from http://bmj.com/cgi/content/full/317/7167/1233 - The British Medical Journal
BMJ 1998;317:1233-1236 ( 31 October )
History BMJ 1998;317:1233-1236 ( 31 October )
Education and debate
"A calculated risk": the Salk polio vaccine field
trials of 1954
Marcia Meldrum, DeWitt Stetten memorial fellow in the history of the biomedical
sciences.
National Institutes of Health, Bethesda, MD 20892 The polio vaccine field trials
of 1954, sponsored by the National Foundation for Infantile Paralysis (March
of Dimes), are among the largest and most publicised clinical trials ever undertaken.
Across the United States, 623 972 schoolchildren were injected with vaccine
or placebo, and more than a million others participated as "observed"
controls. The results, announced in 1955, showed good statistical evidence
that Jonas Salk's killed virus preparation was 80-90% effective in preventing
paralytic poliomyelitis.1
The statistical design used in this great experiment was singular, prompting
criticism at the time and since. Eighty four test areas in 11 states used
the textbook model: in a randomised, blinded design all participating children
in the first three grades of school (ages 6-9) received injections of either
vaccine or placebo and were observed for evidence of the disease. But 127 test
areas in 33 states used an "observed control" design: participating
children in the second grade (ages 7-8) received injections of vaccine; no placebo
was given, and children in all three grades were then observed for the duration
of the polio "season."1
The use of the dual protocol illustrates both the power and the limitations
of the randomised clinical trial to legitimate therapeutic claims. The placebo
controlled trials were necessary to define the Salk vaccineintroduced by a lay
organisation that has taken an activist position against the counsel of its
virological advisersas the product of scientific medicine. The observed control
trials were essential to maintaining public support for the vaccine as the product
of lay faith and investment in science. Here I examine the process by which
the trial design was negotiated and the roles of the several actors.
Summary points ------------------------------------------------------------------------
The 1954 polio vaccine field trials used a singular statistical design
------------------------------------------------------------------------
Over 600 000 schoolchildren were injected with vaccine or placebo
and over a million others participated as "observed" controls
------------------------------------------------------------------------
This dual protocol illustrates both the power and the limitations of randomised
clinical trials to legitimate therapeutic claims A problematic
vaccine
On 23 January 1953, Jonas Salk of Pittsburgh presented the results
of his tests of a "killed virus" polio vaccine on 161 children
to the Immunization Committee, a scientific advisory committee to the National
Foundation for Infantile Paralysis. 2 3 The foundation, created in 1938 by
President Roosevelt and his law partner, Basil O'Connor, was a lay governed
organisation based on grassroots fundraising and volunteer effort. For 15 years
a portion of the dimes and dollars collected in the annual "Mothers' March"
had been devoted to research: epidemiological studies of poliomyelitis, identification
and classification of the three strains of the virus, development of practical
culture methods. These projects had strong support among scientists, but for
the foundation's staff and volunteers they were necessary stepping stones to
the development of an effective vaccine.4 Salk's work seemed promising to O'Connor,
and to Thomas Rivers, the dean of the foundation's scientific advisers. The
children had shown no ill effects and the levels of polio antibodies in their
blood had risen. Almost immediately, O'Connor and Rivers began planning for
a major field trial. 4 5
Several of the senior virologists on the Immunization Committee, notably the
Nobel laureate John Enders of Harvard and Albert Sabin of Cincinnati, thought
these plans precipitate. They questioned the relation of antibodies to permanent
immunity and doubted the safety of a vaccine prepared from virulent poliovirus,
whatever "inactivation" method was used. Enders described Salk's work
as "most encouraging" but cautioned that "the ideal immunizing
agent against any virus infection should consist of a living agent exhibiting
a degree of virulence so low that it may be inoculated without risk"6that
is, an attenuated strain that would create immunity by producing a subclinical
case of the real disease, as in the classic cowpox/smallpox model. 3 5
Despite these objections, O'Connor believed that his organisation had a mandate
from its volunteers and donors to proceed. 3 5 As Harry Weaver, the foundation's
director of research, wrote: "The practice of medicine is based on calculated
risk .... If [we wait until more] research is carried out, large numbers
of human beings will develop poliomyelitis who might have been prevented from
doing so."4
The virologists' critique was only one obstacle to the field trial. Since paralytic
polio was a disease of relatively low incidence, the experimental population
would consist of school age children, the group with the highest case rate;
the foundation decided to target the first three grades in the 272 counties
with the highest incidence of the disease. Volunteers from the foundation would
work through state and local health departments and schools to gain parental
consent and deliver the children for injection. 7 8 The use of a placebo control
group seemed to be too much of a "calculated risk," one that parents,
teachers, and health officials would reject; in Salk's words, "a `beautiful'
... experiment over which the epidemiologist could become quite ecstatic
but [which] would make the humanitarian shudder."9
The foundation enlists support
On 9 November 1953, O'Connor announced that the field trials would
begin in the spring and that an "observed control" plan would be used,
in which one group of children would receive vaccine, while others in the same
age group would not be injected but only observed.10 Hart Van Riper, the foundation's
medical director, asked the nation's health officers for advice and support.11
Carrying the imprimatur of medical expertise, yet necessarily responsive to
public fears, the health departments constituted a potential counterweight to
the virology community.
Within a month, departments in 38 states had responded, most enthusiastic
about the prospect of a vaccine and ready to use the observed control plan.
A number of state officials, however, saw it as a problem that the project was
sponsored not by scientists but by a lay organisation. They questioned the impartiality
of an evaluation run by the foundation and the rigour of the proposed design.
3 12
To meet these objections, and those of the doubting virologists, O'Connor and
Van Riper asked Thomas Francis of the University of Michigan to direct an independent
evaluation of the trials, supported by funds from the foundation, but otherwise
autonomous. 3 4 13 14 Francis, a highly respected virologist who had conducted
field trials of influenza vaccines, was supportive of the killed virus preparation.
"I think I shall do it," Francis admitted in a letter on 29 December15;
but before taking the job, he mobilised support among the state health officers
to engineer a change in the trial design.
"Polio pioneers"some of the many children who took part in trials
of poliomyelitis vaccine
CREDIT: CORBIS Two types of controls
In a public statement on 8 January 1954, the foundation still adhered
to the observed control plan; but on 15 February, six days after Francis
was formally appointed to head the evaluation, O'Connor announced that two types
of controls would be used in the field trials: "observed controls"
in 34 states and "placebo controls" in 11: "a combination
of the two procedures [will] assure a valid evaluation of the trial vaccine."
16 17 This change in plans was the result of a month of manoeuvring on Francis's
part.
He had requested an "advisory group" meeting on 11 January. This
new group was entirely distinct from the foundation's scientific advisory committee,
which was excluded from these deliberations. As well as the senior staff of
the foundation, a selected list of state health officers, paediatricians, clinical
polio specialists, statisticians, and virologists attended. Their charge was
not to debate the merits of Salk's work but to take the vaccine project from
the laboratory into the field. Part of the January group later became an advisory
committee for the field trial evaluation, and the state health officers constituted
a separate body to advise on "technical aspects" of the project. 1
18 19 Because the health officers were divided, Francis's role was critical.
The 11 January meeting began with briefings from the foundation's staff
on plans to date. Rivers assured the group that the foundation would do its
best to guarantee the safety of the vaccine.18 After general discussion, the
group subdivided for the afternoon into three groups designated as "clinicians,"
"statisticians," and "health officers." Though each of the
groups made several recommendations, I will focus here only on their statements
regarding trial design. The clinicians' report assumed the use of observed controls,
and the statisticians' group, unsurprisingly, recommended the use of "a
blind injected control" wherever the "proper facilities" made
such a design possible.18
Francis himself joined the "health officers" group. He listed in his
notes several health departments that would support an injected, or placebo,
control design: Massachusetts, New York, Michigan, Ohio, Illinois, California.
Each was a populous state with a well organised health department headed by
a nationally respected physician. Perhaps, he mused, a "double study"
could be done in these states: placebo controls in the second grade, observed
controls in the first and third grades.20
Rewriting the design
When the "health officers" met in the afternoon of 11 January,
Francis found the group willing to endorse an even broader design. The participants
included Francis, health officers from California, Illinois, New York, and Massachusetts,
and two friendly virologists. Their report began emphatically: "It was
the consensus of the group that [placebo controlled] studies were necessary
... that rather than limit the controlled study to the second grade it
would be better to take the first three grades of school and select individuals
... on an alternate basis."17
The Health Officers' Advisory Committee which met in Atlanta at the end of January
was a select group of doctors from eight states who were supportive of placebo
control. Francis told the group that he had decided to accept the job of directing
the evaluation "with the understanding that a number of the states have
indicated that they would like to, and would be able to, carry on injected [placebo]
control studies." The majority of the states, 36, preferred to adhere
to the observed control design. Francis suggested that if a shortage of vaccine
developed (which seemed quite likely at that point) supplies should be reserved
for the placebo control areas; the group agreed with a formal recommendation
that those areas be given "priority on available vaccine." Someone
asked whether the placebo control plan would make it more difficult to obtain
the parents' consent. The group decided that it could rely on the widespread
fear of the disease; members agreed that "it would not be difficult to
sell as there is a high attack rate in the three grades [and] there would still
be a 50% chance of a child receiving the vaccine."19
In Francis's mind, the placebo control study was now his primary interest, and
he reiterated this point in the summary report. Indeed, he seems to have stage
managed the January meetings to reorient the project in that direction, selecting
likely allies among the health officers and using their support to rewrite the
trial design. "The best Departments are committed to this [placebo control]
plan," Francis told Van Riper. "The assurance and faith of those committed
must be maintained."21 Although it might be necessary to exploit parents'
fears to obtain their consent and to allow large numbers of children to face
the polio season without protection, the use of a randomised and blinded controlled
trial would effectively counter the criticisms of scientists such as Enders
and Sabin, legitimise the sponsorship of the lay governed foundation, and gain
the support of the leaders of the medical community, exemplified by the nation's
leading state health officers.
A national event
But the observed control trials were not a sideshow to the main event, an unnecessary
"deviation" from good methodology. Thirty six health departments,
representing a large segment of public opinion and the rank and file of the
medical profession, were committed to that plan and their participation was
necessary to the field trial. If the Salk vaccine trials were to succeed, it
was essential that they be a great national event, enlisting volunteers, doctors,
and parents in one united effort that represented the culmination of 15 years
of work and faith. Given the climate of scientific doubt that surrounded the
killed-virus vaccine, it was essential that the field trials offer public, as
well as scientific, validation of its effectiveness.
The National Foundation for Infantile Paralysis had tried to reconcile its scientific
and political problems by working through the state health officers, but this
groupeach official facing the conflicting demands of professional training and
public constituencywas itself divided. O'Connor then enlisted Francis and his
impressive credentials, who, rather than pacify the advocates of placebo controls,
chose to ally himself with them. The ensuing negotiations shaped a dual statistical
design that reflected the multiple meanings of the trial: as scientific demonstration,
political statement, and mass participation event.
Acknowledgments
Competing interests: MM has presented papers at meetings sponsored in part by
pharmaceutical companies.
Funding: The research for this paper was supported in part by a grant from the
Agency for Health Care Policy and Research of the US Public Health Service.
References
1. Francis Jr T, Korns R, Voight R, Boisen M, Hemphill F, Napier J, et al. An
evaluation of the 1954 poliomyelitis vaccine trials: summary report. Am
J Public Health 1955; 45(suppl): 1-50.
2. Minutes of the Committee on Immunization, Hershey, Pennsylvania, January
23, 1953. Jonas Salk Papers, Mandeville Special Collections, University
of California San Diego, Box 254, Folder 2, page 157.
3. Smith J. Patenting the sun: polio and the Salk vaccine. New York: Morrow
, 1990.
4. Benison S. Tom Rivers: reflections on a life in medicine and science (an
oral history memoir). Cambridge, MA: MIT Press , 1967.
5. Meldrum M. Departures from the design: the randomized clinical trial in historical
perspective, 1946-1970 [dissertation]. Stony Brook: State University of
New York , 1994.
6. Enders J. Recent advances in the study of poliomyelitis. Medicine 1954; 33:
87-95.
7. Thomas Dublin to Hart Van Riper. Predicting poliomyelitis incidence for the
1954 field trial. Thomas Francis Papers, Bentley Historical Library, University
of Michigan, (hereafter TF-BLUM), Box 21, Folder VaccineSelection of Counties.
8. Hart Van Riper to NFIP Staff. Brief background statement for the vaccine
field trial, December 1, 1953. TF-BLUM, Box 18, Folder NFIP1954.
9. Carter R. Breakthrough: the saga of Jonas Salk. New York: Trident , 1966.
10. New tests on polio to dwarf old ones. New York Times 1953 November
10:32.
11. Hart Van Riper to Carl N Neupert, November 19, 1953. TF-BLUM,
Box 18, Folder NFIPVan Riper.
12. Thomas Dublin to Hart Van Riper, Response from state health officers regarding
selection of field trial areas, December 9, 1953. TF-BLUM, Folder
NFIPMemos.
13. Stella Barlow to Hart Van Riper, November 16, 1953. TF-BLUM, Box
6, Folder National FoundationVan Riper.
14. Hart Van Riper to Thomas Francis, January 11, 1954. TF-BLUM, Box
6, Folder National FoundationVan Riper.
15. Thomas Francis to Harry Weaver, December 29, 1953. TF-BLUM, Box
6, Folder National FoundationWeaver.
16. Named to direct study on polio vaccine tests. New York Times 1954 February
10:16.
17. School tests set for polio vaccine. New York Times 1954 February 15:25.
18. Minutes of the Meeting of Advisory Group on Evaluation of Vaccine Field
Trials, January 11, 1954. TF-BLUM, Box 18, Folder MeetingNew
York, January 11, 1954.
19. Minutes of the Advisory Committee on Technical Aspects of the Poliomyelitis
Field Trials, January 30-31, 1954. TF-BLUM, Box 18, Folder MeetingAtlanta,
Advisory Committee, January 30-31, 1954.
20. Thomas Francis, handwritten notes. TF-BLUM, Box 18, Folder MeetingNew
York, January 11, 1954.
21. For discussion with Van Riper. Thomas Francis' typed notes, nd. TF-BLUM,
Box 18, Folder MeetingDetroit, February 23-24, 1954.
(Accepted 6 October 1998)
------------------------------------------------------------------------
© BMJ 1998
This article has been cited by other articles:
S. Yusuf
Randomised controlled trials in cardiovascular medicine: past achievements,
future challenges
BMJ, August 28, 1999; 319(7209): 564 - 568.
[Full Text]
© 2003 BMJ Publishing Group Ltd
FAIR USE NOTICE
This site may contain copyrighted material the use of which has not always
been specifically authorized by the copyright owner - or on which copyright
has expired. We are making such material available in our efforts to advance
understanding of environmental, political, human rights, health, economic, democracy,
scientific, and social justice issues, etc. We believe this constitutes a 'fair
use' of any such copyrighted material as provided for in section 107 of the
US Copyright Law. In accordance with Title 17 U.S.C. Section 107, the material
on this site is distributed without profit to those who have expressed a prior
interest in receiving the included information for research and educational
purposes. For more information go to: http://www.law.cornell.edu/uscode/17/107.shtml.
If you wish to use copyrighted material from this site for purposes of your
own that go beyond 'fair use', you must obtain permission from the copyright
owner.