Novel H1N1 Mortality Rate Approaches 1%
As of today, the World Health Organization reports 1799 deaths due to the novel H1N1 swine-flu virus among a global total of more than 182,166 cases. By using these numbers, the overall mortality rate is 0.98% (although the actual rate is somewhat lower). The death rate as of August 6th was 0.82%. Again most of the deaths (~87%) have occurred in the Americas.
And what is the mortality of seasonal flu?
Using numbers from the CDC...
About 5%-20% of the US population is affected each year by seasonal influenza. The US population is approximately 300 million, so that means 15-60 million infected annually. About 36,000 people die of seasonal flu, for an annual mortality rate of 0.06%-0.24%. About 200,000 are hospitalized, for a hospitalization rate of 0.3%-1.3%.
The numbers infected with seasonal H1N1 are substantially larger at present, but the rising mortality rate associated with the novel H1N1 epidemic is concerning. Also disproportionate numbers of otherwise healthy young adults, including pregnant women, have died of novel H1N1. Historically those most severely affected by seasonal H1N1 are children, the elderly, and the immunocompromised.
According to a Stanford-affiliated web page (http://www.stanford.edu/group/virus/uda/), the death rate during the US influenza epidemic of 1918 was 2.5%.
These numbers are significant overestimates.
It's generally believed that well over a million people just in the US have been infected. The number you've taken from WHO is confirmed cases only. The vast majority of people who are infected have had mild symptoms and are not tested unless they become more ill. So the vast majority of deaths have been correctly identified, but the vast majority of cases has not been correctly identified.
So the numerator is about right, but the denominator is much too small. Fixing the denominator would make this death rate much smaller.
Also, the rising mortality rate actually represents a decrease in the rate of testing. As it becomes clearer that H1N1 is not a severe disease, doctors have been less aggressive about testing mild cases. Thus an apparent increased mortality rate is simply a reduction in testing rate.
Thanks for caring, jcm52. The pitfalls of calculating the 2009 H1N1 mortality rate with WHO's numbers was addressed in this later post: http://bmartinmd.com/2009/08/h1n1-mortality-rate-pitfalls.html.
There is a difference between mortality rates and death rates that is frequently confused
Mortality rate is deaths per 100,000 population via a particular disease whether you are ill or not. Case fatality (death rate if you have a particular disease) is probably what most people are interested in.. That is : what are my chances of dying from a particular disease in this case H1N1 ?
Most recent news reports roughly 4,000 deaths related to H1N1, and "millions" infected (actually somewhere near five million.) That is a mortality rate of .0008, which is nowhere near the annual mortality rate of a typical influenza season. There is, at the very least, strong anecdotal evidence that brings into question both the efficacy and safety of the swine flu vaccine, yet there is a huge government backed advertising campaign (I believe that is not too strong a term) to encourage the administering of this vaccine to as many as possible. Why? The seasonal flu, and the vaccine itself, seem to pose far greater risks than H1N1 itself. As Dr. Pitt stated in a previous post, there is a difference between mortality rate, and case fatality, but that somewhat insinuates that the possibility of dying from H1N1 may actually be high, while the likelihood of actually becoming infected in the first place is low. The opposite seems to be true, with H1N1 being highly virulant, and a comparatively large segment of the population either having already had the infection, or likely to become infected. Again, this would indicate both a low mortality rate, and low case fatality, although highly infectious. In no way do I wish to make even one death seem insignificant, but the government and media attention are not warranted by the facts.
correction: In my previous post, I stated the H1N1 mortality rate as .0008, but I neglected to make a final entry into the calculator. Per my original numbers, the rate should be .008, or 8/10th of one percent. This still is at the low level of normal for the seasonal flu. However, news this morning now states 22 million infected since April, with 4,000 deaths. That would change the roughly figured mortality rate to .0018, which is, to repeat, exceptionally far below the seasonal flu mortality rate, and approaching statistical zero.
Readers of this post are advised to move on to later posts that address calculations of the mortality rate or case-fatality ratio associated with the pandemic influenza virus. For example: http://bmartinmd.com/2009/10/millions-of-americans-infected.html, http://bmartinmd.com/2009/09/harvard-expert-low-h1n1-death-rate.html, and http://bmartinmd.com/2009/08/h1n1-mortality-rate-pitfalls.html.
Harvard epidemiologist Marc Lipsitch estimates the 2009 H1N1 mortality rate at 0.007%-0.045%. To Reuters, he described the mortality risk as comparable to that of a "moderate" influenza season (eg, less than 0.1%).
For those interested, the model used by Lipsitch and the CDC to estimate case numbers and hospitalization rates due to pandemic flu is recommended: http://www.cdc.gov/ncidod/EID/index.htm.