A Capsule History of American Research in the Life Sciences
During the latter half of the 20th century, the United States has witnessed substantial growth in the size and effectiveness of its life-science research enterprise. Indeed, the very definition of life science has emerged during this century as the sum of agricultural, biochemical, cellular, developmental, ecologic, evolutionary, molecular, and medical biology. The National Institute of Health was established by the Ransdell Act in 1930 (PL 71-251), but during the 1930s life-science research in university and industry laboratories was conducted with little support from the government. The US Department of Agriculture (USDA) was the only source of federal support for such work. The National Cancer Institute (NCI) was established in 1937, but although its mandate included the funding of research and training in nonfederal laboratories, its expenditures for medical research in 1940 were only $3 million, including both intramural and extramural work. Meanwhile, private sources, such as the Rockefeller Foundation, contributed $17 million, and industry $25 million (NIH 1961). In 1944, Congress pluralized the National Institutes of Health (NIH) to include several disease-oriented institutes in addition to NCI, but at no time between 1938 and 1945 did NIH extramural expenditures exceed $250,000 (NIH 1978).
In the period before World War II, the number of life scientists trained per year was also low; in 1930, only 342 PhDs were awarded in all the life sciences. By 1940, however, change was in the air: Warren Weaver, of the Rockefeller Foundation, noted that "gradually there is coming into being a new branch of science—molecular biology—which is beginning to uncover many secrets … of the living cell" (Judson 1979), and the number of life-science PhDs awarded was 672. It was, however, the events during and after World War II that had the greatest effect on the climate of life-science research. The pressing problems of wartime required solutions on an unprecedented scale. Whole armies became ill with malaria, and drugs for the treatment of infection and trauma were needed in massive amounts. Rates of food production became an issue of international importance. For the first time, life scientists were mobilized on a broad front and given abundant resources with which to tackle the fundamental and practical problems of biology; and both medical and agricultural problems were solved. The successes of those efforts and of comparable work in other fields of science gave credibility to the idea that the entire United States could benefit from institutionalized support for research, as propounded in the 1945 report by Vannevar Bush, Science, the Endless Frontier (NSF 1960).
The postwar years saw the establishment of the National Science Foundation (NSF) and an expansion of NIH. By 1947, the government was investing $28 million per year in medical research, 9 times the investment of 7 years earlier and approaching industry's $35 million. By 1960, NSF was spending $29 million on biologic and medical sciences. From 1956 to 1961, NIH expenditures for extramural research went from $40.5 million to $272.9 million; during the same period, NIH investments for training grew from $17.3 million to $132 million, proportionally an even larger increase (NIH 1961), so funds for training kept pace with support for research. Indeed, an important consequence of Bush's blueprint for federal investment in science was the establishment of a linkage between research and research training. It was a natural consequence of the policy that federally supported research would be conducted primarily
in university-based research laboratories. As the funds for research increased in the postwar years, the number of life-science PhDs granted per year grew correspondingly—from 1,660 in 1960 to 4,980 in 1971, tripling in only 10 years.
Those patterns of government investment had profound effects on both the number and the structure of US universities. Building on the foundations established by the early research orientation of Johns Hopkins University and the expansion of academic medicine, as initiated by the Flexner report (Flexner 1910), the influx of federal support for research helped to change American universities into research-intensive institutions. For example, training was seen as part of the mission of NCI from its beginnings in the 1930s. Recodification of the Ransdell Act during 1944 reauthorized the training activities specified in the act. The training of scientists at the master's and PhD levels became an integral part of research. As new national institutes came into being, the authority for training—research or clinical—was often included as an essential component of their missions and incorporated into their statutory portfolio, as specified in Title IV of the Public Health Service Act. Funds to support the tuition and stipends of students and fellows were now often included as items in the budgets of federal research grants. By the early 1950s, NIH had administratively crafted an elaborate set of training mechanisms, including grants for predoctoral, postdoctoral, and special fellowships and for predoctoral and postdoctoral training; these supported a wide variety of training programs in the biomedical sciences.
The most general and comprehensive statutory authority for supporting research training was added to Section 301(d) of Title III of the Public Health Service Act by an amendment enacted in 1962 as part of PL 87-838. The amendment extended the limited authority of the surgeon general (later the secretary) from supporting simply "such research projects as are approved by the National Advisory Health Council" to supporting "such research and research training projects as are approved …" By the early 1970s, more than 6,000 life-science graduate students were supported by NIH and NSF training grants or fellowships. The National Research Act of 1974 (PL 93-348) established the National Research Service Awards program, providing funds for competitive individual fellowships for graduate students and postgraduate fellows. It also instituted a mechanism by which a committee appointed by the National Academy of Sciences met every 2 years to identify current national research training needs (NRC 1994). The new mechanism led to the termination of some training grants, but the general level of support for biomedical training continued to grow. The sums spent for life-science research training continued to mirror those spent for life-sciences research, as exemplified by the transient drop in the number of PhDs granted per year during the middle to late 1970s, which followed a temporary cessation in the rapid growth of research funding that occurred during the late 1960s. When federal research investments resumed growth in the middle 1970s, the rate of PhD production followed suit. The expansion of training has continued at various rates ever since, as detailed in chapter 2.
The growth of the life sciences has permitted the absorption into the research workforce of a large fraction of the ever-increasing trainees. The ready availability of recent PhDs has also contributed to the success of companies built on the life sciences, such as in the biotechnology industry. Scientists needed to guide company decisions and workers to staff research laboratories were already available when the discoveries of recombinant DNA in the 1970s empowered entrepreneurial scientists to develop processes that would make marketable products of an unprecedented kind. Human proteins could now be synthesized in large quantities outside the human body and used as therapeutic agents of great practical utility. During the 1980s, this
industry grew rapidly, fueled in part by the enthusiasm of Wall Street for the possibilities associated with new markets. New investment from the private sector flowed quickly into the life-science enterprise, increasing both the quantity of scientific research and the perception that such work could be of value to the American people. In 1996, the number of life-science PhDs granted was 7,696; in 1997, federal investment in health research exceeded $14 billion. Private foundations contributed $1.2 billion to biomedical research in 1997, and industry's investment in health research and development exceeded $17 billion (NSF 1996, appendix table 4-31). Meanwhile, the country's investments in plant science and agriculture had also grown: during 1995, USDA invested $1.4 billion in research and development, and industry's investment in agriculture and forestry was $3.5 billion. The life-science research enterprise had become economically important.
In the recent decades, the various sectors of employment for life scientists have expanded at different rates. The fastest growth has occurred in industry, where the number of life-science PhDs has increased from around 5,500 in 1973 to nearly 24,000 in 1995, an average annual increase of almost 7%. During the same period, the pool of postdoctoral fellows and non-tenure-track staff at academic institutions has grown from about 4,000 to over 20,500, an average annual increase of 7.6%. In contrast, federal-laboratory and other government employment has shown modest growth; and the number of life scientists holding faculty appointments in universities and colleges has increased from 28,500 in 1973 to only about 49,000 in 1995, an average annual increase of only 2.5%. Universities remain the largest employers of life-science PhDs, but their share of the pool has diminished substantially during the last two decades (see appendix table F.8 for details).
Our country's investment in the life sciences has produced many important results. Discoveries in agricultural science have improved our understanding of soils and their chemistry and have led to the development of new strains of crop plants that are resistant to diseases and that yield more food per cultivated acre. Such work has contributed to the low cost of food that our country now enjoys. Environmental sciences and forestry have evolved new methods for sustainable managing resources that will help our expanding population to pass on more of its natural wealth to future generations. Medical science has provided fundamental understanding of the molecular basis of numerous diseases, which has led to the elimination of some and the containment of many. Not only preventive approaches, like proper nutrition and immunization, but diagnostic techniques and ameliorative treatments—drugs, surgery, radiation, and physical therapy and psychotherapy—have reduced human suffering and prolonged and enriched human life. Advances in molecular biology not only have spawned the biotechnology industry, which is contributing to the American economy, but also have contributed fundamental knowledge about the structures of genes and the behavior of biologic macromolecules. These advances are yielding new insights into the relationships among organisms and about the continuum of structure and function that connects living and nonliving things. (For more specific examples of the fruits of progress in the life sciences, see chapter 4.) The long-range implications of all this rapidly evolving knowledge are hard to predict, but many additional benefits are now on the horizon.
The Structure of the Life-Science Enterprise
The spectacular successes of the life sciences have emerged from a professional structure that evolved to meet the needs stemming from rapid growth. The lives of professors, industrial biologists, agricultural and medical researchers, postdoctoral fellows, and graduate students in the
1990s very different from those of comparable scientists 30–40 years ago. A typical academic research laboratory in earlier times included a professor, perhaps a technician, and sometimes a graduate student. Today, many life-science laboratories include 20 or more people, most of whom are in the process of training to become independent scientists. The chapters that follow present data on many aspects of the changes. To make the later chapters more meaningful for readers who are not themselves life scientists, we describe here the training of a life scientist and the major professional events in a life scientist's career—the work toward a PhD, in many cases postdoctoral training, the passage to a job, and the pursuit of research support—and then sketch the research environment. Space limitations require that this treatment be brief, so it is restricted in scope and detail; the descriptions are intended not to be detailed, but to illustrate what it is like to be trained and to work in today's biologic research enterprise.
It is important first to recognize the breadth of knowledge that is now encompassed by the term life sciences. At one extreme, we find physical and chemical studies of the molecules that make up living things: organic molecules—such as fats, carbohydrates, and proteins—that are the stuff of which all living things are made. The life sciences then range up through the study of genes and of the DNA and RNA from which they are constructed and expressed to studies of macromolecular assemblies and organelles and the cellular processes that they accomplish. Cells are sometime studied as organisms in their own right (for example, bacteria, protozoa, and some fungi) and sometimes as components of multicellular plants or animals, which must in turn be analyzed not only as organisms, but also as entities that develop from a single fertilized ovum and must interact with other plants and animals in their environments. Whole systems of interacting organisms must be studied to understand an ecologic niche. And the evolutionist would argue that none of the above studies makes sense unless viewed in the context of the slow changes in genetic makeup that constitute biotic evolution. All those aspects of the life sciences are linked by the universality of the genetic and biochemical bases that underlie them, but it is clear that there are many ways to study the complexities of life.
The life sciences can be thought of in three categories: the agricultural sciences, the biomedical sciences, and a harder-to-label cluster of basic biologic sciences that address life processes themselves. This report includes data from all those categories, and we have tried to address the interests of every federal agency that supports training and research in biology, broadly defined. It might appear at times that NIH and the biomedical sciences have dominated our considerations. That appearance has been difficult to avoid because of the size of the NIH budget and the resulting number of young and established life scientists that it supports. Indeed, patterns of support that are initiated by NIH often serve as models for programs funded by other agencies. We hope that our discussions and recommendations will be relevant to all the life sciences, not simply those with a biomedical bent.
The Shape of Graduate Education
All new graduate students in biology must select from a panorama of topics, like that sketched above, a specific subset that can reasonably be mastered within the 5–10 years that are commonly devoted to a PhD degree. Graduate work almost always begins with courses, but many programs strive to get their students into a research environment as soon as possible. The intent is partly to distinguish graduate from undergraduate education and partly to let students see what the life of a scientist is like. Coursework usually dominates the first year or more of graduate study and trickles on through years 2 and 3. A preliminary examination usually evaluates competence to continue training, and
the passage of a general examination in the second or third year permits admission to candidacy for the PhD degree. A graduate student usually identifies dissertation supervisor in the first or second year and begins thesis research shortly thereafter.
It is uncommon for graduate biology students to pay their educational expenses from their own resources (see table 2.1 in chapter 2), because there are numerous alternatives: salary grants to individual students, training grants to departments or programs, research grants to faculty members who can then support a graduate research assistant, teaching assistantships from the college or university, and in some cases loans to help to postpone expenditures until more lucrative employment is available. Most graduate students teach at some time during their training, but the duration of this teaching experience usually depends on whether they can obtain support from a research-oriented source that allows them to complete their thesis work without the complications of teaching at the same time.
The duration of graduate training is variable, depending in part on the subdiscipline in question: molecular biology and cellular biology tend toward 7 years (elapsed calendar time from the bachelor's degree to the PhD degree and about a year less as a registered student in the program), but training that requires extensive work in the field or an analysis of populations over a long term takes longer. The mean time to completion of a life-science PhD has increased from 6 to 8 years over the last 25 years. (Chapter 2 presents more detailed data on the graduate and the postdoctoral experience.)
The Postdoctoral Experience
Graduate students in biology who desire a career in research often pursue further training at the postdoctoral level. According to data from the National Research Council's Survey of Doctorate Recipients (SDR, see, for example, NCR 1996), the fraction who go on to this level of training more than quadrupled from 1973 to 1993; in 1995, 53% of life-science PhD recipients pursued further training as postdoctoral fellows within 1–2 years of earning their degrees. Three reasons for postdoctoral training's becoming so common in the life sciences have been suggested: building a successful research career requires such a magnitude and diversity of knowledge that additional training in a second research environment is helpful; funds are often available for postdoctoral stipends, making the second training stage relatively available and additional outlays by the postdoctoral fellow unnecessary; and the competition for jobs with more independence and security is intense. Thus, the improvements in one's curriculum vitae (CV) that result from the additional research experience and publications characteristic of postdoctoral work are very important for one's prospect of permanent employment. The relative importance of those factors is discussed in chapter 5.
Some postdoctoral fellows apply for and receive their own funding from a government agency or a private foundation. Such fellowships are particularly desirable because the recognition that accompanies them carries implicit and explicit messages of intellectual and professional independence and because the salary money makes a candidate more attractive to a host laboratory of high quality. Other postdoctoral fellows are supported by salaries specified in the research budgets of their new host laboratory. To some extent, scientists in the latter group are more obliged to work on the projects for which their new mentors have been funded than on projects of their own choosing. However, because postdoctoral fellows commonly select their host laboratories on the basis of an interest in the science that is done there, that constraint is usually of minor importance, at least at first. Many young scientist find that the first 2 or 3 years of postdoctoral experience is exceptionally rewarding. Researchers at this stage of professional development are already experienced
enough to get good work done fast, but new enough to the subdiscipline of their new host laboratories to find their work both challenging and valuable. The combination of scientific competence with a new scientific project is heady, constructive, and useful. Many senior scientists look back on their postdoctoral years as among the best of their scientific careers.
The graduate experience and postdoctoral training are formative in developing a sense of how science should be done. Virtually all graduate training and most postdoctoral work are carried out in the academic environment of a university or medical school, so the experiences of young life scientists are heavily weighted toward the loosely structured environments characteristic of basic-research laboratories. That situation might contribute to the preference that many postdoctoral fellows show for continuing their careers in an academic environment.
In recent years, it has become common for postdoctoral training to last at least 3 years. That situation is now having an important on the lives of older postdoctoral fellows because most postdoctoral fellowships last for only 2 or 3 years. For those who derive their stipends from host laboratories or institutions, the support rarely extends more than 5 years. A distinction should be made between ''postdoctoral training", when a young life scientist is learning new approaches or techniques, and "postdoctoral employment", when training is largely over and the young scientist is continuing to work at this professional rank, improving his or her CV and/or looking for a more permanent and independent job.
As the length of the postdoctoral experience increases, the issue of job security can become more important. Moreover, starting postdoctoral salaries are usually rather low and increase only modestly with additional years of experience (the recommended NIH postdoctoral salaries for a person with up to 5 years of previous postdoctoral experience have recently been increased to just over $20,000 per year at the beginning of their NIH-supported postdoctoral work and just under $30,000 per year at the end; fringe benefits are also modest). Few universities have a professional structure that provides additional financial support for postdoctoral fellows, and although they are welcomed in scientific professional societies, they are neither students nor established professionals. That situation provides strong motivation for most postdoctoral fellows to try to find a different form of employment within 5 years of obtaining their PhD degrees.
The Pursuit of a Job
After a period of postdoctoral training and the publication of several papers as evidence of scientific accomplishment and expertise, most postdoctoral fellows apply for positions that carry some measure of future prospects and permanence: tenure-track academic posts, jobs in companies or government laboratories, or positions in alternative professions that will enable them to use their scientific training or research skills. In recent years, the job market for life-science PhDs has tightened considerably. The number of positions in academic institutions, the largest employers of life-science PhDs, has not increased as fast as the number of applicants. Junior faculty positions for which the field of research is not narrowly defined generally attract several hundred applicants, and good jobs in industry and in primarily undergraduate, teaching-intensive colleges are just as competitive. Of course, some young scientists with extraordinary credentials get jobs immediately, but many others with impressive CVs are now finding the professional transition very difficult (for a more complete treatment of this important issue, see chapter2 and 3).
In response to the tightening job market, there has been an expansion in the range of positions that young biologists will look at seriously. The extent of this "alternative" job market is not at
present very clear, but some of the major research centers are beginning to provide symposiums and conferences on the careers available to life-science PhDs outside the conventional spheres of employment. The reaction among postdoctoral fellows has been mixed (as discussed in chapter 5). The problem for an individual postdoctoral fellow remains how best to be distinguished from the competition. To maximize their marketability most candidates try to publish as much as they can in journals that are widely read. Job seminars get brightly polished, and candidates practice presenting themselves favorably. Even with strong credentials and a broad perspective on the suitability of diverse employment opportunities, however, it often takes several years to get a good job. This difficulty is almost certainly an important factor in the increasing duration of postdoctoral "training".
The Pursuit of Research Support
For applicants who get positions in industrial or governmental laboratories, resources for research are usually included. For new employees in academic institutions and research institutes, the next career step is usually to obtain funding that will support scientific work. Many job offers include some funds with which to set up laboratories, so initial purchases of equipment and often the first year or so of research supplies are already available, but the expectation for most new employees in these research institutions is that they will apply for and obtain their own research funding. The details of an application vary from one granting agency to another, but a research proposal usually includes a description of the scientific context and significance of the proposed experiments and a detailed account of how the work will be done. Construction of such a proposal takes anywhere from a few weeks to a few months, and the probability of success of first applications is not high, ranging from less than 10% in some agencies to 35% in others. Such figures, of course, vary from year to year and depend primarily on the state of the economy and the attitude of Congress toward research.
Staying funded is not much easier. It is important to remember that obtaining grants has been difficult for many years; there are few investigators still submitting proposals whose work is not of good quality. The competition is therefore intense for all investigators, young and old, and achieving a rank in the top one-third is not easy. A successful proposal requires not only imagination, skill, and hard work, but also good fortune. It helps to be in the right intellectual place at the right scientific time. If a proposal is radically different from the scientific mainstream, it can be dismissed as "risky". If it is not sufficiently involved with current methods and ideas, it can be dismissed as old-fashioned. There is also some luck in the rather arbitrary choice of who reviews a particular proposal. Most reviewers are highly accomplished scientists, chosen by well-meaning grant administrators for their expertise and fair-mindedness. However, when the people who review a proposal know and respect both the subfield in question and the work of the applicant, the chances of a fundable score are likely to improve.
It is also important to recognize the importance of funding for life scientists working outside government or industrial laboratories. Most universities, medical schools, and research institutes require grants to individuals for the pursuit of a particular project: if there is no grant, there is no (or very little) support for research. Furthermore, one's livelihood is often affected by a grant, dramatically in some instances. In most colleges of arts and sciences and related university divisions, a salary is provided for only 9 months of the year, the time when a principal investigator is engaged in teaching and related university activities. Salary for the summer months can be sought from a research grant, and sometimes a fraction of a principal investigator's academic-year salary will be included as well, on the grounds that the faculty member is using that
portion of his or her time on research-related activities. In medical schools and other medical research institutions and in private institutions to a greater extent than in public ones, research personnel are expected to obtain substantial portions of their salaries from grants throughout the year. Thus, the motivation to write successful proposals is high indeed. Given all those factors, it is no wonder that many principal investigators spend a large fraction of their time seeking the funds with which to do research.
The Character of the Research Environment
Given the diversity of biologic research, there is a huge range in how life-science research is conducted. Some is done "in the field", with a heavy emphasis on the observation of organisms in their natural settings. Some is done in the field, literally; selected plants are grown in experimental plots side by side with control strains to assess their relative susceptibility to disease, drought, or nutritional deprivation. Some is done in laboratories that could serve a chemist or a physicist as well as a biologist. The following generalizations should, however, be reasonably applicable to all.
A principal investigator builds a research group by defining the scientific questions to be addressed, specifying the methods to be used, obtaining necessary funding, finding the suitable research environment, and attracting the research personnel, usually a mixture of students, technicians, and postdoctoral fellows. The day-to-day jobs of the principal investigator include those of a research manager: making decisions about expenditures and personnel matters, evaluating data, planning the next experiments or observations, providing training for less experienced personnel, and directing the whole enterprise toward the completion of research manuscripts for publication. Ancillary tasks include the writing of grant proposals and such research-related articles as reviews of the literature, critiques of work of other principal investigators, and the committee work associated with the host institution. Many principal investigators must also teach and administer activities distinct from their own research projects.
The research personnel in the group usually work on more-specific tasks that pertain to the construction of research tools or the acquisition and analysis of data. Group sizes usually range from a few workers to around 20; some exceptional research groups are much larger. It is common for the social structure of the research environment to be quite free, permitting and even encouraging iconoclastic and innovative contributions from anyone in the group. Rarely is the judgment of the principal investigator always right, and the details of a particular experiment or observation are sometimes known only to the people doing the work. The ebb and flow of criticism and suggestion between the principal investigator and the laboratory members is one of the things that make a free social structure so effective for the progress of science. The give and take is one of the most instructive and constructive aspects of a laboratory environment; it is a key reason why research training must be obtained "on the job" in an apprentice situation, not in a classroom. The give and take is also of great value for the quality and quantity of science that gets done; mistakes in judgment or knowledge are often corrected quickly without the emotional stress that can develop in a more structured environment. It is the rare (and foolish) principal investigator who is offended by constructive disagreement.
One of the most important aspects of the laboratory group structure is its flexibility and intellectual mobility. In fast-moving fields like the modern life sciences, the intellectual ossification that can accompany a major administrative structure, such as the environment suitable for an expensive instrument, impedes the readjustments of position and direction that are necessary for innovative work. Flexibility of
structure has been one of the great strengths of life-science research in the United States. Research groups can vary widely from the model described above, depending on the discipline, the size of the group, the personality of the individuals involved, and the institution; but even this variation is probably constructives: it allows the country's research enterprise to encompass many approaches within the framework of research that is supported by grants to individual life-science investigators. The resulting pluralism has contributed to the ability of American life-sciences to explore the biologic landscape fast and economically. Even the research structures found in many companies can be described by this model, although they include a different range of constraints, depending on the scientific and economic goals of the companies.
Flexner A. 1910. Medical education in the United States and Canada. New York: Carnegie Foundation.
Judson HF. 1979. The eighth day of creation: makers of the revolution in biology . New York: Simon and Schuster.
NIH (National Institutes of Health). 1961. Basic data relating to the National Institutes of Health. Bethesda, MD: NIH.
NIH (National Institutes of Health). 1978. NIH almanac. Bethesda, MD: NIH.
NRC (National Research Council). 1994. Meeting the nation's needs for biomedical and behavioral scientists. Washington, DC: National Academy Press.
NRC (National Research Council). 1996. Summary report 1996: Doctorate recipients from United States universities. Washington, DC: National Academy Press.
NSF (National Science Foundation. 1960. Science, the endless frontier, a report to the president on a program for postwar scientific research. Washington, DC: US Government Printing Office [Reprint of 1945 publication.]
NSF (National Science Foundation). 1996. Science & Engineering Indicators 1996. NSB 96-21. Washington DC: US Government Printing Office.