Cross protection between the first and second waves of the 1918 influenza pandemic among soldiers of the Canadian Expeditionary Force (CEF) in Ontario

[tetano]

Vaccine. 2015 Nov 4. pii: S0264-410X(15)01576-5. doi: 10.1016/j.vaccine.2015.10.120. [Epub ahead of print]
[h=1]Cross protection between the first and second waves of the 1918 influenza pandemic among soldiers of the Canadian Expeditionary Force (CEF) in Ontario.[/h] Bogaert K[SUP]1[/SUP].
[h=3]Author information[/h]

[h=3]Abstract[/h] [h=4]OBJECTIVES:[/h] This research analyses morbidity and mortality during the 1918 influenza pandemic among Ontario soldiers in the Canadian Expeditionary Force (CEF). This paper asks: did exposure to influenza during the first wave confer protection against illness and death during the second wave of the pandemic?
[h=4]METHODS:[/h] Pneumonia and influenza (P&I) cases and deaths among Ontario soldiers were transcribed from the 1918 Admission and Discharge books for the CEF. Following the methods of Barry et al. [10], hospital admission and mortality rates for P&I were compared for new recruits (<1 month service) and seasoned soldiers (>1 month service) in order to assess the possibility of cross protection during successive waves of the pandemic.
[h=4]RESULTS:[/h] The first wave of the 1918 influenza pandemic occurred between March and May of 1918, with the second wave erupting from September to December. Mortality in the second wave was more severe, with a case fatality rate of 4.7%, which was more than double the rate of 2.3% from March to May. Seasoned soldiers experienced 82.5% protection from illness due to P&I illness in the fall, and 84% protection from death.
[h=4]CONCLUSIONS:[/h] The morbidity data for the soldier population of Ontario, data unavailable for civilians, confirms the presence of a herald wave in Ontario. The findings support the hypothesis that exposure to influenza during the first wave of the pandemic had a protective effect during the second more deadly wave in the fall. Regional heterogeneity characterized the pandemic among soldiers in Ontario. Conscription practices may have funnelled vulnerable recruits, such as rural farmers, into training camps after the first wave of the pandemic, but prior to the second wave.
Copyright ? 2015. Published by Elsevier Ltd.


[h=4]KEYWORDS:[/h] 1918 influenza pandemic; Cross immunity; Military data; Morbidity; Mortality; Vulnerability

PMID: 26546737 [PubMed - as supplied by publisher]

 

[gsgs]

K.Bogaert , PhD thesis, Nov.2015 about the origin of 1918 flu , 186 pages

https://macsphere.mcmaster.ca/bitstr...015Nov_PhD.pdf

CASUALTIES OF WAR? AN ETHNOGRAPHIC EPIDEMIOLOGY OF
THE 1918 INFLUENZA PANDEMIC AMONG SOLDIERS IN CANADA

she has an extensive list of all the ships to/from Canada in summer 1918 and analysis
of flu-cases on board. I think she concentrates too much on Canada while the key
to understand the 2nd wave lies elsewhere.

She argues that influenza was present in Canada and Taubenberger found the 2nd wave virus
also already in summer,spring in USA, so the 2nd wave was just a continuation of the first wave.

However, Taubenberger only published HA, we don't know whether he also examined the other segments.
There could have been some reassortment in July 1918, keeping the same HA, but making
it more virulent.
The origin of the 2nd wave was quite well tracked (Brest ?), it didn't start at distant,
different locations and times.


1. In what ways did the war effort intersect with pandemic influenza to affect soldiers in
the Polish Army Camp at Niagara-on-the-Lake, across Ontario, and on troopships bound
for Europe?

# I'm not so much interested in this. The 1st wave came to Europe probably from USA
# on multiple ships

2. What patterns of morbidity and mortality characterize the first two waves
of the pandemic in Ontario?s military hospitals?

# we do have the mortality statistics from USA,England which should include much more cases

3. Were all soldiers equally vulnerable
to illness and death from influenza?

# I would reformulate it : how much immunity was there from wave 1 to wave 2, does it apply
# to the seasonal wave in April 1918 in USA


Most histories of the 1918 influenza pandemic in Canada discuss shipboard
epidemics in passing, and tend to focus on outbreaks during the second, fall wave of the
epidemic. Chapter 5 of this thesis examines in detail the events on board HMT Pannonia
and HMT Atreus during the summer of 1918, showing how these influenza outbreaks
foreshadowed the experience of Canadian soldiers on board HMT Huntsend, Victoria, and
City of Cairo in the following September and October.

# we do not care so much about "wave 1" , do we ? Where and when and how did wave 2 begin ?
# the wave 1 virus might had established another different influenza strain into the human
# population, as seen in 1889,1957,1968,2009. No big difference from the "view" of the virus,
# its evolution. But we, the hosts suffered 50M deaths plus many more headaches etc.


I think , it does not matter much, how Canada handled the spread through
incoming or outgoing ships. We know, <noone escapes the pandemic influenza>
so Canada might just have delayed it a bit.
However, Canada helped, to sequence birdflu (A/mallard/Alberta/...), which is appreciated.
Closest to 1918-flu is A/duck/Manitoba/1953/H10N7, showing that the 1918 virus had come
from American birds (in most of the 8 segments), not Eurasian.

> There is mounting evidence that influenza in Canada in 1918 was influenced by
> social inequality, where heightened mortality characterized the experience of Indigenous
> groups (Herring 1993; Lux 1997; Kelm 1999a; Sattenspiel and Mamelund 2013;
> Mamelund, Sattenspiel, and Dimka 2013), along with impoverished working class
> immigrant neighborhoods in urban centers (Jones 2007; Swedlund 2010; Herring and
> Korol 2012).

no, I think in general the converse is true. Except the problem with immunity-naive
indigenious populations. (source: bureau of the census, mortality statistics)
1918 flu attacked everyone, poor and rich, simple and noble, white and black.
I mean, more so than flu usually did in the years before 1918.

> While the case fatality rate of P&I among soldiers in Ontario during the fall wave
> was higher than in the spring,

much higher !

> it is important to remember that the majority of soldiers
> who contracted P&I survived their infection.

well, "majority", that's just 50%, not very becalming

> Viral panic can obscure who is actually vulnerable during a pandemic,
> which is mediated by social, economic and environmental
> factors (Herring 2009).

who breathes in the virus.
You can hardly avoid it.
There is no really practicable strategy to avoid it for the masses.

> There is also a tendency in the literature on the 1918 influenza pandemic to
> attribute heightened mortality during the second wave of the pandemic to a genetic
> mutation in the virus (Crosby 2003; Humphries 2013).
> As discussed in Chapters 4 and 5,
> pandemic influenza was already present in Canada well in advance of the outbreak of the
> fall wave. Chapter 4 provides novel evidence in Ontario that soldiers who were likely
> exposed to P&I during the spring of 1918 by virtue of their military service were
> protected from illness and death during the second wave of the pandemic, which suggests
> a close relationship between the circulating strains of influenza (Barry, Viboud, and
> Simonsen 2008; Rios-Doria and Chowell 2009; Korol 2011).

yes, antigenically. But we know, that one single mutation can severely affect the virulence.
The PB2 and NS segments had been identified to be responsible for the virulence of 1918-flu
in animal experiments. Antigenics is just about HA (and to a lesser degree NA).
not about PB2 or NS or G225E

> While it is possible that an as yet undetected
> change in the virus resulted in heightened viral virulence, it is implausible that this was
> the sole determining factor contributing to heightened mortality during the second wave
> of the pandemic.

well, it's what we usually see with influenza. The same virus results in the same virulence,
[ except maybe 1951 Liverpool and Canada/North-USA ]
given similar immunity-status of the host, of course.
I'm not aware of counter-examples here, which should exist if it were true.
Given the large amount of flu-sequences at genbank.

The importance of a complex array of social, economic and
Ph.D. Thesis
K. Bogaert McMaster University
Anthropology of Health

> environmental factors operating during the second wave, such as temperature and
> humidity facilitating the spread of influenza, along with crowding in major urban centres
> associated with poverty, and congregating indoors to avoid cold weather, and the opening
> and closing of schools, deserve consideration (Hallman 2009; Slonim 2010; Earn et al.
> 2012; He et al. 2013; Sattenspiel and Mamelund 2013; Mamelund, Sattenspiel, and
> Dimka 2013).

but it was always the same. Worldwide. First wave = mild, Second wave = many deaths.

> social conditions , ..., other respiratory infectious diseases

I don't see these connections in the mortality statistics. The associations in the literature are vague

> Seasoned recruits were probably protected from illness and death because of exposure to
> the herald wave of the pandemic during their tenure in the military, combined with
> acquired immunity to bacterial infections like pneumonia (Barry, Viboud, and Simonsen
> 2008; Brundage and Shanks 2008)

I would suspect that viral immunity is at play here, not bacterial. I've not seen papers
about such bacterial immunity. And pneumonia was rare in influenza 1918 1st wave.

> democratic disease

> Pathogens alone cannot account for epidemics ? a
> complex array of bio-social factors are always at play

but we still have that _amazing_ history of influenza pandemics. And always -where we were
able to identify- had been identified a new reassortment, a new HA, where we had no immunity,
as the cause of the pandemic. Bio-social factors would be local, but it's a <pan> demic.

> tuberculosis

that connection is very dubious, AFAIK.
And we had no tuberculosis in 1957,1968,1977,2009 , but still pandemics.

> Significant social, cultural economic and political factors included human
> behavioral responses to the pandemic, such as the implementation of quarantine, seasonal
> patterns of behavior, the availability of resources such as medical care, food and housing,
> and other factors such as the movement of people in conjunction with the war effort

these may have delayed the spread, but not really affected it in the long term.
Again, it's a global thing. Most deaths happened outside Canada, and in non-war-countries.
History of multiple influenza pandemics had shown that those above factors do not really play a role.
Decisive is instead the immunity to that virus.
While future genetic research may
help clarify the relationship between the different waves of the 1918 influenza pandemic
circulating around the world at different places and times, further research should also be
directed towards exploration of the social, environmental and other host factors that may
have made the second wave of the pandemic so deadly.

The April-1918 peak in pneumonia deaths by US-State is most pronounced in
MI,MN,MT,WI - so just towards the border with Canada.
Also in CO,IL,IN,ME,OH,PA,RI,UT,VT,WA, but almost no increase in April
in CA,KS,LA,MD,MO,SC,TN,VA.
Well, it may have started in KS in Jan but already declined in Apr because of widespread immunity.

 

[gsgs]

I had to reread about Haskell,KS (Jan.2018),Ft.Riley(Mar,1918) --> other US Camps --> France...

But, there was also a spring wave among the civilists in USA, with peak in April.
While the normal seasonal wave was apparently early (Dec.-Jan).
That April-wave was reported in a paper about NY with a higher than usual
mortality in young adults. Not as high as the fall-wave, but still more than in 1917.
There is also data from Indiana,Philadelphia,Kansas,Massachusetts
(deaths from p+i by month and age) which seem to support this. (I only
checked Indiana and Philadelphia so far).
Now, that April-wave was most pronounced in MT,MI,MN,ME,WI , just the North of USA,
the border to Canada. Also CO,IL,PA,OH,UT , not so much CA,KY,LA,MD,SC,VA
KS,MO had relatively many p+i deaths constantly from Jan to Apr, no increase
in Apr.
They really should have examined deaths from p+i by month and age already
for 1918 in some countries !

http://magictour.free.fr/US18S1.GIF

http://translational-medicine.biomed.../1479-5876-2-3

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1182402/
> These findings are inconsistent with the prevailing hypothesis of a spring 1918 Kansas origin

seasonal influenza was mild in Kansas in 1917/1918 as compared to 1916/17
monthly deathcounts for influenza in Kansas :
1917:170,147,103,44,19,5,2,3,2,1,16,30
1918: 58,47,57,38,23,12,6,2,7,1240,1014,1569
hard to believe that there were "dozends" of influenza deaths in Jan-Feb 1918 in Haskell County alone.
And that the April wave in NY,PA,OH,IL,MI,Toronto was "created" here.
Typically flu is seeded already ~6 months before in Asia, USA.
A pandemic virus could be faster, the 2009 ****** causes a May/June wave in NY but not much
in other States until fall. Did it go from Haskell, Jan.1918 to Michigan,Toronto in April into
the civil population just by soldier-movements from camp to camp ? The April wave was also clearly seen
in distant OR,WA,MT away from Europe-war. Looks more like influenza B, which we often have late in April.
Does it kill proportionally more younger adults than seasonal flu in 1917 ? Maybe. The first recognized
flu-B outbreaks were in 1936,1940, I'll check ...

 

[gsgs]

here is the note mentioned by Barry, about Haskell in Public Health Reports ,vol.33, week 14 ,1918
----------------------------------------------------------
April 5. 1918
502
INFLUENZA
Kansas-Haskell
On March 30, 1918, the occurrence of 18 cases of influenza of
severe type, from which 3 deaths resulted, was reported at Haskell,
Kans.
--------------------------------------------------------

Barry also quotes the Santa Fe Monitor from February 14th, 1918, reporting about
several cases of pneumonia.
(What did they report about the 1916/17 flu, which was more severe in Kansas ? )

> If the virus did not originate in Haskell, there is no good explanation for how it arrived there.
> There were no other known outbreaks anywhere in the United States from which someone
> could have carried the disease to Haskell, and no suggestions of influenza outbreaks in
> either newspapers or reflected in vital statistics anywhere else in the region.

there were seasonal waves each year since 1889/90. It may just have been worse in Haskell
for some reason. This is reflected in the vital statistics. There were deaths from influenza
and pneumonia marking the wave. It was most pronounced in March,April in the North,
KS,MO,KY had almost constant cases from Jan. to April, in the South there was more in Jan
and then it declines, while the Apr. wave is almost absent. So it may have started in the South,
Texas, camp Travis/TX, camp Cody/NM and I speculate it came from Mexico.

> Both contemporary epidemiological studies and lay histories of the pandemic have identified
> the first known outbreak of epidemic influenza as occurring at Camp Funston, now Ft. Riley,
> in Kansas.

source ?
We do have the weekly pneumonia case counts for the military camps.
http://www.ncbi.nlm.nih.gov/pmc/journals/347/
These were young soldiers, so it probably was the new Spainflu virus, 1st wave,
which attacked the young adults. The counts for the camps coincide with the
monthly pneumonia death counts for the US-States and cities
from "mortality statistics, 1918", table 10.
There was a March-April wave, while the rest of the season was unremarkable.
There may have been a normal wave from another virus in Jan.
Now, the counts for the camps don't show Funston as the first. Funston/KS had a
wave parallel to camp Dodge/IA , from week 11 until week 25 with peaks in week 18,21.
Pretty much parallel with the Camp-average, which peaked in week 17, and with the civilist wave.
Earlier were camp Cody/NM, camp Travis/TX with peak in week 10 or before.

 

[gsgs]

On March 15, 1916,[33][34][35] Pershing led an expedition into Mexico to capture Pancho Villa. 10,000 men penetrated 350 miles (560 km) into Mexico.
from March 14, 1916, to February 7, 1917
Pershing was permitted to bring into New Mexico 527 Chinese refugees
After departing from Fort Jay at Governors Island in New York Harbor under top secrecy in May 1917, Pershing arrived in France in June 1917.

[Germany had sent a poorly encrypted telegram to their Washington-ambassador, to
be forwarded to the Mexicans for a joint mission should USA enter WW1.
The British decrypted it and used it to convince USA on entering WW1 against Germany
and Pershing was chosen as commander in that war.]
Trump : My Grandfather fought with Pershing in the Mexican Insurrection

> the [Pershing-] expedition remained in Mexico until February 1917 to encourage Carranza's
> government to pursue Villa and prevent further raids across the border.

so, if Spainflu1 had come from Mexico, then it probably did so after Feb.1917 or with the return
of the Pershingistas.
If it had come in early 1918, then there would just not have been enough time to cause
a full wave in most of USA in March/April.
Remember, Mexflu came at Easter 2009 from Cancun to New York by airplane and caused
a wave in NY in May/June , but not much in the rest of USA until fall.

Let them examine the age-distribution of Texas p+i deaths from Dec.1917-Feb.1918 to resolve this !
The data should be available.
And, if USA keeps them secret, then (some of) the genealogy people may provide them from their
databases against money in some years ...
Maybe the 100th anniversary of Spainflu will be a good opportunity for Trumpaganda
to blame it on Mexico and to justify the expenses for the wall, which Mexico won't pay.

I wonder whether there was pigflu in Pancho Villa land ...

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

while that Mar/Apr 1918 wave (Spainflu1) cannot be seen in the normal vital statistics
for the Southern US-States, it _can_ be seen in some Southern cities like
New Orleans,Birmingham,Atlanta,Louisville,Kansas-City, Saint Louis, Nashville.
In general it was more prevalent in urban areas than rural. See the Ontario-picture above.
Maybe because it didn't make it to rural areas before summer came.
Maybe because it was transported by railway to the big centers first.
Maybe because it arrived at the ports by ship, maybe it started in NY and was too slow
to reach the South before summer.

Was this Spring - "herald" wave also seen in other parts of the world ?
I don't know. The usual vital statistics goes by years and thus the 1918 spring
is overturned by the 1918 fall.
We had it in Spain,Birmingham/UK,Switzerland in Summer- but that was probably introduced through
US-troups from Brest. It was reported to have slowed the German 1918 "Michael" offensive, (1918,Mar.25)
but that was probably just a misquote from Ludendorff's memoirs.

 

[gsgs]

http://demingnewmexico.genealogyvill...Cody/hst15.htm

The first increments of troops sent to this camp were National Guard men from Minnesota,
Iowa, Nebraska, North Dakota, and South Dakota. About 12,000 of these National Guard
troops arrived be tween the 1st and 30th of September, 1917. The strength of the camp
was gradually increased by increments from other camps, so that the average strength
for the month of December was approximately 21,000

During the latter part of 1917 there was a sharp outbreak of influenza. [in Camp Cody, NM]
In the month of November the rate for this disease was 223.67. It increased to 329.64 in
December. It declined during January, 1918, to 177.30. It further declined during the month
of February and slightly increased in the month of March with a still further increase to
180.64 in April.
A large number of cases of pneumonia and "other respiratory diseases had been reported
in the latter part of 1917 and in the early part of 1918. These diseases declined during
February and March and did not show the increase in the month of April that occurred
in so many of the camps

------------------
this was clearly before the outbreak at Camp Funston in KS and even before the "outbreak"
in Haskell County. It may have been the reason, why the South didn't get that April-wave so much.
Or it may have been the same virus, that later went North ?!

 

[gsgs]

from the "report of the surgeon general of the army"
just searching for keyword influenza

[1915] War and famine have paved the way for typhoid fever, smallpox, typhus fever, and
influenza in severe forms, and these epidemic diseases have all pre"\ ailed in Mexico
and among the refugees who have flocked to this side of the border.

Acute follicular tonsillitis and influenza have prevailed extensively during the colder
months, especially at Douglas, Ariz.

mfluenzu.There were 3,305 cases of influenza, Regulars and militia included,
icli is about 7 per 1,000 less than for the whole Army in 1915.
From January I to April I. 1917, 2,722 more admissions occurred:
1,943 of these were from the El Paso and El Nogales districts.
These last figures only indicate cases so far recorded from sick
1 grounded reports in this office, and many reports are still to be received.
The en are therefore far too low.

Pneumonia.&#151;In November and December, 1916, the sick and mortality rates
both Regulars and militia began a disquieting rise on account of lobar pneumonia
broncho-pneumonia, the latter-named disease generally a complication of measlea
influenza.

In the early months of 1917, January, February, and March, pneumonia of be
kinds reached the proportions of a veritable epidemic in the mountainous west&#128;
districts of El Paso, Deming, New Mexico, and Arizona.
Pneumonia was coincide
with an extensive epidemic of measles, mumps, and influenza among the civil popu
tion as well as among the troops.
Lobar pneumonia was occasionally and broncho-pneumonia generally a complic
tion of measles or influenza.

The following recommendations were made, and while some of them were made in
inuary, 1917, they relate to conditions in 1916, and therefore it is thought proper to
elude them: "January 5, 1917.
"As a means of reducing the epidemic of pneumonia and influenza, which has become
aite serious,

"January 6, 1917.
"1. The pneumonia situation has reached such a condition of seriousness that it is
lought proper to recommend every procedure that will have a tendency to reduce it.
"2. The histories of many of the cases show that they experienced a chilling of their
odies. which was soon followed by pain in the side and sickness.
Many of these men go to town in the early afternoon with only a sweater, which,
after all, is a very po garment for warmth, and going home late at night probably
in a jitney they becor chilled.
Furthermore, in going to town they usually land in the picture show, whe
they inhale an atmosphere impregnated with pneumonia, influenza, and
probab cerebrospinal meningitis germs.
"3. Therefore it is recommended that the Alabama troops be not allowed to be abee
from their camps after dark, and that all troops be prohibited from attending pictu
shows or other places of assemblages until this epidemic of influenza, pneumonia, ai
measles is over."

'mild but very contagious epidemic of influenza occurred early in the year, lasting
it one month and affecting some 16 per cent of the corps.

At one dusty (amp in this
country, two severe dust storms were followed on each occasion by a
rapid increase of influenza, a great many cases of which were compli-
cated by streptococcus pneumonia.

Influenza, or snipping fever, was very prevalent.
Nearly every animal purchased had the disease in a more or
l<^s virulent form and frequently associated with
it were the usual complications like pneumonia, pleurisy,
laminitis, and gastrointestinal catarrh.

the big Melbourne .pdf has a report about flu in horses in France during the 1918-wave


[ what I think that should be done: calculate the civilian deaths from P+I in these outbreaks
in 1916-1917 and 1917-1918 by month and age, so to see whether a shift towards
younger deaths already occurred in some places in 1917 or whether this all just only started
in Mar/Apr 1918 , but then again it must have cocirculated since some time, because
that Apr.-wave was omnipresent in almost all US-cities and Toronto (except maybe
LA,Baltimore,Boston,)
I'm not sure, but I think this Mar/Apr.1918 age-shift was not seen in Europe , where the first wave
occured May-Jul, probably brought by US-soldiers]

 

[gsgs]

here is the best,quick evidence against that Kansas origin of 1918-flu :
http://magictour.free.fr/CHICA18A.GIF
The first case in Camp Funston was on March.4 (the army-cook) !

pneumonia was a reportable disease in Chicago in early 1918, also in some
other US-cities, but not in New York.
See the weekly "Public Health Reports" .

------------------
Jan. and the first part of Feb. had been unusually cold in USA,
(file g:\noaa\ny1910.bmp)
maybe that explains some of the severe cases in Haskell.
--------------