my horses and ones I have looked after in my field on behalf of their owners.
Mine:
27 ulcerated eye
18 ringbone
16 kissing spine
12 liver failure
But we've had some real oldies in the field!
42 died in field
40 losing condition/deteriorating/stiff
Currently got a couple of very late 20/early 30s pottering about. These are the horses that make me feel cheated for losing mine earlier. I currently have a 16, an 11, and a 3yo, so I'd like them to live forever, please!
I'm strictly Python and/or D3 I'm afraid! I normally design the visualisations that PowerBI/Tableau then include in their options so am one step removed from what most people use it for .OMG - DirectorFury, I think I want to hug you. Love a box and whisker sensitivity graph.
Actually love all data visualisations. Thanks for this. Not a PowerBI freak by any chance as well are you?
Yep, if we'd recorded data on horses that were still alive I could've run a survival analysis which would have been interesting, though I think that's doable with the data we have anyway. You've also got the problem of self-selection -- people who reply to this thread might have known less horses that died (time it takes to type a long post), might be more willing to contribute, etc.DirectorFury - amazeballs!!!!!!
I am wondering whether the data we have are skewing the result a bit to show an earlier average age of death. As well as those deceased, we all know lots that are alive so they don't figure in the calculations. Some will be a good old age, and if we are basing this on those that have died within our experience, won't this slant it to include more young horses? Or am I just wrong??
Yes, sadly it does confirm what I;ve heard said anecdotally, also about certain ages at which there seems to be increased mortality, but if they get past these ages they are then 'safer' for the next couple of years. The matter of breeds may also be quite telling, so good that this is included. I remember someone saying that these stages seemed especially predominant in thoroughbreds.It reflects what I've seen over 50 years of being around horses, that the vast majority never get anywhere near their mid twenties.
I started the thread because I thought social media was making a lot of people feel very guilty that their horses didn't make it to old age. I really hope this thread helps with that.
We humans bred horses up in size from what was natural to make them more useful to us, and in doing so we made them very fragile
What size were the 40 year olds?
We use essential cookies to make this site work, and optional cookies to enhance your experience.