Content uploaded by Richard N Danila
Author content
All content in this area was uploaded by Richard N Danila on Nov 04, 2014
Content may be subject to copyright.
Public Health Chronicles
P H R / N–D 2007 / V 122 803
LESSONS LEARNED FROM
THE 1918–1919 INFLUENZA
PANDEMIC IN MINNEAPOLIS
AND ST. PAUL, MINNESOTA
M O, AB
S F. S, MPH
R N. D, PD, MPH
R L, MD
Those who cannot remember the past are condemned
to repeat it.
— George Santayana
Spanish Inuenza of 1918–1919 killed more than 50
million people worldwide over the course of two years.1
The true origin of the 1918 inuenza pandemic is
unknown. During World War I, propaganda in war-
engaged countries only permitted encouraging news, so
as a neutral party, Spain was the rst country to publicly
report on the health crisis.1 Thus, Spanish Inuenza
became a popular term. However, historical research
has shown that Spain was an unlikely candidate as the
initial source and some suggest that it originated in
Kansas in the spring of 1918.
Inuenza pandemics have occurred regularly every
30 to 40 years since the 16th century. Today, inuenza
experts consider the possibility of another inuenza
pandemic, not in terms of if but when. Due to the
high likelihood of an inuenza pandemic, planning is
underway in many U.S. states and other countries. We
reviewed the responses of two neighboring Minnesota
cities during the 1918–1919 pandemic to gain insight
that might inform planning efforts today.
Many of the components of current pandemic
inuenza plans were utilized to some degree in Min-
neapolis and St. Paul during 1918–1919. Coordination
between different levels and branches of government,
improved communications regarding the spread of
inuenza, hospital surge capacity, mass dispensing of
vaccines, guidelines for infection control, containment
measures including case isolation and closures of pub-
lic places, and disease surveillance were all employed
with varying degrees of success. We focus on medical
resources, community disease containment measures,
public response to community containment, infection
control and vaccination, and communications.
PANDEMIC BEGINNINGS IN MINNESOTA
Minnesota’s rst Spanish Inuenza cases were identi-
ed in the last week of September 1918. As in the rest
of the country, Minnesota’s rst cases “were directly
traceable to soldiers, sailors, or [their] friends.”2 Every
military base and military hospital in the Minneapolis-
St. Paul area was severely affected. Case isolation was
slowly implemented at both Fort Snelling and the
Dunwoody Naval Detachment (military installations
in Minneapolis). On September 30, the rst day of
isolation, cases numbered in the hundreds.3
Inuenza cases were not limited to enlisted men
for long. In Minneapolis, the number of civilian cases
outstripped the number of military cases for the rst
time on October 9, less than two weeks after the rst
case was identied in the state (700 civilian cases; 675
cases at Fort Snelling).4 Inuenza had become a report-
able condition in Minnesota on October 8 in response
to the growing epidemic.5
MEDICAL RESOURCES
Two major issues contributed to the gravity of the pan-
demic: the war effort and limited scientic knowledge.
World War I was not only costly, it required much of the
medical community to be stationed overseas. In 1918,
little was known about inuenza. While this lack of
knowledge did not negatively impact infection control
actions, effective treatment and prevention methods
were not fully utilized.
When inuenza rst appeared in Minnesota on
September 27, the state was ill equipped for a health
crisis.2 Although World War I was coming to an end,
more than four million Americans were mobilized and
the nation’s resources were directed to supporting the
war effort. An editorial in the Minneapolis Tribune daily
newspaper described the lack of physicians and nurses:
“The medical fraternity is severely taxed already. So
many physicians and surgeons have gone to Europe or
to training that those at home have more than they can
attend to comfortably and to good advantage.”6
The number of inuenza patients that needed the
attention of physicians and nurses overwhelmed St.
Paul and Minneapolis clinicians. The war’s consider-
able drain on the medical profession was compounded
by other factors that hindered nurse and physician
mobilization. Methods to keep them healthy while
caring for inuenza patients were ineffective. Many
804 P H C
P H R / N–D 2007 / V 122
health-care providers fell ill, and some died. At one
point, Minneapolis’s City Hospital reported that “nearly
half of the nursing staff has been ill with inuenza in
the last three weeks.”7 This bleak situation discouraged
some clinicians from providing their services. Dr. H.M.
Bracken, Secretary of the Minnesota State Board of
Health, reported to Dr. Rupert Blue, U.S. Surgeon
General, on his campaign to recruit physicians for the
inuenza effort: “A number who we have called for
have made excuses and have not come at all.”8 Other
physicians who were recruited by Dr. Bracken simply
did not show up.9
Dr. Bracken attempted to secure senior medical
students for inuenza work. Dr. Bracken worked not
only with the U.S. Surgeon General but also with the
Surgeon General of the Army, the Committee on
Education and Special Training, and the Dean of the
University of Minnesota Medical School for three weeks
and still was unable to obtain senior medical students
for assistance, because each party insisted that someone
else had to authorize it. In the end, Bracken failed to
receive any medical students.10
Not surprisingly, Minneapolis and St. Paul hospitals
proved to be inadequate to handle the large num-
ber of patients. Minneapolis’s City Hospital and St.
Paul’s St. John’s Hospital were solely devoted to treat-
ing inuenza patients. Non-inuenza patients were
transferred to other area hospitals. This inadequacy
was not entirely due to the lack of beds and supplies;
there simply were not enough healthy nurses. At City
Hospital, Superintendent Dr. Harry Britton reported
that the “hospital was caring for about 150 cases, and
had about 70 on the waiting list. It had beds available
for that waiting number, but not nurses.”11
In St. Paul, a system was set up between St. John’s
Hospital and other hospitals to insure an adequate
number of nurses to care for inuenza patients, but
unfortunately this system failed. Dr. F.C. Plondke, St.
John’s Hospital’s Medical Director, complained that
the other hospitals were abandoning their promises
to assign help from their nursing staff. “The other
hospitals had refused to furnish a single nurse to aid
the fteen who are caring for ninety patients at St.
John’s from their individual nursing staffs.”12
In 1918, medical science maintained that inuenza
was bacterial in origin. Physicians at Fort Snelling
claimed that the “bacillus inuenza of Pfeiffer,” which
is today known as Haemophilus inuenzae, was the cause
of Spanish Inuenza.1,13 Nevertheless, despite this lack
of understanding about viruses, advice to curb infection
was relatively accurate. The Minnesota State Board of
Health recommended the use of handkerchiefs to cover
sneezes and coughs, plenty of fresh air, avoidance of the
sick and of crowds, and to contact a physician if ill.14
CommunItY dIsease ContaInment
As inuenza was beginning to take hold in the civil-
ian population, there was disagreement between the
Minneapolis and St. Paul health commissioners, Dr.
Guilford and Dr. Simon, respectively. Their approaches
varied; Dr. Guilford tended to be broadly proactive
to prevent cases, whereas Dr. Simon tended toward
initiating activities in response to individual cases. Dr.
Guilford believed that closing public places was the
best course of action and that isolation of individual
cases was useless.15 Dr. Simon asserted that isolation of
inuenza cases would be more effective in preventing
the spread of disease.14
The St. Paul Health Department and the Minnesota
State Board of Health met Dr. Guilford’s strong advo-
cacy with opposition. Dr. Bracken, siding with St. Paul,
questioned, “If you begin to close, where are you going
to stop? When are you going to reopen, and what do
you accomplish by opening”?11
Debate between the two cities on the merits of clos-
ing schools caused further strain. Dr. Simon held that
St. Paul’s school nurses were the best defense against
the spread of the disease, and that closing schools
would allow cases to go undetected as the children
would not be under any medical supervision. Dr.
Guilford disagreed, pointing out that 30 school nurses
would not be able to adequately care for the 50,000
pupils in the Minneapolis public school system during
a pandemic.16 Minneapolis closed the schools on two
separate occasions (October 12 to November 17, and
December 10 to December 29, 1918).
Despite Dr. Simon’s conviction that the closing of
public places would be ineffective, on November 6 St.
Paul government ofcials overruled him and enacted
a closing order for the whole city, including schools,
theaters, churches, and dance halls. The St. Paul Citi-
zens’ Committee—consisting of 15 physicians, church
leaders, and community members who were appointed
by Dr. Simon—which was concerned by the record of
218 new cases on November 5, as well as 36 deaths
between November 4 and November 5, 1918, recom-
mended this policy change (Figure 1).17 The number
of new cases began to decline 10 days later, with only
24 new cases, and the next day, Dr. Simon reopened
St. Paul businesses and churches.
Minneapolis and St. Paul both attempted to combat
inuenza by limiting crowding in places with restricted
access to fresh air. Both cities enacted streetcar regula-
tions aimed to keep the air in the streetcars fresh by
mandating open windows and limiting the number
of passengers to 84 (streetcars had a seating capacity
of 46).5,17,18 Because the measure limiting the number
of car passengers, implemented on October 26 in St.
P H C 805
P H R / N–D 2007 / V 122
Paul, was deemed successful, Minneapolis enacted a
similar regulation on October 30.17 As an experiment,
Dr. Bracken also proposed that St. Paul regulate the
business hours of stores and theaters to keep streetcar
congestion to a minimum. Once again, Minneapolis
followed St. Paul’s example on October 16, 1918, by
regulating the hours of retail stores, ofce buildings,
and wholesale stores.19
There were several complaints that the mandate in
Minneapolis to keep three streetcar windows open at
all times caused people to get sick due to winter colds.
A compromise was reached by Dr. Guilford allowing
streetcars with heating and ventilation systems to close
their windows once the temperature dropped to 32
degrees Fahrenheit.20
St. Paul also targeted elevators as places where
inuenza could easily be transmitted due to the tight
quarters and limited fresh air. Buildings with fewer
than six stories were no longer permitted to use their
elevators.21
Public response to community disease containment
The measures used to contain inuenza greatly affected
the day-to-day lives of citizens. While some accepted
the changes imposed on them, others protested regula-
tions that they considered unfair. Some called for more
stringent methods, while others blatantly broke the new
rules that were intended to protect them.
The closing of public places in Minneapolis was
announced in advance, so people rushed to complete
those activities that would soon be banned, resulting
in the very same crowded conditions the ban sought
to prevent. “Downtown theaters were packed last night
with patrons who took advantage of their last chance
to see a performance until the ban is lifted.”22 While
some St. Paul citizens were relieved that Dr. Simon
initially pledged to keep public places open, others
felt this was wrong. “Fear of inuenza contagion in
crowded places has reduced the patronage of St. Paul
motion picture theaters by nearly half, according to
reports to Dr. H.M. Bracken.”23
Many sporting organizations responded negatively
to closing orders. For example, in November 1918, the
bowlers of St. Paul drew up a petition that requested
permission to begin bowling again.24 Minneapolis
football teams chose to ignore the ban and attempted
to play against each other in front of large crowds.
Police were called in to disperse the crowds and halt
Figure 1. Influenza cases in St. Paul as recorded by the St. Paul Health Department
in the St. Paul Daily News, 1918–1919a
aCases were not uniformly reported on Sundays, so Monday’s data may be inflated.
806 P H C
P H R / N–D 2007 / V 122
the games.25 Minneapolis teams found a way to play
despite the closing order. Because Minneapolis high
school football games were banned, practice games
were scheduled with St. Paul teams.26 Several estab-
lishments serving alcohol and food deliberately broke
the closing order to continue their regular business.
“One saloon was discovered with the back door route
open.”27
The elevator regulations in St. Paul were particularly
unpopular. “Some of the downtown hotels objected to
stopping their elevators, saying that they would lose
guests. This caused a change in the ruling to permit
hotel elevators and those in apartment houses to oper-
ate.”28 Many insisted it was unhealthy for the sick to
be forced to climb stairs in their impaired state, while
others felt concerned that people would be shut off
from fresh air if they were not allowed to use their
elevators. Consequently, the city compromised and
all elevators were back in use starting November 9,
1918, although only one person per 5 square feet was
permitted.29
The Hennepin County School Board (where Min-
neapolis is located) was exceptionally deant to the
closing order. The school board was concerned for the
health of the students as well as the “12,000 dollars a
day” that the closing orders cost because teachers con-
tinued to be paid, and extra school days would have
to be added to the school year.30 Against the explicit
orders of Dr. Guilford, and the pleading of several
Parent-Teacher Association ofcers, the school board
reopened schools on October 21, only to be shut down
on the same day under threat of police action.31
In St. Paul, all inuenza cases were supposed to
be reported to a physician, who in turn was required
to isolate the case in his or her own home and notify
the health department. Several problems sprung up
with these requirements that hampered surveillance,
the care of patients, and protecting people from get-
ting sick. For one, both physicians and patients were
often hesitant to bring attention to cases. “Physicians
are not reporting their cases to prevent homes from
being quarantined.”21 (Note: At the time of the 1918
inuenza pandemic, the separation of the ill from the
general population, what is now referred to as isola-
tion, was termed “quarantine.”) The ill also sought to
evade isolation in their homes by not seeking medical
attention, or only seeking medical attention when they
became gravely ill. “Hundreds of persons in the city do
not call for medical assistance until the second, third,
or fourth day and in many cases pneumonia already
has developed when medical attention is rst given.”29
Stafng shortages made isolation even less desirable.
Because there were a limited number of inspectors to
release houses from isolation, houses were not released
promptly from isolation.32
Starting on November 15, St. Paul telephone opera-
tors went on strike. According to the Pioneer Press daily
newspaper, “Less than one third the new cases [are]
being reported to the health department,” as a result
of the telephone strike.33 This strike not only affected
the reporting of cases, but also isolation, as well as their
release from such a measure.
After all of the difculties involved in establishing
isolation for each case, some agrantly disobeyed
the isolation orders altogether. “Disregard of the city
quarantine yesterday caused the arrest of one man who
insisted on taking his child from the city hospital before
the patient was ready to be discharged. The mother
and father and the child later were found mingling
with other persons in the neighborhood.”29
InfeCtIon ControL and VaCCInatIon
In addition to closing public places and isolating
cases in their homes, both Minneapolis and St. Paul
health departments took other steps to keep people
from getting infected. The use of gauze masks, more
stringent sanitation laws, and vaccination campaigns
were deployed in this effort.
Directions for wearing the masks were issued to the
public. “The outside of a face mask is marked with a
black thread woven into it. Always wear this side away
from the face. Wear the mask to cover the nose and
the mouth, tying two tapes around the head above the
ears. Tie the other tapes rather tightly around the neck.
Never wear the mask of another person. When the mask
is removed . . . it should be carefully folded with the
inside folded in, immediately boiled and disinfected.
When the mask is removed by one seeking to protect
himself from the inuenza it should be folded with
the inside folded out and boiled ten minutes. Persons
considerably exposed to the disease should boil their
masks at least once a day.”21 However, there was incon-
sistent advice on the use of gauze masks. Dr. Bracken,
of the State Board of Health, advocated the wearing
of masks, though he did not wear one himself, saying,
“I personally prefer to take my chances.”34
Medical students working in clinics in each district
of St. Paul distributed gauze masks.12 But the Citizens’
Committee rejected an ordinance requiring the wear-
ing of masks at all times, even though, “All physicians
were united in the opinion that the gauze covering
should be worn in hospitals or in the presence of
doubtful cases.”35 Despite the lack of ofcial orders
requiring the wearing of masks and Dr. Bracken’s
unclear message, many people sought out masks for
P H C 807
P H R / N–D 2007 / V 122
themselves. The Northern Division of the American
Red Cross manufactured tens of thousands of masks.
Minneapolis ordered 15,000 masks from the Red Cross
on October 1, 1918.36 These masks were used by nurses
in schools and hospitals, doctors, hospital visitors, and
those suspected of being infected with inuenza.37
As the number of cases increased in St. Paul,
employers sought ways to keep their workers healthy
and productive. Several companies requested masks
to distribute to their workers. Despite the thousands
of masks provided by the Red Cross, still more were
needed to fulll the demand. The Citizens’ Committee
suggested that companies ask their female employees
to fabricate masks for all their employees.21 St. Paul
introduced new sanitation laws that called for the
sterilization of dishes and cups in restaurants and bars,
and the banning of roller towels and common drinking
cups in public restrooms.38
At least two different vaccines were administered
in Minneapolis-St. Paul, neither of them effective as
neither actually contained inuenza virus. One made
by bacteriologists at the University of Minnesota was
purported to prevent pneumonia.39 The Mayo Clinic in
Rochester, Minnesota, made another vaccine that was
intended to prevent both pneumonia and inuenza.40
This latter vaccination was composed of Streptococcus
pneumoniae types I, II, and III, S. pneumoniae group
IV, hemolytic streptococci, Staphylococcus aureus, and
“inuenza bacillus.”41
Military personnel as well as civilians were inoculated
beginning as early as October 4, 1918.37 Both city health
departments purchased vaccine and distributed it to
physicians at no charge to encourage widespread use.
In Minneapolis, people desiring the vaccine “thronged”
the ofces of doctors hoping to be vaccinated, and
in St. Paul it was reported that “thousands of persons
have been inoculated.”39,42 Some physicians took advan-
tage of their access to vaccine and the public’s fear of
inuenza. According to St. Paul’s Citizens’ Committee,
it was discovered that “a few physicians were charg-
ing a fat fee for inoculations.”29 This was particularly
disturbing as the vaccinations were supplied to the
physicians for free.
CommunICatIons
Postal workers, Boy Scouts, and teachers were enlisted
to provide educational materials to the public and to
teach health precautions. Mail carriers distributed
educational materials on their routes. Boy Scouts
distributed posters to stores, ofces, and factories in
downtown Minneapolis.22 Minneapolis teachers who
were put out of work by the closing of schools were
asked to volunteer for a health education campaign.
The main goals of the campaign were to get rid of
shared drinking cups, which were the precursor of the
water fountain, as well as the roller towels, which were
used to dry hands after washing.43 St. Paul teachers were
sent “to ascertain the plight of families worst affected
by the epidemic.”28 This was accomplished through a
canvas of homes where the teachers learned if anyone
was sick, needed to see a physician, or needed food.27
St. Paul set up a public kitchen, a children’s home,
and an emergency hospital for these cases.21
Limitations
Although the two cities chose different methods of
disease containment, determining which method was
more successful is challenging. Information on cases
in both cities depended on ill individuals seeking the
attention of physicians, who were in short supply. The
physicians were then required to report the number
of new cases each day to their city health department.
The city then reported the total number of cases to
the newspapers, which published the number of new
cases and deaths each day. This chain of information
left much room for error and possible falsication.
Because St. Paul chose to utilize isolation and Min-
neapolis did not, case reporting varied greatly between
the two cities. Individuals with inuenza who had their
status reported in St. Paul had to endure isolation
until they were released with a physician’s approval.
This may have discouraged people from seeking the
attention of physicians, and thus being reported—an
undesired consequence of enforced isolation (Table).
Because those with inuenza were not isolated in Min-
neapolis, more people might have felt comfortable
seeking medical attention. This could explain why St.
Paul had such a high case fatality rate compared with
Minneapolis (Table, Figures 2 and 3).
ConCLusIon
Several factors impede direct comparisons of the two
cities’ approaches. The cities border each other and
Table. Minneapolis and St. Paul influenza cases and
deaths, September 30, 1918, to January 6, 1919
Minneapolis St. Paul
Total deaths 747 645
Total cases 14,411 4,399
Death rate (per 100,000) 264 300
Incidence rate (per 100,000) 4,781 2,049
Case fatality rate (percent) 5.2 14.7
808 P H C
P H R / N–D 2007 / V 122
residents travel back and forth. Although the contain-
ment philosophies differed greatly, in reality St. Paul
government ofcials overruled public health, and
schools and public gathering places were closed in
both cities for varying lengths of time. Although the
effects of isolation vs. closure of public places cannot
be specically determined, other lessons can be learned
from what happened in 1918. Many steps could have
been taken to prevent illness and save lives. Prior plan-
ning, clear orders, as well as consistent and transparent
advice and information to the public may have made
a signicant difference in the number of cases and
deaths due to inuenza in 1918.
There was a paucity of planning for a health emer-
gency when inuenza rst appeared. While the actions
that the two city health departments took to stem the
spread of inuenza align closely with current pandemic
plans, health ofcials had the disadvantage of trying
to conceive and realize plans during a health crisis.
Many current recommendations were implemented,
including the use of masks, the use of vaccines (albeit
ineffective ones), increasing the stringency of sanita-
tion measures, limiting crowding in public places,
and trying to coordinate hospitals, nurses, physicians,
and medical students to maximize resources. As part
of maximizing human resources during an inuenza
pandemic, it is imperative that the safety of health-care
workers is insured. The number of nurses and physi-
cians who fell ill and even died as a result of assisting
in the ght against the pandemic scared other nurses
and physicians away.
Had these ideas been generated prior to such a
large emergency, several problems could have been
averted. The debates and disagreements between dif-
ferent public ofcials and health agencies, as with the
Hennepin County School Board and the Minneapolis
Health Department or between the Minneapolis Health
Department and the St. Paul Health Department, could
have been discussed in advance. Supplies could have
been stockpiled, business leaders and community mem-
bers could have provided input on controversial disease
containment policies, and medical students could have
been put to work in hospitals and communities that
lacked physicians. Unfortunately, these disputes arose
and continued throughout the pandemic.
Clear authority and management by public health
Figure 2. Influenza case rates per 100,000, Minneapolis and St. Paul, 1918–1919a
aCases were not uniformly reported on Sundays, so Monday’s data may be inflated.
P H C 809
P H R / N–D 2007 / V 122
ofcials were generally lacking at the federal and state
levels. It was almost as if the fear of using their author-
ity led Surgeon General Blue and Dr. Bracken to fail
to take decisive action. Surgeon General Blue suggested
to Dr. Bracken, and all other state health ofcials, “the
advisability [of] discontinuing all public meetings,
closing all schools and places of public amusement…
on appearance of local outbreaks.”44 Because this was
merely a suggestion, and local outbreaks were not
dened objectively, Blue’s urgent telegram had no
effect.
On the state level, Dr. Bracken acknowledged that
the St. Paul Health Department “followed his advice” to
not close public places, and went on to say that St. Paul,
“has the power to do the opposite any time it wants
to.”11 This statement forced local health departments
to dene their own rules while attempting to decipher
conicting messages from the state and federal level.
Because clear orders were not being given to public
health ofcials, the public in turn was not receiving
transparent and consistent advice and information.
Should the public wear masks? Why was it allowable
to be next to someone in a streetcar and not in an
elevator? Why were church services closed while Red
Cross workers gathered in crowded conditions in those
very same churches? Was inuenza a life-threatening
condition, or was panic the most dangerous element of
the inuenza pandemic? In Minneapolis and St. Paul.
there was no single message on any of these issues.
In many cases, the public had to decide for itself. In
which case, the effect of the messages that were com-
municated only served to contradict each other.
In reviewing this history, some lessons stand out.
Recent analyses of nonpharmaceutical interventions
during 1918 indicate cities in which multiple interven-
tions were implemented early in the pandemic fared
better.45 Of primary importance is developing a plan
ahead of time that incorporates all levels of govern-
ment health infrastructure and describes clear lines
of responsibilities and roles. Plans for surge capacity
and community containment must be discussed with
stakeholders and consensus must be achieved.
Further, general approaches should be put forth
for public comment and approval. The public health
benet of isolation should be weighed against the pos-
sibility that some people would be discouraged from
Figure 3. Daily death rates per 100,000, Minneapolis and St. Paul, 1918–1919a
aCases were not uniformly reported on Sundays, so Monday’s data may be inflated.
810 P H C
P H R / N–D 2007 / V 122
seeking care. Clear explanations of the reason for isola-
tion, generous employer support, and providing food,
medicine, and social service to those in isolation may
mitigate fears and increase cooperation. The public
must also be educated about the reasoning behind
other health measures (i.e., closures), should those
methods be implemented.
Approaches and plans should be based on scien-
tic data whenever possible, and include input from
ethicists. Unlike in 1918, a pandemic inuenza vaccine
will likely be available today, albeit four to six months
after the pandemic starts. But similar to 1918, the
challenge will be designing an orderly and ethical
distribution of a scarce commodity. Further, experts in
risk communication should assist in developing mes-
sages that are scientically accurate, understandable,
clear, and useful. Finally, we need to take careful note
of local and national lessons from the past so we do
not repeat them.
Miles Ott is a Public Health Graduate Student Worker, Shelly F.
Shaw is an Epidemiologist, Richard N. Danila is a Deputy State
Epidemiologist, and Ruth Lyneld is a Medical Director and State
Epidemiologist. All are with the Minnesota Department of Health
Infectious Disease Epidemiology, Prevention and Control Division,
in St. Paul, Minnesota.
Address correspondence to: Richard N. Danila, PhD, MPH,
Minnesota Department of Health Infectious Disease Epidemiol-
ogy, Prevention and Control Division, Acute Disease Investigation
and Control Section, 625 Robert St. N, P.O. Box 64975, St. Paul,
MN 55164-0975; tel. 651-201-5414; fax 651-201-5743; e-mail
<richard.danila@health.state.mn.us>.
referenCes
1. Barry JM. The great inuenza: the epic story of the deadliest plague
in history. New York: Penguin; 2004.
2. State Board of Health and Vital Statistics of Minnesota. Eighth
biennial report 1918–1919. Minneapolis: State Board of Health
and Vital Statistics of the State of Minnesota; 1920.
3. Spanish Inuenza gains headway here. Minneapolis Tribune 1918
Oct 1; 8.
4. Inuenza gains slowly in city. Minneapolis Tribune 1918 Oct 10;
1,8.
5. Inuenza gains among civilians. Minneapolis Tribune 1918 Oct 9;
1,7.
6. Editorial. Minneapolis Tribune 1918 Oct 3; 16.
7. Sailors may attend inuenza patients. Minneapolis Tribune 1918
Nov 3; 4.
8. Henry Bracken to Rupert Blue, 1918 Nov 13, Minnesota Department
of Health Correspondence and Miscellaneous Records, 1895–1954,
Minnesota Historical Society.
9. Henry Bracken to Rupert Blue, 1918 Nov 2, Minnesota Department
of Health Correspondence and Miscellaneous Records, 1895–1954,
Minnesota Historical Society.
10. Henry Bracken to Dr. Merritte W. Ireland, 1918 Nov 16, Minnesota
Department of Health Correspondence and Miscellaneous Records,
1895–1954, Minnesota Historical Society.
11. Business hours may be changed to curb epidemic. Minneapolis
Tribune 1918 Oct 15; 1,10.
12. Lid on tomorrow includes schools. St. Paul Pioneer Press 1918 Nov
5; 6.
13. 150 cases of inuenza in Minneapolis. Minneapolis Tribune 1918
Sep 30; 1.
14. Mill City closed. St. Paul Pioneer Press 1918 Oct 12; 1,6.
15. Doctors propose drastic lid be clamped on city. Minneapolis Tribune
1918 Oct 11; 1,2.
16. Clash over school order due Monday. Minneapolis Tribune 1918
Oct 20; 1.
17. Push grip ght. St. Paul Pioneer Press 1918 Oct 30; 1.
18. Epidemic controlled in city’s army camps car crowds, 558 pupils
stricken. Minneapolis Tribune 1918 Oct 29; 10.
19. Trade hours set to stem Spanish Inuenza here. Minneapolis Tri-
bune 1918 Oct 16; 1,2.
20. Ban on until deaths decrease to 7 a day. Minneapolis Tribune 1918
Nov 10; 14.
21. Sweeping order against inuenza in effect here today. St. Paul
Pioneer Press 1918 Nov 6; 1,7.
22. Inuenza lid clamped tight all over city. Minneapolis Tribune 1918
Oct 13; 1,10.
23. See less inuenza. St. Paul Pioneer Press 1918 Oct 15; 8.
24. Pins will fall. St. Paul Pioneer Press 1918 Nov 17; 7.
25. “Flu” order stops park grid games. Minneapolis Tribune 1918 Oct
14; 18,12.
26. High school games neither on nor off. Minneapolis Tribune 1918
Oct 15; 18,12.
27. Fail to report grip. St. Paul Pioneer Press 1918 Nov 8; 10.
28. Plan survey of inuenza cases. St. Paul Pioneer Press 1918 Nov 7;
1.
29. All lifts to run. St. Paul Pioneer Press 1918 Nov 9; 1,7.
30. School chiefs face arrest or injunction, city ofcials to use law as
directors defy inuenza ban. Minneapolis Tribune 1918 Oct 21;
1,2.
31. Guilford wins ght to keep schools shut. Minneapolis Tribune 1918
Oct 22; 1.
32. Shows open today. St. Paul Pioneer Press 1918 Nov 15; 1.
33. Inuenza relief work disrupted as a result of telephone strike. St.
Paul Pioneer Press 1918 Nov 16; 1.
34. Inuenza lid to go on city today. St. Paul Pioneer Press 1918 Nov
4; 1,3.
35. Speed grip ght. St. Paul Pioneer Press 1918 Nov 2; 1.
36. Inuenza spread held slight here. Minneapolis Tribune 1918 Oct
2; 1,4.
37. Inuenza halts “U” opening to all but S.A.T.C. Minneapolis Tribune
1918 Oct 5; 1,22.
38. Cafes and bars hit by grip ban. St. Paul Pioneer Press 1918 Dec 14;
1.
39. Serum to be issued. St. Paul Pioneer Press 1918 Oct 19; 6.
40. Epidemic statistics show decline in city. Minneapolis Tribune 1918
Oct 25; 15.
41. Rosenow EC. Prophylactic inoculation against respiratory infections
during the present pandemic of inuenza. Preliminary report.
JAMA 1919;72:31-4.
42. Epidemic brings mercy problems. St. Paul Pioneer Press 1918 Nov
17; 1.
43. Teachers released for school closing period. Minneapolis Tribune
1918 Oct 15; 11.
44. Telegram from Rupert Blue to Dr. Henry Bracken, 1918 Oct 6,
6:49 pm, Minnesota Department of Health Correspondence and
Miscellaneous Records, 1895–1954, Minnesota Historical Society.
45. Hatchett RJ, Mercher CE, Lipsitch M. Public health interventions
and epidemic intensity during the 1918 inuenza pandemic. Proc
Natl Acad Sci 2007;104:7582-7.