Biomedicine and Its Historiography: A Systematic Review

Nicolas Rasmussen

Contents Introduction ...... 1 What Is Biomedicine? ...... 2 Biomedicine’s Postwar Development ...... 7 The Distribution of Activity in Biomedicine and in Its Historiography ...... 14 Conclusion ...... 19 References ...... 20

Abstract In this essay I conduct a quantitative systematic review of the scholarly literature in history of life sciences, assessing how well the distribution of the activity of historians aligns with the distribution of activities of scientists across fields of biomedical research as defined by expenditures by the cognate institutes of the United States NIH. I also ask how well the distribution of resources to the various research fields of biomedicine in the second half of the 20th Century has aligned with morbidity and mortality in the United States associated with the cognate categories. The two exercises point to underexplored areas for historical work, and open new historical questions about research policy in the US.

Introduction

I have taken an unusual approach in this essay to ask a question that, to my knowledge, has not been addressed before: how well, in terms of subject matter, does our historiog- raphy describe the domain of biomedicine? I shall define the subject matter by treating it

N. Rasmussen (*) University of New South Wales, Sydney, NSW, Australia e-mail: [email protected]

# Springer International Publishing AG, part of Springer 2018 1 M. Dietrich et al. (eds), Handbook of the Historiography of Biology, Historiography of 1, https://doi.org/10.1007/978-3-319-74456-8_12-1 2 N. Rasmussen as an actor’s category; that is, I define biomedicine according to the way the concept was used by scientists and policy-makers in the institutions they created at mid-century. I shall briefly describe the early history of the institutionalization and growth of this kind of research activity after the Second World War. Then I shall bring quantitative evidence to bear in a way that is somewhat novel to map the distribution of resources, hence research activity, among the major fields within the biomedicine of the United States during the five decades that followed the War. (I focus on the United States because it was the polity supporting and shaping biomedical science far more than any other and because it exerted additional influence as the West’s leader.) The first major purpose is to ask in quantitative terms how well the distribution of research effort among biomedical fields matches the importance of the different fields of research to the health of the US population – as political leaders and technocrats of the day acknowledged that they should – or whether, on the other hand, there are interesting historical questions about why some fields attracted disproportionately little or much cultural and political favor. This leads me to propose a number of topic areas for further research by historians of biomedical research policy and politics. An additional purpose for charting the distribution of health research resources is that it enables one to address the novel question noted above and an issue central for this collection: assessing how well the biomedical historiography aligns with the historical trajectory of biomedicine itself. That is, I ask whether historians are distributing their efforts to the various fields of the biomedical enterprise in a way roughly matching the past distribution of efforts by biomedical researchers and policy-makers generating the activity itself. Here, too, we will be able to identify apparent mismatches between events and attention and offer some hypotheses about why certain fields appear relatively under- or overstudied by historians. To identify mismatches I apply a quantitative method to the historiography – systematic literature review – rare for the history of science and medicine but standard in certain biomedical fields. In a stand-alone review article, I would follow these three quantitative sections – dealing in turn with the biomedical research enterprise over time, its fit with population health needs, and its fitwiththe historiography purporting to describe it – by drilling down into several particularly active areas of biomedical historiography, looking qualitatively at the prevalent questions being addressed and fashionable approaches to answering them. However, given that this review is part of a collection in which many of the other historiographic articles deal with areas of life science that overlap what I (following the US National Institutes of Health, NIH) have broadly defined as biomedicine, I will leave such qualitative analyses to the other authors. The main aim of my largely quantitative survey is to point out historio- graphic lacunae and raise new questions for historians; it is also intended to demonstrate that quantitative systematic review methodology can be productive, but not to rule out other more qualitative approaches.

What Is Biomedicine?

The first task of any literature review must be to define the subject matter. Bibliometric tools (such as Google Ngram for book contents or ProQuest Historical Newspapers for American journalism) show that as a term in English usage, and Biomedicine and Its Historiography: A Systematic Review 3 arguably too as a generally accepted concept, “biomedicine” appears around 1960 and becomes ubiquitous by 1970. So, while both clinical research and experimental biology with deliberate medical application certainly dates to the nineteenth century, for example, in Koch’s and Pasteur’s famous vaccine trials and Helmholtz’s and Bernard’s work on nerve action, and while furthermore medicine has increasingly borrowed concepts, practices, and authority from laboratory sciences since the late nineteenth century (Vogel and Rosenberg 1979; Starr 1982; Ludmerer 1985; Rothstein 1987; Porter 1998), the postwar period manifests this melding of science and clinical technique on a new scale. After the Second World War, government funding for research dramatically expanded in the United States, quickly over- shadowing private philanthropies such as the and Institute – which before the War had been the preeminent patrons of both biological and , and not only in the United States. In the postwar United States, the position of dominant patron was rapidly assumed by the National Institutes of Health. In the mid-1970s, for example, NIH funding accounted for 40% of US health R&D, as compared with about a 25% share from all other government funders, about 30% from the private sector, and less than 5% nonprofits and philanthropies. The overwhelming dominance of the NIH among all US funders of life science is also attested by a comparison of the agency’s 1970 budget of more than $1 billion with the $49 million research budget of the National Science Foundations’ (NSF) division of Biological and Medical Sciences (Table 1). Indeed, in terms of resources, the NSF as a patron of life science was roughly equivalent to one of the more modest NIH Institutes (Strickland 1972; Brown 1979; Shyrock 1980; Bloom and Randolph 1990; Kohler 1991; Appel 2003; Hamowy 2008). As implied by the mushrooming NIH budget and proliferation of NIH Institutes from the late 1940s through the 1960s, briefly discussed below, health-oriented science became a social priority in the United States like never before. That commitment to biomedical research, by the world’s richest nation amidst a global scientific community shattered by war, made the United States the dominant force in health research globally. Indeed, it was not until the mid-2000s that the US share of world health research and development spending fell beneath 50%.1 Beyond mere spending power, we may also point to literature indicating that in biomedicine as in science generally, during the Cold War many Western countries tended to emulate the “Free World’s” leader in order to demonstrate solidarity and furthermore that the East Bloc often felt compelled to compete with US scientific strengths (Gaudillière 2002; Krementsov 2002; Krige 2006). This cultural leadership in the postwar period offers another reason why the American biomedical research enterprise shaped others throughout the world. Thus, putting aside qualms about provincialism, for

1US public and private combined biomedical research spending still accounted for half of world spending in the mid-2000s (Moses et al. 2015). I am not aware of reliable figures earlier than this, but given the postwar state of European and Asian economies until the 1980s, it would be safe to suppose that before the 1990s the United States would have accounted for far more than half. 4

Table 1 Budgets of NIH Institutes founded before 1970 and NSF research budgets (in current $1000s) over time NIH Institute Founded 1960 budget 1970 budget 1980 budget 1990 budget NCI 1937 91,257 181,454 999,869 1,634,332 NHI (NHLBI) 1948 62,237 160,634 527,488 1,072,354 NMI (NIAID) 1948 34,054 97,342 215,364 832,977 NIDCR 1948 10,019 28,754 68,303 135,749 NIMH 1949 67,470 NA 212,000a 727,000b NIAMD (NIDDK) 1950 46,862 131,761 341,206 581,477 NINDB (NINDS) 1950 41,487 97,315 241,966 490,409 NICHD 1962 76,095 208,953 442,914 NIGMS 1962 148,294 312,468 681,782 NEI 1968 22,828 112,989 236,533 NIEHS 1969 17,423 83,893 229,234 NIH total 339,380 1,061,007 3,428,935 7,576,352 NSF total 158,600c (88,350 res) 462,490 (316,410 res) 975,130 (836,830 res) 2,100,000 NSF BMS/BBS (res) 24,870d 49, 450c 170,000e 300,000f BMS, BBS Directorate of Biological and Medical Sciences/Biological Behavioral and Social Sciences Although total Institute budgets are given here, approximately 90% represents research expenditure. For NIMH, incorporated in the late 1960s as a research division within another non-NIH agency dedicated mainly to service delivery (ADAMHA), research expenditures are given for 1980 and 1990. In 1992 NIMH, together with NIDA and NIAAA, was reconstituted as an Institute when that agency was reorganized. (Sources: NIH figures from http://officeofbudget.od.nih.gov/approp_hist.html) aADAMHA research division budget from http://www.ncbi.nlm.nih.gov/books/NBK235741/ bADAMHA research division budget from http://www.ncbi.nlm.nih.gov/books/NBK235734 cBudget obligations (http://dellweb.bfa.nsf.gov/NSFFundingbyAccount.pdf), and .Rasmussen N. dNew grants, both from Appel (2003) eTotal divisional appropriation (http://library.cqpress.com/cqalmanac/document.php?id=cqal79-1183790); research obligation not available fApproximate estimate from http://www.ncbi.nlm.nih.gov/books/NBK235736/ Biomedicine and Its Historiography: A Systematic Review 5 present purposes I will define “biomedicine” as the areas of research supported and conducted by NIH. This essay will emphasize the half-century from 1945, which is to say the period in which biomedicine as we know it first appeared, up to the recent horizon of what can be considered history. But I will take it as given that many of the particular research disciplines and fields that flourished under the “biomedicine” aegis existed in some form since the early twentieth century. In accepting this postwar actor’s category, I am aware that I am bypassing a significant body of sociological literature – and controversy – seeking to define the essence of the biomedical, and of associated “biomedicalization,” and when exactly the postmodern form of “biopower” emerged (in the 1980s for many authors; Rabinow and Rose 2006; Sunder Rajan 2006; Clarke et al. 2009). However, for purposes of the present review, we do not need to theorize biomedicine deeply; we only need to recognize biomed- ical research and historical work describing it. For this limited scope, the actor’s category works well and reduces the risk of anachronism. So, for present purposes “biomedicine” is defined operationally as what NIH did (and does). As the names of the Institutes indicate – as do the introductory sections of their present web sites – the research supported by NIH is defined not so much by the disciplines of academic life science (cell biology, developmental biology, , neuroscience) as by the disciplines and specialties of academic medicine, which roughly correspond to human organ systems and their . Thus the National Institute (NCI) sought (and seeks) to expand the knowledge base and techniques of the conjoined specialties of hematology and oncology; the National Institute of Mental Health (NIMH) psychiatry; the National Heart Institute (NHI, later, Heart, Lung and Blood Institute, NHLBI), cardiology; the National Institute of Allergy and Infectious Disease (NIAID) the gradually diverging special- ties of immunology and infectious disease; and so forth. Biomedicine’s auto- definition according to medical specialties, by its central agency the NIH, presents an immediate challenge to the historian (and historiographer) of biology. Historians of life science typically define their own subject matter according to the fields one finds in a university faculty of arts and sciences, corresponding to the way that the key figures we have mainly focussed on, biology professors, define themselves. Scientists with very closely related research programs, when employed in medical schools, often find themselves and their departments defined at least indirectly by reference to clinical practice. Indeed we may already sense a shortcoming of the professional historiography, because the list of Nobel Laureates in Physiology or Medicine – a Prize mapping well onto “biomedicine” in the postwar period – shows that many of the century’s leading biomedical scientists held physician’s degrees and have defined their own careers in terms of medical speciality and problem, in NIH fashion. Sampling the Prize every 5 years or so, consider Max Theiler (1951, virology, on yellow fever), Dickinson Richards (1956, cardiology, on catheterization), Frank Macfarlane Burnet (1960, immunology, on transplant compatibility), CB Huggins (1966, oncology, on ), or Earl Sutherland Jr (1971, endocrinology, on hormone 6 N. Rasmussen signalling).2 Such medically oriented figures (with a few exceptions) have received but little attention from professional historians, in that historians of medicine typi- cally attend more to patient care while historians of biology attend more to biologists oriented toward basic science. Thus, a great deal of past biomedical science has fallen between the stools, so to speak, of the scholarly fields of history of biology and history of medicine. In the next section, I will explore the validity of this impression through a systematic review of the historiography concerning the many areas of biomedical research encompassed by NIH activity. All this is not to say that historians of science have failed to notice scientists funded by NIH and therefore biomedicine as viewed from Bethesda. Of course they have studied NIH-funded scientists, as one expects given that the NIH was the dominant funder of life science in the United States, both basic and applied (to use the common- place postwar distinction). World-leading departments in fields we think of as completely removed from the clinic were built on NIH money, for example, the molecular biology hothouse of Stanford’s Biochemistry Department (in the medical school), as ably discussed by Doogab Yi (2015), where luminaries like Paul Berg and Arthur Kornberg solved the riddles of in E. coli bacteria. But historians typically act as if NIH medical agendas had no influence over what they did. The biology, historians suppose, was the same no matter who funded it (so long, at least, as it was scientist-initiated and scientist-evaluated research). This conceit meshes with the internalist premises of Mertonian and Kuhnian history and philosophy of science scholarship from the Cold War era, which in turn reflects the self-conception of many scientists of the time – a construction of good science as independent of whatever practical or political implications it might have that was encouraged by the American National Security State in its early Cold War heyday (Hollinger 1995; Fuller 2000; Shapin, 2008, chapter 3). This purist conceit among biologists played a key role in the cultural Cold War, I have argued (Rasmussen 2014, chapter 1), in the same way that the modernist conceit that good art seeks only to advance the potential of its medium and discipline (e.g., in abstract expressionist painting) found favor among Guggenheim Foundation and State Department decision-makers promoting the image of a Free World where artists and intellectuals were not enslaved to political agendas (Saunders 2000; Osgood 2006; Wilford 2009). This same conceit allowed university physicists developing the maser in classified Navy-sponsored research projects to imagine that they were simply pursuing paradigm-driven science for its own sake (Forman 1992) – and similarly, biologists pursuing the of cancer with tobacco industry funding (Bero 2005; Proctor 2011, chapter 16; Brandt 2012). It is time, I would urge, for historians of life science to shake off the Cold War hangover and broach the question of whether the science really was the same regardless of who funded it, even when it was peer-reviewed. But here I stray from my central task of comparing the biomedical

2http://www.nobelprize.org/nobel_prizes/medicine/laureates/, accessed 2 July 2015. I have not counted how many Physiology or Medicine Laureates have higher medical, dental, or veterinary degrees, but the proportion is very substantial [why not count and give stat to us?]. Another interesting question is whether that proportion has changed over time. Biomedicine and Its Historiography: A Systematic Review 7 territory’s distribution to its historiographic map. After we have a more reliable measure of key divergences between the attention of professional historians of bio- medicine and the historical terrain they aim to describe, I will venture some specula- tions to explain these divergences.

Biomedicine’s Postwar Development

Before we compare the recent historiography to the past biomedical research enter- prise it describes, a review of that enterprise’s emergence is in order. For the United States, the rise to prominence of biomedical research on the political scene has been comparatively well studied. Biological and medical research contracts of the war- time Committee on Medical Research (a branch of Vannevar Bush’s science agency, OSRD) were in 1946–1947 adopted wholesale by the NIH’s director Rolla Dyer under Surgeon General Thomas Parran, the expansionist chief of the late New Deal Service (PHS). While the National Science Foundation proposed by Bush was delayed in Congress by disputes over its exact form and purpose (until 1950), by the end of 1947, the NIH had transformed itself from a research sponsor mainly of in-house research at its Cancer Institute, as well as the home to smaller PHS laboratories with infectious disease and other public health functions, to a patron of academic life science research on a scale an order of magnitude larger. Riding a wave of popular enthusiasm for the idea that the main threats to health and longevity could be conquered through science, a throng of new National Institutes of Health was conjured by lobbyists such as Mary Lasker and their Congressional allies – the National Institute of Mental Health in 1946 (but funded and formally established in 1949), the Heart, Microbiology, and Dental Institutes in 1948, and the Diabetes and Stroke Institutes in 1950 (Kevles 1977; Fox 1987; Strickland 1989; Kleinman 1994; Hamowy 2008). All boasted a campus laboratory in Bethesda but sponsored much more research extramurally through grants. As shown in classic work by Stephen Strickland (1972), based largely on Congressional and other political discourse, the rapid growth of the NIH within PHS during this period emerged as an unintended quasi-compromise from the clash over Harry Truman’s 1948–1950 effort to institute national health insurance. This initially popular plan was notoriously defeated by an all-out lobbying and advertis- ing campaign against “socialized medicine” by the American Medical Association and its business allies (Poen 1996; Mayes 2004, chapter 3; Boychuk 2008, chapter 3). Public funding of research to improve medical care (for paying patients) became a palatable alternative to publicly funded medical care, not just for physicians and drug companies but also for health advocates and for politicians concerned to demonstrate both to constituents and to the polarized Cold War world that the United States cared about its citizen’s health. Given an impoverished postwar Europe encumbered by a greater dedication to health care and other social welfare, this concerted effort by the world’s richest nation quickly made it the world leader in life science research and likewise clinical technique. This is not to suggest extreme divergence between the United States and Western Europe, however. On the one hand, other Western nations 8 N. Rasmussen

2,500,000 0.7

0.6 2,000,000 0.5

1,500,000 0.4

0.3 1,000,000

0.2 500,000 0.1

0 0

o NIH Budget ($1960) YEAR R01 success rate

Fig. 1 NIH grant success rates and budget (constant 1960 $1000s), 1960–1995. (Sources: For NIH appropriations, http://officeofbudget.od.nih.gov/approp_hist.html; for NIH grant success, http:// www.nih.gov/UploadDocs/Estimated_success_rates_1962-2008.xls (both downloaded February 7, 2013); and for historical (urban CPI) inflation values from the US Bureau of Labor Statistics, http://www.usinflationcalculator.com/inflation/consumer-price-index-and-annual-percent-changes- from-1913-to-2008 (downloaded June 9, 2013)) seeking to retain great power status, like France, also emulated the American high- technology biomedical turn in life science (Gaudillière 2002). On the other, the American Federal government found indirect ways to subsidize health care, such as through the Veterans Administration and hospital construction, long before Medicare was instituted in 1966 (Hamowy 2008). Under the NIH aegis in the 1950s and 1960s, the life sciences as a whole participated in what Roger Geiger (2008) has called the “hypertrophy of disinter- ested research” at universities and other scientific institutions in the early Cold War era. NIH funding grew exponentially from the early 1950s, reaching a peak in 1966 at just over $1bn constant 1960 dollars. But the growth of the research enterprise outstripped Federal funding; already in 1966, at peak funding, grant success rates dipped below 1/2 and dropped to 1/3 over the next 5 years as competition intensified (Fig. 1). Therefore in the same period that the Mansfield amendment reduced military funding for physical sciences at American universities, Federal funding for life sciences declined in real terms – and was perceived to move in a more narrowly pragmatic direction too, so that (for instance) at Stanford Paul Berg worried in 1970 that NIH might no longer support any work on bacterial viruses because it was too remote from medical application (Yi 2015, 55). NIH funding did not return to its 1966 level in real terms until 1972, with an infusion of money from the 1971 National Cancer Act (“Nixon’s war on cancer”), and despite this infusion remained at an approximate plateau of a little more than $1 billion 1960 dollars for more than a decade. Real NIH research spending did not start to expand Biomedicine and Its Historiography: A Systematic Review 9 well beyond the 1966 level until after 1984. This late 1960s funding stall and long plateau, combined with a biomedical enterprise still expanding after two decades of consistent growth, produced some very hard years for US researchers in the mid-1970s – as indicated by unprecedented low grant success rates. The distinct existence of this interruption in funding growth for life science research is not well appreciated by historians of biomedicine and still less by other historians of science (e.g., Agar 2012, p. 435); it needs to be incorporated into our understanding of everything that happened in the life sciences around the 1970s, such as the emer- gence of biotech (Rasmussen 2014). Moreover, it intriguingly suggests that US biomedical research spending tracks the intensity of the Cold War conflict, escalating from around 1950 to the late 1960s and then again from the middle 1980s to the early 1990s – the periods labelled by some historians as the First and Second Cold Wars (Halliday 1983; Painter 2002). The American biomedical funding plateau corre- sponds to, but is 3–4 years longer on both ends than, the 1970s era of economically rational détente between these wars as usually defined. This parallel, and the possible role it suggests for biomedical research as another weapon in the Cold War(s), deserves further investigation by historians of science, medicine, and policy. As for describing the changing distribution of biomedical research effort in the United States over time, this is easily done (to a reasonable approximation) by tracking the budgetary history of the various NIH Institutes. A simple glance at Table 1 reveals, for instance, that was an especially favored area in 1960 and so too was heart disease. Both research fields remained generously supported through the subse- quent two decades, while mental health research appears to have lost favor between 1960 (when its share of resources exceeded heart disease research) and 1980 (when its share was less than half). These observations are not novel, although they still do bear further investigation by historians of US science policy. However, new historical questions could be opened if we could identify in some unbiased way those cases where research areas were especially over- or underfunded. This in turn would require, for comparison, some measure of unproblematic or politically rational funding distri- bution (not that rational political decisions require no explanation!). To enable this, in Table 2, I compare the relative contributions to the health burden of the United States, as reflected by national mortality and morbidity statistics, of the various disease groups, with the budget allocations of the several NIH Institutes dedicated to research on cognate biomedical problems. That the burden of disease should guide biomedical research expenditure proportionately is not only almost axiomatic for health policy today but has been generally accepted for the whole of the postwar period (e.g., among the Senators in the 1948 appropriation hearings to fund the freshly established National Heart Institute, as well as the Surgeon General Leonard Scheele).3 The task of correlating causes of death and

3Second Deficiency Appropriations Bill, Hearings Before the Subcommittee of the Committee on Appropriations, US Senate (80th Cong., 2nd sess.), on H. R. 6935 (Washington: GPO, 1948), 17 June 1948, p. 141 (testimony of Oscar Ewing and Leonard Scheele), pp. 141–143 (statement of Senator Claude Pepper). Table 2 Share of NIH budget, US disease burden, and historiography. (By Institute domain) 10 ABCDE F G H I NIH 1960 NIH 1970 NIH 1980 NIH 1990 NIH % total % total No. [% of No. [% of Institute Institute Institute Institute Institute mortality morbidity, included] included] 1960 name/ budget, budget, budget, budget, [cognate illness cognate illness cognate cognate 1990 name current current current current ICD codes], group, United historical historical $1000s [% $1000s [% $1000s [% $1000s [% United States States literature, literature, share] share] share] share] 1960//1980 1981–1982, 1990–2015, all 1990–2015, elite YPLL65 journal journal NCI/NCI 91,257 181,454 999,869 1,634,332 15.6%//20.9% 14.5% 1926 [12.7%] 50 [16.0%] [26.9%] [17.1%] [29.2%] [21.6%] [140–205] NHI/ 62,237 160,634 527,488 1,072,354 41.5%//40.9% 13.0% (heart 1404 [9.2%] 6 [1.9%] NHLBI [18.4%] [15.1%] [15.4%] [14.1%] [400–02, disease only) 442–47, 450] NIAMD/ 46,862 131,761 341,206 581,477 3.2%//3.7% 1.4% 894 [5.9%] 13 [4.2%] NIDDK [13.8%] [12.4%] [9.9%] [7.7%] [260, 592–94, 540–41] NINDB/ 41,487 97,315 241,966 490,409 11.3%//8.6% 2.1% 2643 [17.4%] 19 [6.1%] NINDS [12.2%] [9.2%] [7.1%] [6.5%] [330–34] NIAID/ 34,054 97,342 215,364 832,977 3.9%//3.2% 3.2% 1742 [11.5%] 74 [23.7%] NIAID [10.0%] [9.2%] [6.3%] [11.0%] [480–93] NIMH/ 67,470 NA 212,000 727,000 3.3%//2.9% 7.4% 4878 [32.1%] 128 [41.0%] ADAMHA [19.9%] [6.2%] [9.6%] [581, suicide] (research) NICHD/ 76,095 208,953 442,914 5.2%//1.9% 17.2% 1707 [11.2%] 22 [7.1%] NICHD [7.2%] [6.1%] [5.8%] [750–76//

740–79] Rasmussen N. Total 100%a 87.8%b 80.1% 76.3% 84.0%//82.1% 58.8% 15,194 312 aIn this calculation a value of $190 m is imputed to ADAMHA research, midway between the final NIMH appropriation of about $227 m for 1966 and the earliest ADAMHA research budget figure available to me of $153 m for 1977 (http://www.ncbi.nlm.nih.gov/books/NBK235734/table/ttt00015) bInstitute budgets may not sum exactly to 100% total NIH Budget, because the latter reflects only current appropriations and also includes central administration funds, while the Institute budgets reflect commitments and may include fund transfers. Furthermore, the 1960 total budget includes NIDCR, while the 1980 and 1990 budgets also include many other Institutes. It appears that NIH expenditures exceeded appropriations slightly in 1960 Biomedicine and Its Historiography: A Systematic Review 11 disability to NIH Institutes is not especially difficult, as the names of the Institutes correlate nicely with the coding of the disease groups used by the Centers for Disease Control (for long another branch of the Public Health Service) in describing the national vital statistics. For example, of the 15 leading causes of death for 1980, the second, malignant neoplasms (ICD classification codes 140–208 on death certifi- cates) correspond neatly to the brief of the National Cancer Institute; the third, cerebrovascular events (ICD codes 430–438, and no other neurological disease classes are listed among the top 15 causes) correspond neatly to the brief of the National Institute for Neurological Disorders and Stroke, while the first, heart diseases, combined with the fifth, chronic obstructive pulmonary disease, and the ninth, atherosclerosis (ICD codes 380–398, 402, 404–429; 490–496; and 440, respectively), correspond neatly to the brief of the National Heart, Lung and Blood Institute. As crude indicators of the mortality burden of the (psychiatric) conditions lying within the National Institute of Mental Health’s domain, I have, following others, combined suicide and liver disease, since alcoholism and other substance abuse are the latter’s major cause (morbidity is discussed separately below). Column F records the contribution to national mortality of the causes of death corresponding to NIH Institutes, and to offer some sense of how the causes of mortality are shifting through the postwar period, these are given both for 1960 and 1980. Some Institutes, such as General Medical Sciences and Dental and Craniofa- cial Research, do not correspond to major causes of death not otherwise recorded, but these are minor components of the NIH portfolio; indeed six of the Institutes (Cancer, Heart, Infectious Disease, Diabetes and Kidney, Neurology and Stroke, and Mental Health), account for approximately 100% of the NIH budget in 1960 and, with Childhood Development, still account for more than 75% of the budget in 1990, despite the addition of many of new Institute level budgetary units by that time and the splitting off of disease domains to new Institutes, such as the loss of Blindness from NINDS in 1968 and Arthritis from NIDDK in 1986. This analysis focuses mainly on these dominant branches of the biomedical enterprise (see Table 2, columns A–E). While mortality statistics, available through the causes of death recorded on death certificates, are a relatively accessible and politically potent measure of the relative social burdens of various diseases, a nation governed by the rationality of public health (as currently framed) would instead apportion its health research resources according to morbidity – that is, suffering and lost productivity due to illness, independent of death. Unfortunately, however, there is no single generally accepted measure of health burden. For example, the main measure used today in the United States, the DALY (disability-adjusted life year), differs from that used in much of the former British Commonwealth, the QALY (quality-adjusted life year). Even the standardized DALY measure specified by the WHO for international comparisons in recent Global Burden of Disease studies produces highly variable results (Polinder et al. 2012). Such measures also require detailed preference questionnaires, surveys, or other sociological data on particular health conditions that are not retrospectively obtainable. Indeed, morbidity measures were first explored only in the late 1930s, largely by public health advocates for increased attention to chronic diseases, and 12 N. Rasmussen remained contentious and experimental through the 1960s (Weisz 2014). The first National Health Examination Survey, the earliest baseline against which subsequent US health status trends can be judged, was only conducted in 1959–1962.4 National morbidity figures have only been routinely published since 1981 and are based on a simple measure calculable from the same death certificate data as mortality, the YPLL65 (years of productive life lost before age 65; age 75 from 1996). Although it also fails to assess impaired health not associated with shortened life expectancy, YPLL does improve on raw mortality as a health burden measure by accounting for the age of victims. Because nothing better is available, and also because they represent the health policy decision base circa 1980 (the approximate midpoint of the postwar twentieth century), I present YPLL65 US morbidity figures for 1981 in column G of Table 2 as a second rough measure by which the rationality of the distribution of health research funds may be assessed. Columns H and I refer to the distribution of historical research efforts and will not be discussed until the next section, after we have concluded our analysis of the match between biomedical research effort and the burden of cognate illness categories. Taking each described budget year as a rough measure of funding distribution for the decade in which it stands as midpoint, these figures cover the period from 1955 (before which funding patterns had not stabilized) to 1995 – essentially the entire postwar twentieth century. Therefore we may draw some broad conclusions about the postwar distribution of biomedical research efforts based on the budget shares of the seven NIH Institutes described in columns A–E in Table 2, as compared with the shares in national mortality and morbidity of the corresponding illnesses in columns F–G. For example, as many historians have noted, cancer research claimed a large and fairly stable share of research resources throughout the postwar period (the dip in 1970 is more apparent than real, as it immediately preceded a major boost from the National Cancer Act of 1971). However this 27% (1960) or 29% budget (1980) share was not dramatically far out of step with the illness’s impact on health (about 16% of national mortality in 1960 and 21% of mortality in 1980 but 15% of morbidity). Heart disease also attracted a consistently large share of research resources, and likewise this large share is easily explained by the condition’s visibly large health burden. Indeed, one would be justified in asking why in 1960 (before morbidity was formally considered, and mortality dominated the epidemiological visibility of disease), heart disease research attracted only 60% of the funding won by cancer research, despite killing more than twice as many Americans. But asking why heart disease was relatively underfunded is nearly the same question as asking why cancer research attracted particular favor, because these two research areas together accounted for nearly half of the biomedical research enterprise at the time. On the other hand, research into metabolic disorders such as diabetes and kidney disease (whose chief cause is diabetes) seems to have enjoyed even greater relative favor in 1960, when the research share of the fields under NIAMD aegis was fourfold higher than their cognate mortality share, and 1980, when what had become

4Data available at https://www.icpsr.umich.edu/icpsrweb/ICPSR/series/00197 Biomedicine and Its Historiography: A Systematic Review 13

NIDDK enjoyed a budget share two and a half times its mortality share (and seven times morbidity share, although YPLL does not capture arthritis). This analysis suggests that some historical explanation is needed to account for the flourishing of this field, which is dedicated to the main chronic diseases of aging other than cancer and heart disease, but to my knowledge no historians have interrogated it. As I argue below, this area of biomedicine is in in fact radically understudied by historians. We might also observe that infectious disease appears to have commanded a disproportionately high share of research funding as compared with mortality and also morbidity (in 1980), but that this share was declining until the 1980s – a reversal due to an influx of new AIDS research funding, one presumes as for neurology and neurological disorders, this biomedical field seems to have attracted a share of research funding in both 1960 and 1980 that is very close to the mortality burden attributed to cerebrovascular disease at both times, which suggests both that mortality was a key determinant of health research priorities in both 1960 and 1980, as expected, and also that there is no very interesting problem to study in why this field’s research funding share declined 42% over the 20-year interval (because cognate mortality declined proportionately). Still, that morbidity was starting to play some role in attracting health research funding by 1980 is implied by the National Institute of Child Health and Development budget share of about 6%, three times the mortality contribution of congenital and early childhood illness – but still only one-third the contribution to morbidity at the time. The emerging role of formal morbidity assessment in the establishment and funding of this Institute in the early 1960s, together with the other arguments offered for funding research in perinatal medicine and childhood development, is another worthwhile but largely unexplored question for historians of biomedicine. Mental health research funding presents special problems of assessment. At face value it would seem that the National Institute of Mental Health was grossly overfunded in 1960, when it accounted for one-fifth of the total NIH budget, double the budget of the National Institute of Allergy and Infectious Diseases (whose cognate illnesses had a similar contribution to mortality) and even somewhat more than the National Heart Institute (whose disease domain claimed more than ten times the mortality share attributable to mental illness, even excluding the cardiopulmo- nary causes of death that did not officially fall within its portfolio before its 1969 reorganization as the NHLBI). By contrast in 1980 NIMH had seen a 70% decline in budget share or more precisely in the research budget of the mental health agency to which NIMH then belonged (and which would again spin off NIMH in 1992, along with the National Institute of Drug Abuse and National Institute on Alcohol Abuse and Alcoholism, formerly programs within NIMH), from six times the national mortality share of suicide and substance abuse to about two and a half times that mortality share. It would seem that this branch of research had lost the political favor it had once enjoyed in the 1950s and early 1960s and by 1980 was attracting research effort more proportional to mental health’s contribution to the national health burden. Advocates of mental health research (and psychiatry in general) have long argued, of course, that the health burden of mental illness far exceeds what is visible in mortality-based statistics; indeed, at the height of early postwar enthusiasm for all 14 N. Rasmussen things psychiatric, Surgeon General Parran urged Congress to establish the NIMH on the grounds that “mental diseases equal all physical diseases in subtracting from the total vigor, the total fitness of our population” (Grob 1996). Yet as noted earlier, no national morbidity data were available at the time to support such a claim, although it was a widely recognized problem that around half of the national hospital beds were occupied by long-term mental health patients. Even today the measures of health burden that elevate the morbidity profile of mental health, and have been criticized for exaggerating its contribution to total burden such that antidepressant drugs may be displacing more cost-effective public health interventions, still tend to place anxiety and depression in fourth place, far behind heart disease and at about the same level as metabolic disorders (in developed countries) (Brhlikova et al. 2011; Polinder et al. 2012; Bemme and D’Souza 2014). In sum, historians have good reason to investigate why mental health attracted so much cultural attention and political favor in early postwar United States. They also have good reason to ask how and why the field lost favor in the late 1960s and 1970s, a question much less thoroughly addressed. To sum up, in this section we have examined the shifting distribution of biomed- ical research funding, and therefore (approximately) research effort, among the major fields of biomedicine, and also the relationship of research effort to national burden of disease. This comparative exercise has been fruitful in raising new hypotheses and problems for investigation by historians of biomedical science as well as US research policy. We have confirmed the observation of many historians that cancer appears to have attracted disproportional support. We have raised the more novel question of why heart disease, responsible for three times as many deaths, never achieved the same national attention and resources. We have pointed out that the cluster of diseases associated with the other main chronic diseases of aging (diabetes, kidney disease, and arthritis) appear to have, on the other hand, garnered a disproportionately great share of resources and suggested this as an area for historical investigation. We have also suggested that a similar, large over- investment in mental health research occurred in the early postwar period, as well dramatic decline by the end of the 1960s, both meriting further historical investiga- tion. We have noted that funding for infectious disease research was disproportion- ately large in the postwar period and that its share was in decline until the 1980s, suggesting that historians might investigate the mobilization of resources in reaction to AIDS to explain this reversal. We have also suggested that the emerging role of morbidity burden, as opposed to mortality, as a metric used in research policy decision-making from the 1970s, is another topic deserving historical investigation.

The Distribution of Activity in Biomedicine and in Its Historiography

Much as we can compare the budget shares of various NIH research areas to the shares of national morbidity and mortality claimed by cognate disease groups, in order to identify important issues for the history of health research policy, we can Biomedicine and Its Historiography: A Systematic Review 15 compare NIH research area budget shares to the share of the historical literature devoted to biomedical activity in the same cognate disease groups. That is, we may ask about the distribution of the efforts of historians of biomedicine – historiographic “mindshare” (if I may) – as compared with the distribution of past activity within the biomedical enterprise itself. Taking the share of the budget claimed by each NIH Institute as a tangible (if still approximate) measure of postwar American biomedical research activity in that scientific domain is not hard to defend, given (1) the way NIH dwarfed by an order of magnitude other sponsors in the main biomedical research fields it funded and (2) the indisputable premise that research activity within modern biomedicine correlates (as a first approximation anyway) with the quantity of money in a field available to pay for salaries, facilities, supplies, and graduate fellowships. This measure can be extrapolated internationally to some extent – albeit one not assessable on the basis of current historical literature – given that US biomedical research spending throughout the postwar twentieth century represented the bulk of world spending (as noted above). The other side of this comparison, measuring the distribution of historians’ efforts among biomedical research fields, is a more complex matter and one without much precedent in the historiography. However, systematic literature reviews have become a commonplace method in biomedical research in recent years. I have adopted such an approach to study the historiography of biomedicine, inspired not just by the attractions of reflexivity but also by the impressive body of “evidence-based medicine” studies that assess questions of bias in the clinical trial literature (e.g., the degree to which, ceteris paribus, published clinical trials sponsored by the manufacturer of a novel drug more often report significant benefits for the new drug as compared with independent trials) (Lundh et al. 2012). Mine can be thought of as an analogous project to measure bias in historical literature. The foundation of any systematic review is an extensive and representative database of the relevant field of literature, together with a sampling method that is unbiased with respect to the question at hand. To identify recent historical publica- tions on biomedical research fields, in June 2015 I searched the EBSCO History of Science, Technology and Medicine database established by the History of Science Society and Wellcome Trust, limiting searches to journal articles published since January 1, 1990 (mainly because before this date journal coverage in the database is very incomplete and also because little of the postwar period was remote enough to be treated as history before the 1990s). The search terms, allowed anywhere in the bibliographic entry, were “Research” and any of five other terms distinguishing the NIH research area in question. These five search terms were derived iteratively from the mission statement on each relevant NIH Institute web page, exploring close cognates and variants to achieve the maximum number of total hits for the com- pound search. British and American spelling variants were both used, together counting as only one of the five search terms; for example, historiography concerning the National Cancer Institute’s research domain was identified with search terms Research AND Cancer OR Neoplasms OR Tumor OR Tumour OR Oncology OR Hematology OR Haematology, while that of the National Heart (Lung and Blood) Institute was identified with terms Research AND Cardiology OR 16 N. Rasmussen

Cardiovascular OR Cardiac OR Cardiopulmonary OR Heart.5 Boolean mode was employed, no language was specified (although the database is mainly Anglophone), and the automatic related term search function was deactivated. The number of hits for each Institute-cognate biomedical research domain is given in column H of Table 2, together with percent of total hits in searched domains – although the latter should not be regarded as a reliable measure of the total literature because the database contains many duplicate entries, and there was no effort to restrict hits to sources actually dealing with biology or medicine. It must also be considered that at least as many of the historical items retrieved in these searches were short retrospec- tive items published in biomedical research journals, typically written by practi- tioners in these fields, as were scholarly articles by professional historians. While such auto-historiography of biomedicine merits investigation in its own right, it lies beyond the scope of the present review. To assess biomedical research area mindshare in the professional research liter- ature for history of science and medicine, I searched within the total hits retrieved above by limiting results to a subset of elite scholarly journals. This elite subset, intended only as an unbiased sample of the professional historiography, was derived from the journals ranked in the top half (of 59) categorized in the field history and philosophy of science, according to the 2013 Thomson-Reuters Web of Science Journal Citation Index (which was the latest available at the time). Only regularly publishing journals with a scope including all of biology or medicine were included (thus excluding Journal of the History of Neurosciences, as well as Annals of the History of Computing) and also specifying history as their central purpose (thus excluding the American Journal of Bioethics and, perhaps regrettably, Social Studies of Science). While somewhat arbitrary, the eight included journals do represent a plausible sample of the professional historiography of biomedicine: Bulletin of the History of Medicine, British Journal for History of Science, Isis, Journal of the History of Biology, Journal of the History of Medicine and Allied Sciences, Medical History, Social History of Medicine, and Studies in History & Philosophy of Science (Part C). This elite subset happened, conveniently, to be picked out by limiting the much larger overall search results in each domain to items with the term “history” in the publication title/subtitle; these results then were manually searched. Articles specified as book reviews or introductions (with no further designation of historical subject matter) were excluded and also any item shorter than five pages. All duplicate records were eliminated. No article was included unless it specified “20th century” in its keywords or clearly referred to events in the twentieth century in its title or keywords (excluding birth and death dates). In the very few cases that an

5The other search term clusters were as follows: Mental Health, Research AND Psychiatry OR Behaviour OR Behaviour OR Mental Health OR Psychopharmacology OR Insanity; Kidney and Metabolic Disease, Research AND Diabetes OR Kidney OR Metabolic OR Gastrointestinal OR Arthritis; Infectious Disease, Research AND Immunology OR Bacteriology OR Virology OR Parasitology OR Allergy; Neurology, Research AND Neurology OR Brain OR Stroke OR Spinal OR Cerebral; Childhood Disease, Research AND Childhood OR Neonatal OR Pediatric OR Pediatric OR Congenital OR Embryology. Biomedicine and Its Historiography: A Systematic Review 17 article was listed in multiple research domain searches, it was allowed to count for each. The number of hits for each Institute-cognate biomedical research domain in the elite journals is given in column I of Table 2, together with the percentage share of total elite items identified in all domain searches. A number of conclusions emerge concerning the distribution of effort in the twentieth-century biomedical historiography as compared with distribution of research effort within postwar biomedicine itself. Restricting myself to cases of pronounced mismatch, where the share of historical attention is more than double or less than half the share of a biomedical field’s research activity in the past, these include the following. First, the historiography of biomedicine pays attention to psychiatry and related fields to a degree far exceeding their importance in past biomedicine. Even compared with the early postwar period, when the biomedical fields associated with the National Institute of Mental Health exhibited special dynamism thanks to an excessive (I have suggested) 20% share of health research funding, the historiography redoubles this favor with a surprising 42% of the elite journal literature. That historians pay special attention to mental health research of the past may owe something to the relative accessibility of the field’s content to scholars trained in the humanities or to the closeness of the field’s content to themes subsequently fashionable in the humanities (e.g., social control, deviance, race, gender construction). Probably both. Explaining such disproportionate attention among historians is an intriguing topic for future reflective research, although not as important in my view as correcting it with new empirical work elsewhere. Second, and similarly, the historiography of biomedicine pays exaggerated attention to fields associated with infectious disease. Microbiology at the end of nineteenth century could not be more worthy of historical attention. As historians have emphasized time and again, the successes of this field of lab research not only revolutionized thinking and practice in medicine but also catalyzed the wider convergence of the two as biomedicine during the middle third of the twentieth century. However, the bacteriological revolution was already victorious by the first decades of the twentieth century; for example, between 1900 and 1938, the US national death rate from tuberculosis declined by three quarters and that from pneumonia two-thirds, before any impact attributable to antibiotics.6 That infectious disease research, including virology and allergy as well as bacteriology, still accounted for a much greater share of the postwar US biomedical research budget than can be justified by its contribution to national mortality can be plausibly explained by a number of mutually reinforcing explanations. For example, the penicillin-driven retreat of bacterial infection as a feared cause of death after the Second World War inspired ambitions that all remaining infectious diseases, such as those due to viruses, could similarly be cured (and not just prevented). Also polio, that dreaded scourge of children, was almost eliminated from American experience,

6Forrest Linder and Robert Grove, Vital Statistics Rates in the United States, 1900–1940 (Federal Security Agency; GPO 1947), available at http://www.cdc.gov/nchs/data/vsus/vsrates1900_40.pdf; accessed 2 July 2015); Mortality Tables: Table 12. 18 N. Rasmussen practically overnight. At the same time, the simplicity of bacteria and viruses made them attractive experimental objects for more basic disciplines like biochemistry and the emerging, glamorous field of molecular biology, which achieved some spectac- ular scientific successes. Further, immune disorders became an intellectually fash- ionable field, even if not responsible for a large burden of disease – partly due to metaphoric relevance to Cold War problems of political control, according to some historians (Rasmussen 1993; Anderson and Mackay 2014). But a plausible hypoth- esis about why historians should redouble past popular and political enthusiasms for microbiology-related research is less obvious.7 Perhaps some of this interest is driven by the recognition of historians in the late twentieth century that the period of freedom from fear of infectious disease was transitory, since it would soon return as a prominent problem in the developed world (e.g., with AIDS and mounting bacterial drug resistance), while less developed countries never achieved as much benefit from postwar advances. The roots of pressing social and medical problems of the present ought quite properly to attract extra attention from historians. Another, somewhat less charitable explanation is historiographic momentum: topics already attracting attention from historians tend to exert a continuing attraction to historians, because training and previous work by established historians instructs budding historians and shapes their consciousness of the past. Taken too far, this process can be likened to mapmakers relying only on old maps to draw new ones, as in Scholastic cartography. Third, the biomedical research fields centered on heart disease, corresponding to the NHLBI’s remit, are drastically understudied relative to their importance as activities within postwar biomedical research. At first thought, it seems hard to explain the relative neglect of heart disease by historians since it was an area of research that not only expanded so much in the postwar era that it soon rivalled cancer research in terms of resource allocation but also attracted enormous glamor, for example, in spectacular cardiac techniques and blockbuster pharmaceu- ticals modifying the heart’s action (e.g., beta blockers) and blood properties (e.g., pressure, clotting, fat profile). Here perhaps the likeliest explanation is a type of historical accident that we may call “inertia,” the flip side of what I called “momen- tum” above: historians often take up subjects close to their doctoral supervisors’ expertise, but those few who have studied heart disease have for whatever reason trained fewer students who went on in the field. And why were there few senior historians of medicine in the heart disease field to begin with? Here it may be that the relative neglect of heart disease within medical research before the Second World War has had a follow on effect upon historians of medicine. Thus, in the 1960s and 1970s, there was as yet not much heart disease research old enough to qualify as

7It might be supposed that historians have mostly attended to the early twentieth-century golden age of bacteriology, when it achieved its most dramatic successes. However, it seems this is not the case: the majority (38/74) of elite journal articles returned by the searches in the review dealt substantially with events after 1945. My takeaway impression is that the center of gravity of the retrieved literature in this area lies in the 1930s and is motivated by questions about how postwar biomedicine took the shape it did. Biomedicine and Its Historiography: A Systematic Review 19 history – certainly not much compared with microbiology, still in its heyday with the afterglow from penicillin, the polio vaccine, and earlier conquests of infectious disease (and long established as historical genre of the Microbe Hunters ilk). The 1970s were when many of the senior generation of historians of science and medicine, now in or entering retirement, wrote their dissertations – and the period when the field was expanding and many academic positions were filled at research universities. Thus I would suggest in the realm of our disciplinary ecology what population biologists call a “founder effect,” perhaps coupled with incidental causes that propagated rather than corrected the historiographic neglect of the heart over time. Fourth, the fields of metabolic disorders, kidney disease, and diabetes, corresponding to the NDDK’s remit, are also drastically understudied relative to their importance as activities within postwar biomedical research. In the case of kidney disease and metabolic disorders (a cluster of conditions associated with chronic diseases of aging), we might venture that, in addition to the founder effect and inertia among historians just discussed, there may operate another effect based in professional formation. Just as historians may seek to share in the glory of especially glamorous past events by becoming authorities on them, they may fear obscurity by studying areas of science and medicine forgotten by the public and neglected by fellow historians (whose opinion, after all, determines the bulk of professional reward). This last hypothesis, attraction to glamorous and still memorialized past events, may of course apply to any marked case of selective historical attention and historical neglect, not just infectious disease and kidney disease. Why the introduc- tion of the polio vaccine drew so much more acclaim in the past than renal dialysis, and left a more lasting aura of glamor, is another intriguing question worthy of historical investigation. Here we may begin to consider the role of professional historians in managing society’sofficial memories of science and medicine, some- times unthinkingly accepting a script written by past journalists as a starting point. But we can only ask such questions, which arise from comparing public and professional commemoration (both past and present) with some independent mea- sure of importance, if we stray from the well-worn paths left by the existing historiography and look afresh at the landscape we purport to be describing.

Conclusion

There is nothing inherently wrong with historians researching the topics most frequently studied by other historians. To some extent this is inevitable; as just noted, young historians often adopt the study areas of their mentors as part of their apprenticeship, giving rise to momentum (and, as a side effect, inertia). Furthermore it is vital to the spirit of historical scholarship that the stories arising from existing historical writings, popular and professional, be subject to scrutiny and evidence- based reinterpretation –“revisionism,” to those uncomfortable with scrutiny of convenient myths. However, there is something wrong if we do nothing more, treading ever-deeper historiographic ruts fixed on our crude inherited maps, while 20 N. Rasmussen oblivious to vast, largely unexplored parts of past biomedical terrain lying nearby. It is not my intention to suggest that too much history of microbiology and history of psychiatry has already been done, and sufficient research. The amount of biomedical activity in the recent past so vastly dwarfs the activity of its historians – think of the ratio of living biomedical researchers to historians of biomedicine! – that even on many relatively well-studied topics we have only scratched the surface. Even so, I do hope that the large lacunae on our maps of the past, highlighted in this review, will tempt some adventurous historians of biomed- icine into less explored territory.

References

Agar J (2012) Science in the 20th century and beyond. Polity, London Anderson W, Mackay IR (2014) Intolerant bodies: a short history of autoimmunity. Johns Hopkins University Press., Baltimore Appel T (2003) Shaping biology: the National Science Foundation and American biological research, 1945–1975. Johns Hopkins University Press, Baltimore Bemme D, D’Souza NA (2014) Global mental health and its discontents: an inquiry into the making of global and local scale. Transcult Psychiatry 51:850–887 Bero LA (2005) Tobacco industry manipulation of research. Public Health Rep 120:200–208 Bloom FE, Randolph MA (eds) (1990) Funding health sciences research: a strategy to restore balance. National Academies Press, Washington, DC. Available at http://www.ncbi.nlm.nih. gov/books/NBK235742/ Boychuk G (2008) National health insurance in the United States and Canada: race, territory, and the roots of difference. Georgetown University Press, Washington, DC Brandt AM (2012) Inventing conflicts of interest: a history of tobacco industry tactics. Am J Public Health 102:63–71 Brhlikova P, Pollock AM, Manners R (2011) Global burden of disease estimates of depression–how reliable is the epidemiological evidence? J R Soc Med 104:25–34 Brown R (1979) Rockefeller medicine men: medicine and capitalism in America. University of California Press, Berkeley Clarke AE, Mamo L, Fosket JR, Fishman JR, Shim JK (2009) Biomedicalization: technoscience, health, and illness in the U. S. Duke University Press, Durham Forman P (1992) Inventing the maser in postwar America. Osiris 7:105–134 Fox DM (1987) The politics of the NIH extramural program, 1937–1950. J Hist Med Allied Sci 42:447–466 Fuller S (2000) Thomas Kuhn: a philosophical history for our times. Press, Chicago Gaudillière J-P (2002) Inventer la biomedicine: La France, l’Amérique et la production des savoirs du vivant (1945–1965). Éditions la Découverte, Paris Geiger R (2008) Research and relevant knowledge: American research universities since World War II. Transaction, New Brunswick Grob G (1996) Creation of the National Institute of Mental Health. Public Health Rep 111:378–381 Halliday F (1983) The making of the Second Cold War. Verso, London Hamowy R (2008) Government and public health in America. Edward Elgar, Northampton Hollinger DA (1995) Science as a weapon in Kulturkampfe in the United States during and after World War II. Isis 86:440–454 Kevles DJ (1977) The National Science Foundation and the debate over postwar research policy, 1942–1945: a political interpretation of science – the endless frontier. Isis 68:5–26 Biomedicine and Its Historiography: A Systematic Review 21

Kleinman D (1994) Layers of interests, layers of influence: business and the genesis of the national science foundation. Sci Technol Hum Values 19:259–282 Kohler R (1991) Partners in science: foundations and natural scientists, 1900–1945. University of Chicago Press, Chicago Krementsov N (2002) The cure: a story of cancer and politics from the annals of the Cold War. University of Chicago Press, Chicago Krige J (2006) American hegemony and the postwar reconstruction of science in Europe. MIT Press, Cambridge, MA Ludmerer K (1985) Learning to heal: the development of American medical education. Johns Hopkins University Press, Baltimore Lundh A, Sismondo S, Lexchin J, Busuioc O, Bero L (2012) Industry sponsorship and research outcome. Cochrane Database Syst Rev (12):Art. No.: MR000033. https://doi.org/10.1002/ 14651858.MR000033.pub2 Mayes R (2004) Universal coverage: the elusive quest for national health insurance. University of Michigan Press, Ann Arbor Moses H 3rd, Matheson D, Cairns-Smith S, George B, Palisch C, Ray Dorsey E (2015) The anatomy of medical research: US and international comparisons. JAMA 313:174–189 Osgood K (2006) Total Cold War: Eisenhower’s secret propaganda battle at home and abroad. University of Kansas Press, Lawrence Painter D (2002) The Cold War: an international history. Routledge, London Poen MM (1996) Harry S. Truman versus the medical lobby: the genesis of medicare. University of Missouri Press, Columbia Polinder S, Haagsma JA, Stein C, Havelaar AH (2012) Systematic review of general burden of disease studies using disability-adjusted life years. Popul Health Metrics 10:21. https://doi.org/ 10.1186/1478-7954-10-21 Porter R (1998) The greatest benefit to mankind: a medical history of humanity. Norton, New York Proctor R (2011) Golden holocaust: origins of the cigarette catastrophe and the case for abolition. University of California Press, Berkeley Rabinow P, Rose N (2006) Biopower today. BioSocieties 1:195–217 Rasmussen N (1993) Freund’s adjuvant and the realization of questions in postwar immunology. Hist Stud Phys Biol Sci 23:337–366 Rasmussen N (2014) Gene jockeys: life science and the rise of biotech enterprise. Johns Hopkins University Press, Baltimore Rothstein W (1987) American medical schools and the practice of medicine : a history. Oxford University Press, New York Saunders FS (2000) Who paid the piper?: the CIA and the cultural cold war. Granta, London Shapin S (2008) The scientific life: a moral history of a late modern vocation. University of Chicago Press, Chicago Shyrock R (1980) American medical research past and present. Arno, New York Starr P (1982) The social transformation of American medicine. Basic, New York Strickland S (1972) Politics, science, and dread disease: a short history of United States medical research policy. Press, Cambridge, MA Strickland S (1989) The story of the NIH grants programs. University Press of America, Lanham Sunder Rajan K (2006) Biocapital: the constitution of postgenomic life. Duke University Press, Durham Vogel MJ, Rosenberg C (eds) (1979) The therapeutic revolution: essays in the social history of American medicine. University of Pennsylvania Press, Philadelphia Wiesz G (2014) Chronic disease in the twentieth century a history. Johns Hopkins University Press, Baltimore Wilford H (2009) The mighty Wurlitzer: how the CIA played America. Harvard University Press, Cambridge, MA Yi D (2015) The recombinant university: genetic engineering and the emergence of Stanford biotechnology. University of Chicago Press, Chicago