Prisoner Longevity News Item: Math Question

I read an article in this morning’s paper which stated that prisoners live longer than non-prisoners.

San Diego U-T Article
Similar item from USA Today

In the article, they only gave statistics on death rate. The death rate per 100,000 was less in prisons than the general population. There was no attempt to control for differences in the two populations.

My questions: is it a valid conclusion to say that a group with a lower death rate lives longer? (I suspect “NO!” but I’m not a math person)
Do prisoners, in fact, have longer lifespans than non-prisoners? And if so why — it is like indoor cats versus outdoor cats (fewer dangers, a more controlled environment and better care)?

This would be my guess. Watching the paint dry just isn’t that dangerous an activity.

In response to question 1, no. Crude death rates, like the ones reported in that article, are not very useful statistics. The full report, found here:

specifies that they do control somewhat for population composition by looking at the general population between the ages of 15 and 64. Apparently 99% of prisoners also fall into that age range. However, a quick look at a life table shows that - unsurprisingly - 64 year olds have a much higher probability of dying (0.0140) in a given year than do 15 year olds (0.0003). If the prison population is younger than the US population in general, which wouldn’t surprise me, then prisoners may have a similar or higher probability of dying than a demographically similar person in the general population.

I’m not sure about question 2. I imagine that it would actually be fairly tricky to work out. It would be hard to say who exactly would be the appropriate comparison population, since people who end up in prison likely differ from people who don’t end up in prison in some important, health-affecting ways.

Comparing the news report to the original study makes it apparent that the mis-interpretation that lower death rates mean that inmates are “living longer than people on the outside” was appended by a mathematically illiterate reporter.

This is complete rubbish.

For starter, the average age of prisoners is around 23-26, straight away you can see the problem of comparing this with the average population between 15 and 64.

Next is that the average prison terms renders it extremely unlikey that the prisoner will die, despite what you may think, the average time served in prison is less than 5 years.

In the UK, and average citizen is classed in medical terms as being elderly at around 60, and for prisoners this figure is 47, but there is debate as to wether this should be reduced to 45.

This is complete rubbish.

For starters, the average age of prisoners is around 24-28, straight away you can see the problem of comparing this with the average population between 15 and 64.

Next is that the average prison terms renders it extremely unlikey that the prisoner will die, despite what you may think, the average time served in prison is less than 5 years.

In the UK, and average citizen is classed in medical terms as being elderly at around 60, and for prisoners this figure is 47, but there is debate as to wether this should be reduced to 45.
The reason that this is significant is that patient on a GPs roll are funded on certain criteria, so that greater allowance is made to fund a GP who has a disproportionate number of elderly folk on their roll, other criteria that are considered when funding patients in our Health system include diabetes, socio economic levels in the area.

One significant trigger point for increased funding is age, and its recognised that older prisoners will be more ill more often and require more intervention, and will die earlier than the general population.

The figures I have seen suggests a death rate that is at least twice the rate for prisoners as against the general population aged between 15-44 years, and to my mind this is rather optimistic, given that the median age range for prisoners is around 25.

Even the normalised figures suggest the rate for ex-offenders is around three times as high as the control, and for those doing non-prison community, the rate is even higher.

None of these should come as a surprise, as known suicide risk factors include mental illness, drug use, unemployment, and that’s without the known risk taking behaviour of the individuals concerned, and the poor health that goes along with abuse and sefl neglect.

This report relates mainly to suicide,

http://www.homeoffice.gov.uk/rds/pdfs/hors231.pdf

Yes, that report is completely bogus. As others pointed out, the age distribution is widely different. In a normal population of people aged 15 to 64, you’re going to have a pretty even distribution - the average age is probably around 40. In a typical prison population, you might still have a range of ages 15 to 64, but the distribtuion will be heavily biased towards the young. 90% of the prisoners will probably be under 25 and the average age will be around 20. So all this report proves is that older people have a higher deathrate than younger people.

There is also the fact that many prisoners are in better physical shape than the general population. Nearly all prisons have an exercise facility, right there on the property. And ‘working out’ is a popular pastime in prison – there aren’t that many other things to do with your time!

The report does conclude that:

As has been said, even with age adjustment for direct comparison, it wouldn’t matter. The incarcerated population is just too different for any meaningful conclusions to be drawn from a death/mortality rate.

Not true at all. There are some prisoners who lift weights, but the favored recreational activites for the majority are watching television, eating junk food, and hanging out with their friends. And their poor lifestyles show - you’d be amazed at the general unhealthiness of prisoners. Qadgop could give you more specifics but the rates for various diseases and infirmities among prisoners is far higher than a comparible group on the outside.

Prisoners don’t drive cars.

The populations in the study really aren’t comparable in any meaningful way - “males aged 15 to 64” is a rather large, heterogenous group. I’d bet the prison population has a different age distribution along with many, many other different factors.

It isn’t that no meaningful conclusions can be drawn, only that the particular conclusion quoted in the news account–that prisoners live longer–can’t be drawn.

Given the age adjustment in the study that you yourself cited, we draw the interesting but opposite conclusion–that prisoners drop like flies relative to the general population of the same age. No doubt the reasons are as given by Little Nemo, to which I would add that many prisoners have long histories of alcohol and drug abuse before, and sometimes even after, their incarceration.

That’s what I mean, the prisoners in this study have long histories of all kinds of things. Some measured, some not, but regardless, most of them aren’t accounted for in a simple death rate. I’ll concede that the report supplies some basic information, but beyond that I don’t think it’s conclusive.

And during, too – smuggling drugs into our prisons, or even prisoners setting up their own stills, is fairly common. Guards spend a lot of time watching for such things, but they happen anyway.