OT, 50.8% Increase in DEATHS in NY, compared to first week Jan 1999, "experts" attribute it to wanting to see 2000 (yet no decrese for weeks prior to support this claim)

greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread

http://www.usatoday.com/news/ndssat02.htm

01/15/00- Updated 03:45 PM ET

NYC deaths up in early 2000

NEW YORK (AP) - The city's death rate jumped in the first week of January, suggesting to experts that some terminally ill patients held onto life just long enough to see the year 2000 arrive.

''The will to live can be pretty powerful,'' said Robert N. Butler, founder and president of the International Longevity Center.

Preliminary numbers from New York's Department of Health show 1,791 people died in the city in the first week of 2000 - a 50.8% increase from the 1,187 deaths for the same period last year and 46% more than the 1,226 deaths in the first week of 1998, The New York Times reported Saturday.

''It's pretty well established that people who are seriously ill will hang on to reach significant events, whether they are birthdays, anniversaries or religious holidays,'' said Richard Suzman, an associate director of the National Institute on Aging in Bethesda, Md.

''In this case, making it into the next century or millennium certainly counts as that.''

The increase was not caused by this year's influenza virus, because the flu has been occurring at normal levels in the city, said Health Department spokesman John Gadd.

National statistics are not yet available, and a spokesman for the city's health department said such data would take considerable time to analyze.

--------------------------------------------------------------------------------

-- Hokie (Hokie_@hotmail.com), January 16, 2000

Answers

A 95% confidence interval means that one result in twenty appears correct even though wrong. No offense Hokie, but I bet this falls into that category (obviously my *personal* opinion, being far toooo lazy to look up the last 20+ yrs. death results for myself.)

Asking Patience,

Frank

-- Someone (ChimingIn@twocents.com), January 16, 2000.


What? Where'd you get that confidence rating? What does it apply to? I hear confidence ratings in polls, which is the "error" figure generated to apply the responses of the sample polled to a larger population. For instance, if I ask 1,000 members of a 10,000 member club who they are gonna vote for as club president, then I can generalize their responses to the entire group, with a certain degree of confidence given statistically derived ranges of error.

Now this article lists actual body counts. There is no error generated by estimations of deaths, and so no confidence interval. They actually counted the number of deaths in the whole population, as opposed to measuring the deaths in one town, then estimating how many died in a greater area.

If you are referring to another article, then please link, because I would be interested to read it.

I would also be interested to read the death rates in the weeks prior to 12/31 for the years mentioned in this article, because those figures are not given and so clearly we lack data to support the conclusion of the 50.8% increase being folks who would have died sooner statistically had they not "held onto life just long enough to see the year 2000 arrive".

-- Hokie (Hokie_@hotmail.com), January 16, 2000.


We know have the way to immortality!

Just declare every day a major world-shattering holiday and these terminal patients could live forever.

That is the point of this article, right? ;-)

-- cgbg jr (cgbg jr@webtv.net), January 16, 2000.


Could this have anything at all to do with the spraying of Malithion? Ummmm...I wonder.

-- eating crow (suzan@monad.net), January 16, 2000.

In order to attribute the higher death rate to terminal patients hanging on until 2000 one would have to know how many were terminal. An analysis by age would be helpful though it would not prove it. Another possible cause might be higher death rates because of the discharge of firearms in celebration. In other words, more info is needed.

-- Evelyn (equus@barn.now), January 16, 2000.


I have heard this theory before--that people tend to die AFTER hoilidays and their birthdays. Why not?

-- Mara (MaraWayne@aol.com), January 16, 2000.

From today's Electronic Telegraph:

'Millennium fever' staved off deaths By David Bamber and James Langton

BRITONS are dying in record numbers this month, having apparently willed themselves to see in the Millennium.

More than 20,700 people died in the first week of the New Year, an upsurge on the previous week and a significantly higher figure than over the same period in the previous five years. The phenomenon has been attributed by health officials to an "unanticipated brand of Millennium fever". Medical experts believe that many would have died earlier, but had hung on because of the significance of the event.

Robert Butler, the president of the International Longevity Centre, a US-based research and education group, said the figures, had been reflected across the world. He said that a number of centenarians had survived into the New Year through sheer resolve because they were motivated by the rare chance to live in three centuries. "The will to live can be pretty powerful," he said.

Figures from the Office of National Statistics show that 20,772 died in the week to January 7 - a 65 per cent rise on the 12,629 people who died in the week to December 31. During both weeks the flu virus was rampant, so the huge increase in deaths cannot be attributed to that alone.

On average 15,900 people have died in the first week of each of the past five years. In New York there has been an unusually high number of deaths during the first week of 2000 - nearly 1,800, more than 50 per cent more than usual.

-- Old Git (anon@spamproblems.com), January 16, 2000.


Hokie, No, this isn't a poll, but it IS a sample of a whole. What they are looking at is the death rates in 3 weeks out of the past say 105 weeks (98 till now) To me that represents a very small sample from which to draw a conclusion. If they wanted to look at just the first week of each year, they should have looked at *many* first weeks, say 20 years worth or so (and accounted for changes in population if significant.)

Also, I used 95% as my confidence interval as a P value of .05 is standard in the social sciences. For me the implication of this is that the chance of ANY study showing a positive result when the true result should be negative is one in twenty. With such a small sample as they used (3 weeks) the chance of error is very high. While I don't have the data to support this, I'd bet that if I had all N.Y.'s weekly death data I could find several occurrences like this that are just statistical flukes and have nothing to do with the millenium.

But then, not having the data, I could be wrong.

Regards,

Frank

-- Someone (ChimingIn@twocents.com), January 16, 2000.


Anyone else find it curious that both articles SPECIFICALLY mentioned that the increase has nothing to do with the flu? Is there really a need to proactively deny the impact of the flu?

There are many other arguments which could be made about the conclusions one could make about these statistics. MAny are mentioned above.

But Flu???

Just how bad is this current flu epidemic anyway???

-- Duke1983 (Duke1983@aol.com), January 17, 2000.


MOST interesting.

-- h (h@h.h), January 17, 2000.


From: Y2K, ` la Carte by Dancr (pic), near Monterey, California

Flu easily impacts mortality statistics. Every year there are deaths from flu; some years more than others. A sizable chunk died from flu in 1918.

-- Dancr (addy.available@my.webpage), January 17, 2000.


Let me ask again, just how bad is this current flu epidemic anyway???

-- Duke1983 (Duke1983@aol.com), January 18, 2000.

From: Y2K, ` la Carte by Dancr (pic), near Monterey, California

He re's how bad.

-- Dancr (addy.available@my.webpage), January 18, 2000.


Moderation questions? read the FAQ