Global deaths from influenza

​a new analysis in the Lancet found that the global influenza-related mortality is likely much higher than previously suggested (see doi.org/10.1016/S0140-6736(17)33293-2 ).

Details:

--they assessed country-specific excess mortality rates (EMR) related to influenza for countries having vital death records and influenza surveillance data

--they then extrapolated data to countries without such records by dividing countries into 3 analytic divisions (stratified by country-specific global health estimates of respiratory infection rates) for 3 age-groups (<65 yo, 65-74 yo, >75 yo), using the WHO Global Health Estimate respiratory infection mortality rates, attributing estimates of influenza-associated mortality from countries with similar mortality risk

Results:

--EMR-contributing countries (33 of them) represented 57% of the global population (4.15 billion of 7.34 billion in the world)

--the 33 countries with EMR data: 2 from sub-Saharan Africa, 13 Europe, 8 in the Americas (3 North America, 5 from Central and South America), 2 from Southeast Asia, 8 Western Pacific

--the estimated mean annual influenza-associated respiratory EMR was (per 100,000 people):

    --<65 yo: 0.1-6.4

    --65-74: 2.9-44.0

    -->75: 17.9-223.5

--overall projected seasonal influenza-associated respiratory deaths: 291,243-645,832 (4.0-8.8 per 100,000 individuals) annually

--highest estimated mortality rate: sub-Saharan Africa (2.8-16.5 per 100,000), southeast Asia (3.5-9.2 per 100,000), and among people >75 yo (51.3-99.4 per 100,000)

--for 92 countries, estimated mortality for kids <5 yo was 9,243-105,690 annually (these 92 countries were chosen because they had high rates of mortality due to respiratory infection)

--their Table 2 has the rates of EMR by country and age group. for those >75 yo, for example, the US has 49.4/100,000 population, behind Australia (20.8), Canada (44.5), Czech Republic (20.4), Germany (21.0), Israel (29.3), Japan (27.5), New Zealand (36.7), Romania (20.2), Serbia (17.9), South Korea (24.9), Switzerland (33.4), and Taiwan (33.4)

Commentary:

--prior WHO estimates were that 250,000-500,000 influenza-related deaths occurred annually (3.8-7.7 per 100,000 people) based on 2005 data, though details of the methodology for these calculations are unclear

--the methodology used in the above study is probably better than in previous studies, since it incorporated an adjustment for differences in influenza death risk between countries and possible differences in underlying health status and access to care. they also did not depend on either influenza being listed as the cause of death on the death certificate or that it had to be confirmed virologically

--there is much temporal variability in influenza-associated deaths, likely from different viral strains each year and their severity, and this is reflected in the wide range of influenza-associated deaths noted above

--clearly there are many limitations to the above study, paramount being the availability of accurate data from around the world, including evaluation by age groups above. also, they did not include circulatory causes of deaths in their estimates above (and some estimates are that about 1/2 of influenza-related deaths are non-respiratory, esp in the elderly, though there might be more variation in this by different countries, eg by the differences in prevalences of cardiovascular disease)

--there is large variability of vaccine effectiveness, being around 48% for the 2015-16 influenza season, 42% for  2016-7, but was only 19% for the 2014-5 season (vaccine effectiveness mostly depends on how good the match is between the actual circulating flu and what the vaccine actually ends up covering). BUT, even if the full protection rate is low, vaccination may well decrease the likelihood of more severe disease. for example, last year's vaccine, per the CDC, the vaccine reduced about 30% of flu hospitalizations (37% in older adults), and 42% of outpatient visits. And this year doesn't look so good for vaccine effectiveness, based on preliminary data from Australia (where flu reached its peak in mid-August), where they found 2.5 times the number of laboratory-confirmed notifications of influenza, especially in adults >80 yo and in kids 5-9 yo, though the severity of infection in people hospitalized with influenza was "on the low end of the historic range", per the CDC report

--why are the US numbers not better (12 countries with lower rates of influenza)?? i have not come across any rigorous studies, but suspect that the issues include: social conditions (there are some small reports which suggest that poorer families living in areas with high housing density or with lots of little kids get the flu more often; studies of other infectious diseases such as strep pharyngitis have found these social conditions to be predictive of higher disease incidence), access to health care (reflecting the maldistribution of health services to poorer and more rural areas), and health insurance/access to affordable health care (no systematic access to health care for all, and this seems to be going in the wrong direction now)

so, especially in severe influenza outbreaks (which are hard to know in advance), these staggering influenza-associated death rates reinforce the importance of systematic vaccination programs. it is probably useful for providers to encourage flu vaccination, both for the individual patients being seen and for protection of elders/higher risk people they might have contact with (there are studies confirming that immunizing caregivers and the more vaccination-responsive young healthy people near them is more effective than vaccinating the elderly themselves, though these studies were before the higher-dose vaccine now used in those >65 yo). i think that we as clinicians do have a strong role in reinforcing this more social message that vaccinating young people is important to protect the larger population and especially the elderly. A related article just came out in the journal Pediatrics showing that mandatory parental counseling by clinicians for immunizations in Washington State led to a 40% decrease in their vaccination exemption rates ( doi: 10.1542/peds.2017-2364  ).  And i think we should undercut the mass media message that flu vaccine effectiveness is not so great this year, given the likelihood that there is at least a shift to less severe cases even if flu itself is not completely avoided.

see http://gmodestmedblogs.blogspot.com/2018/09/maybe-we-should-delay-giving-flu-vaccine.html  which notes that there might be waning effectiveness of the flu vaccine, and suggesting that we should wait to give the vaccine until closer to the onset of the invasion of influenza

to receive these blogs regularly as emails, please contact me at gmodest@uphams.org