31

I recently read this article and although majority of the items are believable, #5 is really shocking.

Is it true that there are more than 50,000 alive people in Japan already over 100?

numaroth
  • 103
  • 4
Zaenille
  • 693
  • 6
  • 9
  • 2
    The higher life expectancy includes the 100+ people, so not really a sanity check. ;) –  Sep 08 '14 at 19:03
  • 1
    The people in the 100+ range are such a tiny fraction of the elderly that it has only a trivial impact on the life expectancy. – Mark Sep 08 '14 at 19:39
  • But the ones below 100 would have a cheated bias too if that were the usual practice in that country. They are not doing it for the 100+ but for the money. –  Sep 08 '14 at 19:48
  • @Mark, can't comment on the other threads, so mentioning here that the cheating can follow a gaussian distribution too, so you won't neccessarily see it. The question is, if there are more than 50k people 100+, this seems to depend on the grade of cheating which imo cannot be infered from public sources. –  Sep 08 '14 at 20:33
  • 1
    According to [Wikipedia](https://en.wikipedia.org/wiki/Centenarian#Centenarian_populations_by_country), Japan has 43 centenarians per 100,000 people. The next highest ratio is South Korea at 29. Now, Japan has [one of the oldest populations on Earth](http://world.bymap.org/MedianAge.html), so it certainly is possible. But it's still quite an outlier. – Jon 'links in bio' Ericson Sep 09 '14 at 19:04

1 Answers1

42

According to the Japanese Ministry of Health, Labor and Welfare, yes. As of September 15, 2013, they reported that there were 54,397 persons over the age of 100.

Though they have previously had issues with their records, recording as alive persons that had actually died years before, such as the case of Sogen Kato, who was on their records as being alive for 30 years after his death.

Compro01
  • 3,195
  • 1
  • 25
  • 29
  • 1
    It is however pretty unlikely that the error on this would be 4,396 people (almost 10%) which is what would be required to answer this question with `no` :D – avalancha Sep 08 '14 at 15:41
  • 5
    @avalancha Considering there is huge financial incentive not to report a death if you don't have to, I wouldn't be surprised if the error was that large in countries that rely on self-reporting of families of the deceased. – Chuu Sep 08 '14 at 16:11
  • 2
    @avalancha Seems like it's a problem for them right now, family keeping dead people just to cash their pension. http://www.nytimes.com/2010/08/15/world/asia/15japan.html?_r=0 – the_lotus Sep 08 '14 at 18:03
  • 23
    As a sanity check on Compro01's numbers, the US has 232 centenarians per million people. Translate that to Japan's 126 million people, and you get about 29k centenarians in Japan. Factor in Japan's 84.6-year life expectancy as opposed to the US's 79.8 (with a roughly normal distribution, a slight shift in the mean makes for a huge change at the edges), and 50k people over the age of 100 is not an unreasonable number. – Mark Sep 08 '14 at 18:40
  • 7
    @Mark Your sanity check says that the number of centenarians in Japan is consistent with the stated life expectancy. But both of those figures could be the result of under-reporting of the deaths of old people. – David Richerby Sep 08 '14 at 19:12
  • 2
    @DavidRicherby, My point was to demonstrate that 50k isn't an unreasonable number. My extrapolation from US statistics was off by less than a factor of two from the official Japanese statistics, indicating that those statistics are probably fairly accurate. There are other ways of cross-checking the numbers (eg. shape of the demographic pyramid, or a disparity between deaths at 99 and at 100), but I don't have the data to perform them. – Mark Sep 08 '14 at 19:48
  • I would note that by 110 years, the number drops off sharply. So called super-centenarians are rare and number less than 1,000 worldwide. http://www.grg.org/calment.html –  Sep 08 '14 at 20:08
  • @the_lotus The article you referenced and said was "right now" is from 2010, which is the same date as the article which Compro01 referenced as described as "previously". The articles suggest there was a "nationwide search", with officials visiting every registered centenarian. Do you have any evidence which suggest that this problem is still happening, especially a statistically significant number (e.g. 10% or 5000)? – ChrisW Sep 08 '14 at 20:57
  • @avalancha that heuristic is incorrect. If it were reported that there were 500 people over 120 years old, I would expect the error to be far more than 10%. – djechlin Sep 08 '14 at 21:56
  • @ChrisW You are right, I did not write my phrase properly. – the_lotus Sep 09 '14 at 12:39
  • @DavidRicherby: Underreporting the deaths of enough old people to push the count under 50,000 would entail ~5,000 mummified corpses stuffed in closets. This seems unlikely, particularly since the vast majority of centenarians will be living in assisted care of some kind, and since cases of this happening are apparently rare enough to make the news. – lambshaanxy Sep 09 '14 at 13:01
  • @jpatokal Don't assume that all cultures are like your own. There are plenty of cultures where the idea of putting your elderly relatives in some kind of home is abhorrent. I've no idea if Japan is such a culture. – David Richerby Sep 09 '14 at 13:06
  • @DavidRicherby: My wife is Japanese and I've lived there for years, thank you very much. As of 2010, only 18% of Japanese 65yo and over lived with their children: http://www.businessweek.com/articles/2013-02-28/in-japan-the-rising-cost-of-elder-care-and-dying-alone That proportion will certainly be less for centenarians, who are highly likely to have severe health problems and whose own children will likely be pushing 80. – lambshaanxy Sep 09 '14 at 13:14
  • Surely on any such issue, relevant questions are: Who collected the statistics? What is the source of the raw data? How accurate is it? Are there reasons why the data might be biased one way or the other? To take a deliberately silly example, if you said that we know that very few people lie on their income taxes, based on a survey in which we asked people whether they lie on their income taxes, I wouldn't take that too seriously: maybe lots of people didn't want to admit they had committed a crime to a stranger. In this case, if the Ministry of Health is not very meticulous in recording ... – Mark Daniel Johansen Sep 09 '14 at 13:37
  • ... deaths, then there could be many people on the rolls who have long since died but are shown as still alive, and so the number of very old people would be exaggerated. Or if, as others mentioned, there is an incentive to not report a death so you can continue to collect retirement. Does that happen enough to skew the statistics? I have no idea. How do they know birth dates? Were accurate records made 100+ years ago and kept, or are these self-reported? Is there an incentive to overstate your age, like to quality for some government benefit? Again, I don't claim to know. – Mark Daniel Johansen Sep 09 '14 at 13:40