On Mon, 27 Apr 2009 09:14:01 -0700, Christopher D. Green wrote: >Stats, stats, everywhere! > http://news.yahoo.com/s/ap/med_swine_flu_glance > >I have head on the news repeatedly over the last couple of days, people >wondering why there have been 103 deaths in Mexico, but none anywhere >else. Is the exported version "weaker"? Are the people to whom it is >exported "stronger"? > >My suspicion is that it is a simple matter of statistics (which >journalists are notoriously poor at). > >Consider: 103 deaths in 1,614 Mexican cases. That is about a 6% death >rate, officially. I strongly suspect that you will find there have been >many, many more cases of Mexicans who were infected but did not become >ill enough to report it or go to hospital. So the death rate is >probably, in actuality, much lower than 6%.
It is also pretty easy to fool oneself with statistics, especially if one doesn't calculate the statistics in way proffessionals do (i.e., epidemiologists, health statisticians, etc.). Conside the following: (1) The 1918 influenza had a "case-fatality rate" of >2.50 which, when compared to the usual case fatality rate of <0.1, is huge. See the article by Taubenberg & Morens at the CDC website for more information about the 1918 flu which may suffer less from media innumeracy: http://www.cdc.gov/ncidod/eid/vol12no01/05-0979.htm#1 (2) What is a "case fatality rate"? Wikipedia provides one definition: |In epidemiology, case fatality (CF) or fatality rate, is the ratio |of deaths within a designated population of people with a particular |condition, over a certain period of time. An example of a fatality rate |would be 9 deaths per 10,000 people at risk per year. This means that |within a given year, out of 10,000 people formally diagnosed with a |disease, 9 died. http://en.wikipedia.org/wiki/Case_fatality Additional definitions are provided at the UCLA School of Public Health website: http://www.ph.ucla.edu/epi/bioter/anthapha_def_a.html Your calculation of a 6% death rate, in this context is actually quite spectacular, being much higher that the case fatality rate for the 1918 flu which killed 20-40 million people worldwide. But your death rate is not a case fatality rate because it is not based on a standard period of time nor do we have good statistics about how many people have been been infected. Given that we don't really know how many people have been infected, any calculations that requires this number should be viewed skeptically. (3) One of the unusual aspects of the 1918 flu is who was killed by it. Quoting from the Taubenberger and Morens article: |Why Did the 1918 Virus Kill So Many Healthy Young Adults? | |The curve of influenza deaths by age at death has historically, for |at least 150 years, been U-shaped (Figure 2), exhibiting mortality |peaks in the very young and the very old, with a comparatively low |frequency of deaths at all ages in between. In contrast, age-specific |death rates in the 1918 pandemic exhibited a distinct pattern that has |not been documented before or since: a "W-shaped" curve, similar |to the familiar U-shaped curve but with the addition of a third (middle) |distinct peak of deaths in young adults ≈20–40 years of age. Influenza |and pneumonia death rates for those 15–34 years of age in 1918–1919, |for example, were >20 times higher than in previous years (35). |Overall, nearly half of the influenza-related deaths in the 1918 pandemic |were in young adults 20–40 years of age, a phenomenon unique to |that pandemic year. The 1918 pandemic is also unique among influenza |pandemics in that absolute risk of influenza death was higher in |those <65 years of age than in those >65; persons <65 years of age |accounted for >99% of all excess influenza-related deaths in 1918–1919. |In comparison, the <65-year age group accounted for 36% of all excess |influenza-related deaths in the 1957 H2N2 pandemic and 48% in the |1968 H3N2 pandemic (33). One of the fears raised by the deaths in Mexico is that so many of them were young adults and not children and the elderly. It may be possible that there are a large number of children and elderly who have died but have not be identified. As of about 2:30pm on Monday, April 27, there is an estimated 148 deaths in Mexico from swine flu. >So, the statistical lesson to pass on to your students during this >eminently "teachable moment"? Calm down. This is not a disaster. It is >very unlikely to become a disaster. There will be some deaths, almost >certainly. There are about 40,000 deaths annually from influenza in the >US in normal years. There is, as yet, little indication that this number >will rise dramatically this year. >From your lips (well, typed text) to God's ears (or eyes). For the video oriented, there are a couple of PBS shows on the 1918 flu: >From the American Experience series: http://www.pbs.org/wgbh/amex/influenza/ >From the Nova series: http://www.pbs.org/wgbh/nova/sciencenow/3318/02.html -Mike Palij New York University [email protected] --- To make changes to your subscription contact: Bill Southerly ([email protected])
