Sei sulla pagina 1di 9

Abe and Yak: The Interactions of Abraham M.

Lilienfeld and Jacob Yerushalmy in the

Development of Modern Epidemiology (1945-1973)
Author(s): David Eugene Lilienfeld
Source: Epidemiology, Vol. 18, No. 4 (Jul., 2007), pp. 507-514
Published by: Lippincott Williams & Wilkins
Stable URL:
Accessed: 14-11-2019 11:12 UTC

Linked references are available on JSTOR for this article:
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

Lippincott Williams & Wilkins is collaborating with JSTOR to digitize, preserve and extend
access to Epidemiology

This content downloaded from on Thu, 14 Nov 2019 11:12:27 UTC
All use subject to

Abe and Yak

The Interactions of Abraham M. Lilienfeld and Jacob Yerushalmy in
the Development of Modern Epidemiology (1945-1 973)
David Eugene Lilienfeld

Abstract: This paper explores the relationship between 2 persons

pharmacy in Jerusalem. When Yak completed gymnasium
much involved in the development of modem epidemiology, Jacob
(high school) in 1924, he went abroad to university; no
Yerushalmy and Abraham M. Lilienfeld. The formation of that universities existed in British Mandate Palestine. He started
relationship is described and the resulting influences by each indi in New York City, though the specific university is not known.
vidual on the other's professional work are discussed. Interactions In 1925, he moved to Baltimore to undertake a doctorate in
between these 2 men contributed to several areas of epidemiologic mathematics at Johns Hopkins, completing his work under
development, including the effects of misclassification on observa Oscar Zariski, himself a recent addition to the Hopkins
tional study data, elements of epidemiologic causal inference, pop faculty. Yerushalmy's thesis topic, Construction of Pencils of
ulation-based linked databases, and the adaptation of statistical Equianharmonic Cubic, reflected his interest in topology and
techniques by the epidemiology community. The impact of their algebraic geometry; he successfully defended in 1930.2
association and its implications for modem epidemiologists are Just as Yerushalmy received his doctorate, the United
considered. States entered the Great Depression. Teaching positions were
difficult to find. Yerushalmy made do as a National Research
(Epidemiology 2007;18: 507-514)
Fellow in mathematics at the University of Illinois for the
1930-1931 academic year, and continued as a fellow at
Princeton University the following year. Not much is known
n 1945, Jacob Yerushalmy was a statistician in the United about this period in Yerushalmy's career, save the absence of
States Public Health Service Tuberculosis Control Bureau. skill in poker.3
In that year he introduced his brother-in-law, Abraham Lil In 1932, desperate for work, Yerushalmy taught math
ienfeld, to his colleague, Morton Levin, then at the New York ematics at Johns Hopkins for a year as an Instructor. One of
State Department of Health. During the ensuing conversation, his students was the son of Lowell Reed, Jr., Chair of the
Levin suggested the brother-in-law join the New York De university's biostatistics department. Using this entry to the
partment of Health for a career in public health. After joining elder Reed, Yerushalmy complained bitterly of the lack of
the Department, Lilienfeld began a series of collaborations work available for a Hopkins-trained mathematician. He
with Yerushalmy (or Yak, from his Hebrew name Yaakov) indicated his willingness to beg, cup in hand, at the center of
that would define various aspects of modem epidemiology, downtown Baltimore with a sign, "Johns Hopkins Ph.D. in
including many of the research and teaching foci in the mathematics." Horrified, Reed directed Yerushalmy to con
departnents Lilienfeld and Yerushalmy subsequently directed tact the biostatistics department administrator regarding a
(Lilienfeld at the University of Buffalo and Johns Hopkins position as an Assistant Professor. Thus Yerushalmy turned
University, and Yerushalmy the University of California, Berke his attention to biostatistics.
ley). This paper examines the association of these 2 public health Yerushalmy remained at Johns Hopkins until 1935,
professionals, my father and uncle. though there is little evidence of career advancement. At
Johns Hopkins, Yerushalmy befriended another member of
Yak the biostatistics faculty, Carroll E. Palmer, trained in demog
Jacob Yerushalmy (Fig. 1) was born in Ottoman-con raphy. For a period of almost 2 decades (beginning in the
trolled Ekron, near Jerusalem, in 1904, the youngest of 8 early 1940s), Yerushalmy would work with Palmer (see
children.' His father, William Zev Yerushalmy, owned a below). Also of significance, in 1933 Yerushalmy met and, 1
year later eloped with, Eva Zemil, the eldest daughter of a
From the Department of Health Policy and Research, Stanford University, Baltimore businessman. In 1935, with his career stuck, Yerush
Palo Alto, California. almy accepted a position as a biostatistician for the Department
Editors' note: A commentary on this article appears on page 515. of Health in Albany, New York.
Correspondence: David Eugene Lilienfeld, 399 Port Royal Ave., Foster City,
His billet allowed Yerushalmy to gain confidence in his
CA 94404. E-mail:
Copyright C 2007 by Lippincott Williams & Wilkins biostatistical skills. It also placed him in contact with Dr.
ISSN: 1044-3983/07/1804-0507 Morton L. Levin, the Assistant Director for Cancer Con
DOI: 10.1097/EDE.0b013e318063eea8 trol. The 2 developed a fast friendship, both professional

Epidemiology * Volume 18, Number 4, July 2007 507

This content downloaded from on Thu, 14 Nov 2019 11:12:27 UTC
All use subject to
Lilienfeld Epidemiology * Volume 18, Number 4, July 2007

ultimately lead to the formation of the Kaiser-Pe

database in northern California for the conduct of
logic studies (see below). However, the entry of t
States into World War II changed many of the
government's funding priorities; child development
was not among them, but tuberculosis researc
early 1943, Yerushalmy's boss moved from the C
Bureau to the United States Public Health Servic
culosis Control Bureau; Yerushalmy followed.
From 1943 until his departure 4 years later to e
a biostatistics department at the University of C
Berkeley School of Public Health, Yerushalmy wo
as the chief statistician at the Tuberculosis Control
1945, Time magazine referred to him as the "Pub
Service's top statistician."8 At the Bureau, Yerushalm
2 areas of research, one (the reproducibility of clinica
with which he would be identified for the remainde
career, and the other (criteria for causal inference in
ogy) which became the core of the coming debate on
FIGURE 1. Jacob Yerushalmy ("Yak") in his office. Undated.
smoking and lung cancer He would also develop
skepticism regarding the quality of data used in any
logic query.
and personal, that would last for almost 4 decades. Yerush
In 1943, conventional medical thought held
almy kindled in Levin an interest in biostatistics, and
physician could diagnose pulmonary tuberculosis on
encouraged cancer epidemiology studies sponsored by
of a chest X-ray. As a nonphysician and earning con
Levin throughout the late 1940s and 1950s. (These studies
less than a physician, Yerushalmy was contemp
included the first case-control study on cigarette smoking
physicians and their techniques. On joining the Tub
and lung cancer conducted in the United States).4 In turn,
Control Bureau, Yerushalmy was immediately s
Levin showed Yerushalmy the need for biostatistical input
when told of
into epidemiologic investigations of chronic diseases, physicians' ability to diagnose tuber
X-ray. Based on his NIH experience, Yerushalmy kne
as cancer.
In 1938, another opportunity beckoned to had to be some misdiagnoses. Working with Palmer
almy undertook a series of pioneering studies int
at the National Institutes of Health (NIH). There, Yerushalmy
came to know Harold Dom, a young demographer leading of reading
the chest X-rays. His findings, pub
1947, indicated
First National Cancer Survey. During the 3 years prior to the that physicians were not only unabl
on a diagnosis
attack on Pearl Harbor, Dorn began developing ideas for among themselves, but that a given p
would them
studies following up on survey findings, discussing change the reading of the same X-ray when s
his colleagues. Through such discussions with second
Dorn, or Yerush
third time.9 In doing so, Yerushalmy dev
almy raised the issue of diagnostic certainty. He ofthat
reasoned sensitivity and specificity. The find
lenged the
one often does not truly know if a malignancy was present medical community, which held the p
diagnosis as beyond reproach. These findings als
until it has been shown by biopsy. As a biostatistician,
Yerushalmy considered the distribution of biopsy implications in the coming development o
tions, along with the possibility of errors. domized clinical trial, indicating the need for ce
It was as the NIH that Yerushalmy also laboratories and double-blinded allocation of treatm
began his
lifelong interest in child development. He was tasked to inves also began a series of discussion
Palmer on causal
tigate childhood mortality. Considering the still-incomplete na inference in epidemiology. The d
ture of the national vital statistics system, such not develop out of the Bureau's research progra
but from biostat
must not have been a desirable one for most ambitious Yerushalmy's previous work on childho
ity. They
isticians.5 Yerushalmy persevered, and his reports showedbegan
an with a consideration of what ev
irreverent and thoughtful investigator.6'7 necessary to make a causal inference. These disc
Three years as a statistician at the NIH continued for another decade and a half, and b
working with
development for
child development data opened another opportunity of modern approaches to discerning ca
In 1947,
Yerushalmy. When Palmer left Johns Hopkins to run the Tuberculosis Control Bureau learne
the field
British of
studies section of the United States Department randomized
Labor trial of streptomycin for treatin
Children's Bureau, Yerushalmy left the NIH to work directed
for him,by the statistician Bradford Hill. P
becoming Palmer's Director of the Division Yerushalmy to England to learn about
of Statistical
Research. It was during his tenure at the Department trip consisted of visits with Hill, t
of Labor
that Yerushalmy's interest in child development center staff, and selected clinical cente
This interest would guide his future research, and during
return, it would
a Sunday visit by both Morton Levi

508 ? 2007 Lippincott Williams & Wilkins

This content downloaded from on Thu, 14 Nov 2019 11:12:27 UTC
All use subject to
Epidemiology * Volume 18, Number 4, July 2007 Abe and Yak

Aliyah for Jews

Lilienfeld, Yerushalmy described Hill as a competent to return to the Holy Land. However, under
tician whose data analyses would be solid, but control, the
he noted much of the Palestinian economy was closed
lack of concern regarding the quality oftothe data.
Jewish His
involvement. Joe struggled to support his family.
conclusion: "I wouldn't believe anything After
Hill Sam
graduated from medical school, he returned to the
about those data." United States, followed shortly by his family's return to New
Shortly after his return from England, Yerushalmy
York City. Onwas
their arrival in 1933, Sam informed them of his
approached about an opportunity in California: the University
marriage to a member of a prominent Baltimore family; Sam
of California at Berkeley was opening a school of public
subsequently established his practice in Baltimore.
health and needed someone to organize the biostatistics
For the next 4 years, Abe attended Erasmus High
department. In 1948, Yerushalmy took on the assignment.10
School. He maintained excellent grades in science and math
ematics, but found an interest in history. In the evenings, he
Abe studied with his father-usually the Talmud, as well as
Abraham "Abe" Lilienfeld (Fig. 2) was born in Newmodem philosophers. The Talmudic discussions refined Abe's
York City on November 13, 1920. His father, Joe, came from
tiinking, forcing him to identify the key variables defining a
a well-to-do family in Galicia (western Ukraine). Joe hadsituation. In 1938, with high school graduation looming, the
trained as a rabbi. As his wife refused to cover her hair
Lilienfelds discussed the issue of Abe's college education. Abe
(expected of all married Jewish women), Joe could not find a
wanted to study business administration at City College. Given
pulpit. The family migrated to the United States in 1914 to
the family's chocolate store business, he thought it a useful field.
escape the draft. They expected victory by the Central Pow
His parents, however, focused on medicine. With developments
ers, and accordingly maintained their wealth in German
in Nazi Germany in mind, they reasoned medical skills were
marks. At the conflict's conclusion, the family was destitute.
portable; a business was not. Accordingly, Abe applied to 2
Until 1930, Abe, his parents, and his older brother Sam lived
colleges: City College and Johns Hopkins. Admitted to both, he
on the Lower East Side of Manhattan in New York City, as
packed at the end of the summer of 1938 and moved in with his
Joe eked out a living as a bookbinder.
brother Sam to attend Johns Hopkins.
With the Great Depression, employment opportunities
College opened new vistas to Abe. He took courses in
for bookbinders disappeared. By then, Sam had entered
Middle Eastern studies given by Foxwell Albright (doyen of
medical school in England (quotas against Jewish students
the field), as well as French and German. He enrolled in the
precluded attending a United States school). Lilienfeld's
premedical curriculum, and he found a particular interest in
mother took matters in hand and purchased a run-down
chocolate store with a soda fountain. Within a year, thebiophysics. In a bid to minimize the financial drain on his
family, he increased his course load and graduated in 3 years.
family had made the shop profitable, sold it and purchased
He applied to the Johns Hopkins School of Medicine and was
another run-down chocolate shop. In another year, that shop
was rejuvenated and sold. This pattern continued until Abetold during the interview that, while he was otherwise qual
entered medical school in 1941, except for a brief interruptionified, he would be rejected because he was Jewish.
in 1932. In the fall of 1941, Abe enrolled at the Albany Medical
In 1932, the Lilienfelds migrated to Palestine in re College. He experienced a difficult first semester; his father
sponse to the World Zionist Organization's call for the Fifth passed away during that term. In the middle of the semester,
Lilienfeld was ready to quit school and return to New York
City to help support his mother. At the train station, his
roommate persuaded him to at least finish the semester. He
did so, and then transferred to the University of Maryland
medical school in Baltimore.
After the attack on Pearl Harbor, all United States
medical schools were nationalized, all medical students in
ducted into the military, and classes held year-round. During
this time, Lilienfeld was introduced to Lorraine, the youngest
daughter of Baltimore businessman Max Zemil. During the
next 2 years, as Abe progressed through medical school, he
developed a strong relationship with Max; in 1943, having
received the approval of Lilienfeld's armed forces com
mander, Abe and Lorraine married. In marrying Lorraine,
FIGRE . Arahm . Lliefel ("be) a hoe. ndaed Lilienfeld came into contact with Lorraine's brother-in
law, Yak.
Abe graduated from medical school in 1944. For the
next year, he persevered through a traditional rotating intern
ship at a Baltimore community hospital. He disliked clinical
medicine, particularly the emphasis on performing proce
dures for income. Having completed his internship, Lilienfeld
FIGURE 2. Abraham M. Lilienfeld ("Abe") at home. Undated.was assigned by the military to Walter Reed Hospital, then

c 2007 Lippincott Williams & Wilkins 509

This content downloaded from on Thu, 14 Nov 2019 11:12:27 UTC
All use subject to
Lilienfeld Epidemiology * Volume 18, Number 4, July 2007

one of the crown jewels in American medicine. He re

The 1948-1949 academic year at Johns Hopkins opened
quested assignment in internal medicine,
newbut the
vistas for onlynot
Lilienfeld, slot
only the course material but also
open was in otolaryngology; for 6 months, he worked
the faculty. He was taughtas an Langmuir, William Cochran,
by Alex
Ernest Stebbins,
otolaryngology resident. Bored and unhappy, and Phil Sartwell, all of whom had major
sought transfer to anything other than otolaryngology.
impact on his career development. He also met a young biosta
When a position opened in the obstetrics-gynecology
tistician who would becomeresimportant in his later career-Irwin
idency program, he transferred. Bross, a postdoctoral fellow in biostatistics. At the conclusion of
While based at Walter Reed, Lilienfeld and
the 1-year his wife
curriculum, Lilienfeld graduated with his MPH de
lived with the Yerushalmys in Bethesda, gree Maryland.
and moved his family Lilien
to Albany, New York for the training
program's second
feld's previous experience in British Mandate year.
Palestine pre
The year spent in Albany
pared him well for this experience; not only conversational in was a significant one for
Hebrew, he also knew many of the "local" Lilienfeld.
idioms Duringfamiliar 1949-1950,
to he displayed a vision of pub
Yerushalmy. The 2 conversed on a wide range
lic health of topics, not
as all-encompassing. Early in the year, on behalf of
only medicine or family. For example, during
a legislator interestedone hot palsy, the New York State
in cerebral
Legislature passed
evening during the summer of 1945, Lilienfeld a bill directing
realized his the state health department
neck was particularly hot. In the resultingtodiscussion,
spend 1 million dollars
he andfor this disease. The Commissioner
Yerushalmy considered how many layers of directed Levinwere
fabric to oversee
in the
a spending. Levin directed Lil
ienfeld to spend
shirt collar worn with a necktie. Their conclusion: the million
more than dollars, expecting the money to
be quickly
20. Accordingly, Lilienfeld resolved that upon spent on block
discharge fromgrants to construct buildings for
the military service, he would wear clip-on ties; at palsy
the clinics.
only Lilienfeld proceeded to de
bowties were available as clip-ons, and he wore a program forthe
for cerebral
rest palsy in New York State. He
of his life. included a research component (producing seminal epidemi
It was in Washington DC that Yerushalmy introduced ologic findings), training for physicians and nurses, and the
Lilienfeld to Morton Levin, who was there for business. creation of a network of clinics (signing long-term leases to
Lilienfeld indicated his displeasure with clinical practice and maximize the number of clinics established throughout the
the fee-for-service health care system then dominant in the state and sponsoring quarterly meetings of the clinic direc
United States. Levin responded with a description of a career tors) to treat cerebral palsy patients. 11-13 The program estab
in public health, and he told Lilienfeld that, should Abe lished New York State as a center for research and treatment
decide to leave clinical practice, Levin would be happy to of cerebral palsy for at least the next 2 decades. Lilienfeld's
find him a preventive medicine training position at the New view of public health and epidemiology, with its ultimate
York Department of Health. focus on the reduction of disease incidence, required a cath
In 1946, Lilienfeld's residency program assigned him olic effort, one ably financed with the bounty provided by the
to gynecology at a local community hospital. Soon thereafter, legislature.
Lilienfeld had lost all interest in clinical practice, the final One component of Lilienfeld's program concerned the
straw coming during a hysterectomy when the surgeon com epidemiology of cerebral palsy. He discussed possible re
mented upon delivering the uterus and tossing it into a search designs with Yerushalmy through the mail and at the
specimen bucket, "That's my wife's new mink coat." Lilien 1949 Annual Meeting of the American Public Health Asso
feld remembered seeing an advertisement on the medical staff ciation. During World War IL, Yerushalmy had spoken with
bulletin board at Walter Reed recruiting physicians for the Joseph Berkson, the biostatistician-physician at the Mayo
Public Health Service (PHS). Since few physicians were Clinic, about the challenges of using vital statistics data to
willing to part with the rewards promised by private practice, account for neonatal deaths. Berkson noted the linkage al
the PHS had difficulty recruiting physicians. Lilienfeld im ready in use within the Mayo Clinic records.'4 Yerushalmy
mediately contacted the recruitment officer. Within weeks, subsequently began to link birth and death records, but he
Lilienfeld was transferred to the PHS. Yerushalmy requested didn't complete the effort before changing jobs. After the
Lilienfeld be assigned to the Tuberculosis Control Bureau, war, Dunn published the concept of record linkage,'5 pre
and Lilienfeld found himself reporting to his brother-in-law. sumably following one or more discussions with Yerush
Yerushalmy immediately dispatched Lilienfeld to the almy, although there are no records to document such dis
field office in Philadelphia. For the remainder of his military cussion. Having heard Yernshalmy's thoughts, Lilienfeld linked
service, Lilienfeld worked on tuberculosis control. At the the cases of cerebral palsy with their birth records." 1"13 Lilienfeld
conclusion of his military service, Lilienfeld contacted Levin later told Yerushalmy of his success-a key piece of information
and asked if Levin had any positions open in his training for Yerushalmy when the latter was asked to consult on a
program. Levin did, and Lilienfeld joined the preventive Hawaiian database study in 1954 during a sabbatical in Hono
medicine training program. For 3 months, Lilienfeld was lulu. Yerushalmy recognized the advantage that all healthcare on
posted to Lake Placid to provide basic public health services. the islands was provided in only a few places, making it possible
In late summer 1948, Levin directed Lilienfeld to enter the to link various childhood health and vital records for persons
MPH program at the Johns Hopkins School of Hygiene and living in the territory. One could then examine a variety of
Public Health. questions dealing with child development.

510 ? 2007 Lippincott Williams & Wilkins

This content downloaded from on Thu, 14 Nov 2019 11:12:27 UTC
All use subject to
Epidemiology * Volume 18, Number 4, July 2007 Abe and Yak

During the mid 1950s, Lilienfeld continued to publish

statement regarding his circumcision status. It turned out men
were generally and
papers on cerebral palsy and other neurologic conditions, unable to state their status accurately when
he suggested that peripartum events were associated with the
with the assessment of a health care provider,
occurrence of these conditions.16'17 Officials atsuggesting considerable bias may be present in some data
the National
Institutes of Neurological Diseases and Blindnesssets.22
thought the
Later in his career, Lilienfeld collaborated with Dia
time appropriate to validate the findings and discussed
mond on whether themisclassification resulted in an overesti
idea with Lilienfeld. He suggested a study mateusing of therecord
relative risk or an underestimate.23'24 Not until
linkage between birth records and medical records.
Baron'sHe 1977noted
formulation of a general model of misclassifi
to Yerushalmy the subsequent development of the National
cation did Lilienfeld consider the issue settled.25
Collaborative Perinatal Project. Concurrently, Yerushalmy
initiated his own record linkage project, the Child Health Methods
Development Studies, in which vital records were linked withof the polio-gamma globulin study un
The analysis
medical records at Kaiser Permanente, a San Francisco-based
dertaken by the CDC during the 1950s and the analyses of
health maintenance organization. This effort pioneered
data used bywhat Modan in his postdoctoral study of polycythe
subsequently became the Kaiser Permanente linked
mia veradatabase,
exemplify the interaction of Abe and Yak with regard
which has supported many epidemiologic studies since
to statistical its
development in the late 1950s. 18,19 In the early 1950s, the CDC invited Lilienfeld to assist
At the end of his training program in 1950, in Lilienfeld
training Epidemic Intelligence Service personnel.26 Fol
searched for a job. His responsibilities now included
lowing one such2 chiltraining session, Langmuir contacted Lilien
dren. Although Levin hoped to keep him in Albany, there
feld about joining CDC, using Lilienfeld's direction of an
were no open positions available. Ernest Stebbins, Dean
epidemiologic at of gamma globulin in the prevention of
the Johns Hopkins School of Hygiene and Public polio as aHealth,
pilot. The study commenced, and in short order the
intervened, arranging for Lilienfeld to serve as were
data a regional
collected and analyzed. Langmuir assembled an
health officer in the Baltimore City Health Department.
advisory boardThe of health officers from New York, Pennsyl
epidemiology department at Johns Hopkins hadvania,sustained the
Illinois, Massachusetts, and Michigan (the most pop
loss of Alexander Langmuir in 1949 to directulous the Epidemic
states in the country) to vet the study findings. In 1954,
Intelligence Service at the Centers for Disease Control
Lilienfeld presented the findings to a meeting of this board.
(CDC). The School needed teaching assistants to cover forpresentation began smoothly as he de
Langmuir's absence, and Stebbins thought Lilienfeld could
scribed the study background, study population recruitment,
function as such until Professor Kenneth Maxcy, the depart
and data collection. When he came to the description of the
ment chair, retired. Lilienfeld would then join the department
statistical methods used for data analysis, several of the
as a full-time junior faculty member. Accordingly, Lilienfeld
board members objected. The key method used, the anal
joined the Baltimore City Health Department.
ysis of variance, was one with which they lacked famil
Interactions iarity. Yerushalmy had suggested use of the technique to
During their careers, Lilienfeld and Yerushalmyin dis Lilienfeld 1953. Concerned about its appropriateness,
cussed topics ranging from analytical methods to causal Lilienfeld consulted with Cochran, who agreed with
inference, including data quality, statistical methods and Yerushalmy's recommendation. At the board meeting, Lil
discerning biologic inferences from epidemiologic ienfeld tried responding to the objections by noting the
assent of both Yerushalmy and Cochran. The board mem
Data Quality bers would have none of it, however. Lilienfeld asked for
Data quality became a lifelong interest of Lilienfeld's; a coffee break.
it was a guaranteed topic for questions by any thesis com During the break, Lilienfeld approached Leonard
mittee that included Lilienfeld. That interest first manifested Schuman, who represented Illinois. Lilienfeld had met
after Yerushalmy's studies into the accuracy of physician Schuman at the previous American Public Health Associ
radiograph interpretation, and it solidified after Yak's return ation annual meeting, and Schuman was the only person
from his 1947 trip to England. By the time Lilienfeld arrived present besides Langmuir whom Lilienfeld knew. He told
in Albany in 1948, the seeds of his interest in data quality had Schuman he needed some help. Schuman replied, "Are you
been planted. Indeed, one of Lilienfeld's first publications sure about the technique? I know nothing about analysis of
dealt with the quality of data on birth certificates.20 variance, so I need to be sure it's OK." "Absolutely. Bill
Bross raised the issue of whether misclassification in a Cochran and Yak Yerushalmy both reviewed its use. They
2 X 2 table biased the odds ratio estimate towards unity.21 thought it was fine." "Then don't worry, Abe. Follow my lead."
Concerns about the questions used in Wynder and Graham's After the break, Schuman spoke up. "During the break,
study of cigarette smoking and lung cancer stimulated a I had a chance to review some of the aspects of this important,
discussion among Bross, Yerushalmy, and Lilienfeld at an innovative study," he began. "I must confess, though, I don't
International Biometrics Society meeting in the mid 1950s. understand all the uproar about the use of analysis of variance
Lilienfeld thought the issue worth examining in the field. In for the statistical analysis. It's a common-enough technique,
particular, no one knew whether misclassification occurred in elementary actually. Students in schools of public health
a study. The test case would be the accuracy of a given man's know all about it, and we use it all the time in the Illinois

? 2007 Lippincott Williams & Wilkins 511

This content downloaded from on Thu, 14 Nov 2019 11:12:27 UTC
All use subject to
Lilienfeld Epidemiology * Volume 18, Number 4, July 2007

Departnent of Health. It seems perfectly acceptable

fellow Johnsto me.
Hopkins alumnus Joseph Berks
Why do we need to spend precious time discussing something from the Mayo Clinic. Yer
so elementary when there are important Berkson bothpolicy
public health denied any relationship betw
issues to be addressed?" With those comments,
andwas nocancer.35 Their luncheon dis
further discussion regarding the statistical analysis stopped
focused on the basis for denying the relationsh
and the board approved the findings.26'27 Thus wasin
son strongly the
the camp of Sir Ronald Fis
analysis of variance introduced to the CDC, and thethe
contributed profes
concept of specificity (the c
sional collaboration between Leonard Schuman and
specific in Abe
its biologic effect) to the Yerus
Lilienfeld began. guidelines. (Indeed, when evidence of an assoc
Another example of the influential interactions of Abe and bladder cancer began to
cigarette smoking
and Yak occurred in the early 1960s. A.
"What now, do we breathe
Chair of the Department of Preventiveurethras?")
Medicine at the
Hebrew University Hadassah Medical School, sent one
During theof his and 1960s, Abe and Ya
students, Baruch Modan, to Lilienfeld forbated
the cigarette
in epide smoking-lung cancer relatio
miology. Modan, a hematologist, wantedin
toa study
haze of smoke from Yak's cigarette an
a hemato
These discussions
logical condition for his dissertation. Lilienfeld forced Abe to refine his thin
identified the
outcomes associated with different treatments in epidemiologic terms. Stimula
of polycythe
cussion into
mia vera as meriting inquiry. Modan's findings thinking about chronic disease
more generally,
conventional wisdom regarding such treatments, Lilienfeld brought concepts fro
and resulted
in the formation of the seminal Polycythemia Vera Study (such as incubation peri
disease epidemiology
Group.28'29 Little known, however, is thedisease
origin settings (such as occupational cancer
of the statis
tical techniques involving life tables, which Modan
vice versa used
(such as in
the use of case-control study
his thesis. Although life tables had been epidemiologic
widely usedstudiesin of infectious diseases)
epidemiology for almost a century,30,31 there had beenin his inaugural Wade Hampton F
attention to the statistical theory underlying The discussions also forced him to
their analysis.
One of Yerushalmy's colleagues, Chin Long of
process Chiang,
deriving hadbiologic inferences, the sub
subjected the various parameters of theserieslife table to mathe
of papers.36'38'39 Further, by refining h
matical analysis. While discussing Modan's Abe wasthesisbetterwithprepared as an advocate again
Yerushalmy at the annual meeting of the industry;
American his Public
1962 paper in The Nation was
Health Association, Lilienfeld asked about arguments
potentialhe had fleshed out in discussions
cal methods. Yerushalmy suggested using almy.40
some of (InChiang's
1964, when the Surgeon Genera
was constituted,
work on life tables, which Modan subsequently did. Lilienfeld was told by friends
Health Service that his 1962 paper had rend
Biologic Inferences "partial" to be appointed to the committee. H
Perhaps no area of interaction between prize" came later
Lilienfeld andin 1964 when he was asked
Staff Director
Yerushalmy was as significant as the discussions the 2forhadthe President's Commissio
regarding the derivation of causal inferences Cancer, and Stroke.41)
ologic data. Working with his former boss, Carroll Palmer,
Yerushalmy initiated one of the key discussions in 1950s
chronic disease epidemiology: what evidence is needed DISCUSSION
discern an etiologic factor.31 The issue arose inThethe context of between Lilienfeld and
the debate on cigarette smoking and lung werecancer.
fruitful Yerush
for both men throughout their c
almy, a smoker (at times, a chain-smoker) and soonfound
Yerushalmy to be a capable voice for many
paid consultant to the tobacco industry, Yak,hadLilienfeld
reviewed found
the a source of useful bio
available information. He did not view it as wholly
cepts consistent
and study design ideas. Through his early
with an etiologic relationship between cigarette smoking
with Yerushalmy, andLilienfeld developed an under
lung cancer. In 1959, Yerushalmy and Palmer proposed a perspective
the statistical set in epidemiology. H
of guidelines, modified from the Henle-Koch postulates,
was able as
to form alliances with such statistic
the basis for deriving causal inferences.32 Evans33(then
Cochran discussed
Chair of the Biostatistics Departm
these guidelines as seminal to the subsequent Surgeon
Hopkins), Jerry Gen
Cornfield, Sam Greenhouse,
eral's guidelines and the Hill guidelines. Like
erin (atSir
NIH). Lilienfeld's ability to inter
Fisher, Yerushalmy opposed the notion ticians
that cigarette smok
bore fruit in the 1960s, including his se
ing caused lung cancer until his dying day.34 Unlike Fisher,
the development of logistic regression and his
Yerushalmy did not dispute the value of theobservational
Framingham data
Study from closure.42" The
more generally. Rather, he viewed the smoking
developed data as not
in part out of his familiarity wit
consistently supportive of a causal association.
techniques developed by Jerry Cornfield exp
At professional biostatistical meetings, during the lateLilienfeld's efforts to advanc
1950s and early 1960s, Yerushalmy would within epidemiology
often lunch with was recognized by the

512 ? 2007 Lippincott Williams & Wilkins

This content downloaded from on Thu, 14 Nov 2019 11:12:27 UTC
All use subject to
Epidemiology * Volume 18, Number 4, July 2007 Abe and Yak

munity, which made him a Fellow in the American They viewed themselves as members of the public health
Association. community. While they had many disagreements, personal
Yerushalmy's need for a voice in the epidemiologic and professional, during their careers, they remained mem
community may not have been as important in the 1950s as bers of the public health community and continued their
it would become in the 1960s and early 1970s. His disdain for interactions as such. At a time of increasing polarization
physicians likely contributed in part to the frictions between within epidemiology and within the scientific community
him and Reuel Stallones, a professor in the epidemiology generally, we can emulate these 2 leaders whose commitment
departnent at the University of California, Berkeley. Stal to public health transcended their personal disagreements,
lones eventually left Berkeley to found the School of Public even for major issues such as whether cigarette smoking was
Health at the University of Texas, Houston. At that point, a cause of lung cancer. For example, Yerushalmy did not
absent Lilienfeld, Yerushalmy had few avenues left into the allow that disagreement to stop his suggestion to Lilienfeld of
epidemiology community. The situation was compounded by analytic techniques for use by Modan. Communication is key
Yerushalmy's stand on cigarette smoking and lung cancer. As the formation and sustenance of a community; both Lilienfeld
the 1960s progressed, the epidemiologic evidence for the and Yerushalmy sought to maintain such communication. We
relationship grew, and, with that growth, those (like Yerush would do well to continue to follow their lead.
almy) not recognizing the evidence as increasingly conclu
sive became intellectually isolated. As much as Lilienfeld
needed and benefited from his interactions with Yerushalmy,
It is popular in the history of medicine and public
so too did Yerushalmy need and benefit from those same
health to focus on social trends and movements as expla
interactions-a truly symbiotic relationship up until Yerush
nations of historical phenomena. In some instances, such
almy's death in 1973.
as the promotion of cigarette smoking and the discovery of
What can we learn from this interaction? There are a
its subsequent health effects, it is difficult to examine the
number of lessons: advocacy and the role of public health in
events in other than a societal context.4850 In the present
epidemiology, the formation of modern epidemiology prior to
instance, however, the reader may see the effect of 2
the development of the cigarette smoking-lung cancer con
individuals, sometimes collaborating, sometimes disagree
troversy, and the significance of community within the field
ing, on the subsequent development of epidemiology. For
to facilitate its advancement.
example, record linkage in the United States can be traced
Advocacy back (at least in part) to Berkson's suggestion through
Both Lilienfeld and Yerushalmy began their careers Yerushalmy to Lilienfeld to link birth and death records as
focused on public health, and both spent considerable por a means of creating a population for epidemiologic obser
tions of their early careers in public health settings. They vation. The later development of the Kaiser-Permanente
were strong advocates for public health. The notion that database built on these efforts. Modem pharmacoepidemi
researchers should not advocate on behalf of public health ology was one result of the availability of the Kaiser
would likely strike each as foreign; advocacy was a given Permanente and other similar databases developed in the
aspect of membership in the public health community. It is 1970s and early 1980s.51 Few today are aware of these
good to be reminded that advocacy by epidemiologists on connections, as little has been published on them.
behalf of public health has a long and respectable history. Although this paper has presented some information
Advocacy by researchers did not inhibit the development of about the intellectual output of both men, each could be the
well-founded health policy in the United States; indeed, it no subject of dissertations in the history of epidemiology and
doubt assisted it. public health. My observations are not exhaustive; there are
likely other interactions I know nothing about. The nature of
Formation of Modern Epidemiology Prior to my relationship with both men (and their relationship with
the Cigarette Smoking-Lung Cancer each other) carried with it the lack of formal documentation.
Controversy Lilienfeld rarely saved correspondence, working drafts of
Yerushalmy's entry into public health began prior to manuscripts, presentations, etc. The same is true for Yerush
the post-World War II cigarette smoking-lung cancer con almy. Much of the evidence reported in this paper is apoc
troversy, while Lilienfeld's coincided with the development ryphal; I trust that this does not diminish the credibility of the
of the issue. Even before the first American case-control findings themselves.
studies were published in 1950, Lilienfeld began to focus on
noninfectious diseases. The second paper of his career exam ACKNOWLEDGMENTS
ined the relationship between alcohol consumption and liver I thank Aifredo Morabia for his helpful suggestions
cirrhosis.47 In general, epidemiology had begun its transfor regarding the manuscript.
mation from an infectious disease orientation to one domi
nated by chronic diseases.
Community 1.
& Accessed July 20, 2006.
As noted previously, both Lilienfeld and Yerushalmy 2.^ Accessed
conducted their work within the context of public health. July 20, 2006.

? 2007 Lippincott Williams & Wilkins 513

This content downloaded from on Thu, 14 Nov 2019 11:12:27 UTC
All use subject to
Lilienfeld Epidemiology * Volume 18, Number 4, July 2007

in 1953: summary of the report of the National Advisory Committee for

pmcll. htm,. Accessed December 25, 2005. evaluation of gamma globulin. JAMA. 1954;154:1086-1090.
4. Kluger R. Ashes to Ashes. New York: Knopf; 1996. 28. Modan B. An epidemiological study of polycythemia vera. Blood.
5. Lilienfeld AM, Lilienfeld DE. Foundations of Epidemiology. 2nd ed.
New York: Oxford University Press; 1980. 29. Modan B, Lilienfeld AM. Polycythemia vera and leukemia?the role of
6. Yerushalmy J. The 1940 record of maternal and infantradiation
mortality inA the
treatment. study of 1222 patients. Medicine. 1965;44:305
United States. Child. 1942;6:195-206. 344.
7. Yerushalmy J. Maternal and infant mortality in the United States,
30. Chiang CL. 1941.
The standard error of expectation of E. B. Wilson approach.
Child. 1943;7:110-116. Hum Biol 1966;38:318-319.
8. http://www.time.eom/trme/archive/preview/0,10987,886565,00.
31. Chiang CL. Lifehtml.
Table andAc
Mortality Analysis. Geneva, Switzerland:
cessed July 20, 2006. WHO; 1976.
9. Yerushalmy J. Statistical problems in assessing methods
32. Yerushalmy ofJ,medical
Palmer CE. On the methodology of investigations of
diagnosis, with special reference to X-ray techniques. Public
etiologic Health
factors in chronic diseases. J Chronic Dis. 1959;10:27-40.
Rep. 1947;62:1432-1449. 33. Evans AS. Causation and Disease. New York: Plenum; 1993.
34. Stolley PD. When genius errs: Fisher and the lung cancer controversy.
id=div00052&toc.depth=l& Accessed July 20, 2006. Am J Epidemiol. 1991;133:416-425.
11. Lilienfeld AM, Parkhurst E. A study of the association 35. Berksonof factors
J. Smoking and of
cancer of the lung. Mayo Clin Proc. 1960;35:
pregnancy and parturition with the development of cerebral 367-385. palsy: a
preliminary report. Am J Hyg. 1951;53:262-282. 36. Lilienfeld AM. The epidemiologic method in cancer research. J Chronic
12. Lilienfeld AM, Parkhurst E, Patton R, et al. AccuracyDis. of 1958;8:649-654.
medical information on birth certificates. Public Health Rep. 1951;66:
37. Lilienfeld AM. Epidemiologic of infectious and non-infectious disease:
some comparisons. The first Wade Hampton Frost lecture. Am J Epide
13. Lilienfeld AM, Perlstein AM. Cerebral palsy. In: Proceedings of the miol. 1973;97:135-147.
Conference on Chronic Diseases: Preventive Aspects. 1952.
38. Lilienfeld AM. On the methodology of investigations of etiologic factors
14. Melton LJ III. History of the Rochester epidemiology project. Mayo Clin
in chronic diseases: some comments. J Chronic Dis. 1959;10:41-46.
Proc. 1996;71:266-274.
39. Lilienfeld AM. The distribution of disease in the population. J Chronic
15. Dunn HL. Record linkage. Am J Public Health. 1946;36:1412-1416.
Dis. 1960;11:471-483.
16. Lilienfeld AM, Pasamanick B. Association of maternal and fetal factors
40. Lilienfeld AM. The case against the cigarette. The Nation. March 31,
with the development of epilepsy. I. Abnormalities in the prenatal and 1962.
postnatal periods. JAMA. 1954;155:719-724.
41. President's Commission on Heart Disease, Cancer, and Stroke. GPO,
17. Lilienfeld AM, Pasamanick B. The association of maternal and fetal
Washington, 1964.
factors with the development of cerebral palsy and epilepsy. Am J Obstet
42. Walker SH. Estimation of the probability of occurrence of disease as a
Gynecol. 1955;70:93-101.
function of several variables. Ph.D. Dissertation. Baltimore, MD: Johns
18. Accessed
Hopkins University; 1965.
July 20, 2006.
43. Walker SH, Duncan DB. Estimation of the probability of occurrence of
19. Accessed July
20, 2006. disease as a function of several independent variables. Biometrika.
20. Lilienfeld AM, Parkhurst E, Patton R, et al. Accuracy of supplemental
medical information on birth certificates. Public Health Rep. 1951;66:191 44. Levy D, Brink S. A Change of Heart. New York: Knopf; 2005.
198. 45. Truett J, Cornfield J, Kannel W. A multivariate analysis of the risk of
21. Bross IDJ. Misclassification in 2 X 2 tables. Biometrics. 1954;10:478-486. coronary heart disease in Framingham. J Chronic Dis. 1967;20:511-524.
22. Lilienfeld AM, Graham S. Validity of determing circumcision status by 46. Cornfield J. Joint dependence of risk of coronary heart disease on serum
questionnaire as related to epidemiological studies of cancer of the cholesterol and systolic blood pressure: a discriminant function analysis.
cervix. JNatl Cancer Inst. 1958;21:713-720. Fed Proc. 1962;21:58-61.
23. Diamond EL, Lilienfeld AM. Effects of errors in classification and 47. Lilienfeld AM. Korns RF. Some epidemiological aspects of cirrhosis of
diagnosis in various types of epidemiological studies. Am J Public the liver. Am J Hyg. 1950;52:65-81.
Health. 1962;52:1137-1144. 48. Gostin LO, Brandt AM, Cleary PD. Tobacco liability and public health
24. Diamond EL, Lilienfeld AM. Misclassification errors in 2 X 2 tables policy. JAMA. 1991;266:3178-3182.
with one margin fixed: some further comments. Am J Public Health. 49. Gostin LO, Arno PS, Brandt AM. FDA regulation of tobacco advertising
1962;52:2106-2110. and youth smoking. Historical, social, and constitutional perspectives.
25. Barron BA. The effects of misclassification on the estimation of relative JAMA. 1997;277:410-418.
risk. Biometrics. 1977;33:414-418. 50. Brandt AM. Emerging themes in the history of medicine. Milbank Q.
26. Etheridge EW. Sentinel for Health. A History of the Centers for Disease 1991;69:199-214.
Control. Berkeley: University of California Press; 1991. 51. Lilienfeld DE. Chronology of pharmacoepidemiology. In: Hartzema
27. National Gamma Globulin Evaluation Center. Lilienfeld AM (Director). BH, Tilson HH, eds. Pharmacoepidemiology. 4th ed. Cincinnati: Harvey
Evaluation of gamma globulin in prophylaxis of paralytic poliomyelitis Whitney Books; In press.

514 ? 2007 Lippincott Williams & Wilkins

This content downloaded from on Thu, 14 Nov 2019 11:12:27 UTC
All use subject to