USA flu-deaths 2009/10

[Missouriwatcher]

Re: USA flu-deaths 2009/10

gsgs: Have you calculated the deaths/hospitalization rate from publicized CDC data? It appears that it is continuing to go up, even as the number of jurisdictions reporting is fluctuating.

http://www.cdc.gov/flu/weekly/weeklyarchives2009-2010/data/AHDRA49.htm

[gsgs]

Re: USA flu-deaths 2009/10

no,isn't it the usual delay ?

wikipedia now gives 2792
will it go over 4000 ? maybe

compare with Australia...

at peak they had 60 deaths
at 7 weeks after peak they had 170 deaths
now they have 190 deaths

USA is 7 weeks after peak now


reported USA panflu deaths week 35-49: (wave 2)

30,47,38,51,91,81,117,129,189,186,177,175,87,124,5 6

=1578 so far
my estimated total for wave 2 = 1800
multiply by ~5 to include indirect deaths ...............................................

[gsgs]

Re: USA flu-deaths 2009/10

remember 2003/4 , the Fujian flu,
which also started early

[Missouriwatcher]

Re: USA flu-deaths 2009/10

Yet, all jurisdictions are not reporting, so wouldn't one need to look at percentages (just as you do when figuring MMWR cities not reporting)? Week 49, only 34 reported out of, I believe, 54. Does this mean that those were the only jurisdictions with deaths and hospitalizations, or does it mean they did not send a report to the CDC?

[gsgs]

Re: USA flu-deaths 2009/10

I assume they just had no death in that week

--------edit-------------

ahh, you may have thought that for deaths, but they also give
the number of juridictions for hospitalizations and that number is much
larger, hard to imagine that some jurisdictions have no hospitalization in some week.

so, I think we should divide the numbers of hospitalizations and deaths
by the number of reporting jurisdictions



34,35,35,35,37,37,37,37,38,36,36,37,36,35,32
413,532,813,952,1599,2231,3520,4346,4924,4703,3330 ,3216,2273,1858,749=35459

30,34,38,37,38,37,39,38,40,40,40,40,38,35,34
30,47,38,51,91,81,117,129,189,186,177,175,87,124,5 6=1578

[Missouriwatcher]

Re: Weekly MMWR - Updated through Week 49

"On August 30, CDC and the Council of State and Territorial Epidemiologists (CSTE) instituted modified case definitions for aggregate reporting of influenza-associated hospitalizations and deaths. This cumulative state-level reporting is referred to as the Aggregate Hospitalization and Death Reporting Activity (AHDRA). During August 30--October 31, a total of 17,838 hospitalizations associated with laboratory-confirmed influenza virus infections were reported to CDC through AHDRA. On average, 31 states each week reported laboratory-confirmed hospitalizations during that period.

States report weekly to CDC either 1) laboratory-confirmed influenza hospitalizations and deaths or 2) pneumonia and influenza syndrome--based cases of hospitalization and death resulting from all types or subtypes of influenza. Although only the laboratory-confirmed cases are included in this report, CDC continues to analyze data both from laboratory-confirmed and syndromic hospitalizations and deaths."

http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5844a4.htm

I think this reporting is different than the MMWR report. Looking at weeks 40-49 and dividing deaths by hospitalizations, the rates are as follows:

40: .036
41: .033
42: .03
43: .038
44: .04
45: .053
46: .054
47: .039
48: .067
49: .075

http://www.cdc.gov/flu/weekly/weeklyarchives2009-2010/data/AHDRA49.htm

If an average of 31 jurisdictions (states plus territories) report, than numbers fluctuate, but percentages wouldn't? I agree that deaths may continue to be high after the peak, so maybe this isn't the way to look at the data.

[gsgs]

Re: USA flu-deaths 2009/10

the longer they are hospitalized, the more likely they die

[alert]

Re: USA flu-deaths 2009/10

the longer they are hospitalized, the more likely they die

Not necessarily. The more time that passes, the hgier percentage of illness that have reached their conclusion (recovery or death), so simply dividing deaths by hospitalizions will produce an increasing rate, as you are no longer counting those currently hospitalized as surviving.

I don't know if a longer hospitalization increases the death rate - there are many people dying within hours of admission of severe hemorrhagic pneumonia. On the other hand, it might increase the risk of a co-infection, and might indicate more complications. I think there are several cancelling factors on that one.

[Missouriwatcher]

Re: USA flu-deaths 2009/10

"Deaths are calculated as a percentage of hospitalizations using data from the Aggregate Hospitalization and Death Reporting Activity (AHDRA) surveillance system. This system has up to 35 states reporting. Using data of hospitalizations and deaths from AHDRA, by age group as reported through week 43, we calculated deaths as a percentage of hospitalizations to be: 0-18 years of age ? 1.489%; 19-64 years of age ? 5.556%; 65+ years of age ? 4.946%..."

http://www.cdc.gov/h1n1flu/pdf/details_eip_methods_plus_schematic.pdf

"EIP is a network of 10 state health departments (CA, CO, CT, GA, MD, MN, NM, NY, OR, TN) and their collaborators in local health departments, academic institutions, other federal agencies, and public health and clinical laboratories; infection control professionals; and healthcare providers. Preliminary cumulative rates of laboratory-confirmed, influenza-associated hospitalizations reported for children aged 0--4 years were 3.1 per 10,000 population by EIP and 7.3 per 10,000 population by the new sites. Rates for other age groups were as follows: 5--17 years, 1.5 by EIP and 2.9 by the new sites; 18--49 years, 1.2 by EIP and 1.2 by the new sites; 50--64 years, 1.3 by EIP and 1.2 by the new sites; and >65 years, 1.0 by EIP and 1.1 by the new sites."

http://www.cdc.gov/ncpdcid/deiss/about_eip.html

So will their estimates again change since the death percentage appears to be going up?

http://www.cdc.gov/h1n1flu/estimates/results_2009_h1n1.htm

[gsgs]

Re: USA flu-deaths 2009/10

http://www.cdc.gov/h1n1flu/estimates/results_2009_h1n1.htm

9807 was their estimate of deaths until Nov.14 or their estimate for the
whole season using the data until Nov.14 ?

http://www.cdc.gov/flu/weekly/weeklyarchives2009-2010/data/AHDRA49.htm

Nov.14 was week 45 in the CDC-enumeration of weeks.

27363 hospitalizations until week 45 --> estimate of 10000 deaths

35459 hospitalizations until week 49 --> estimate of 13000 deaths

(?)

[Missouriwatcher]

Re: USA flu-deaths 2009/10

I take it to mean roughly 10,000 deaths up to Nov. 14. How does one figure new numbers if all jurisdictions don't report? On CDC site, it says there are "up to 35 states that report". On HHS site it says AHDRA "collects information from 50 states". Somewhere I found there are 54 jurisdictions that report (I take that to mean 50 states, DC, PR, Guam and US Virgin Islands).

Originally stated in Nov, "CDC estimated that between about 2,500 and 6,000 2009 H1N1-related deaths occurred between April and October 17, 2009. The mid-level in this range was about 3,900 2009 H1N1-related deaths."

Now stated in Dec, "CDC estimates that between about 7,070 and 13,930 2009 H1N1-related deaths occurred between April and November 14, 2009. The mid-level in this range is about 9,820 2009 H1N1-related deaths." http://www.cdc.gov/h1n1flu/estimates_2009_h1n1.htm

Will these factors change in January??? Probably.