The American biomedical research enterprise "is on an unsustainable path" that will lead to "long-term decline" unless "some fundamental features" undergo reform, four influential scientist-administrators state in a lengthy article published today. Prominent among the authors' concerns is "a hypercompetitive atmosphere in which scientific productivity is reduced and promising careers are threatened."
Writing in Proceedings of the National Academy of Sciences (PNAS), the four authors—Bruce Alberts, former president of the National Academy of Sciences and former editor-in-chief of Science; Marc W. Kirschner , founding chair of the Department of Systems Biology at Harvard Medical School; Shirley Tilghman, former president of Princeton University; and Harold Varmus , Nobel laureate and current director of the National Cancer Institute—call on the bioscience community to "confront the dangers at hand and rethink" how academic research is funded, staffed, and organized.
"The great majority of biomedical research is conducted by aspiring trainees: by graduate students and postdoctoral fellows," the authors write. "[T]he training pipeline produces more scientists than relevant positions in academia, government and the private sector are capable of absorbing." The oversupply of scientists, and the resulting scramble for limited resources, has "diminished the attraction of our profession for many scientists—novice and experienced alike."
Among the authors' stated goals is "to gradually reduce the number of entrants into PhD training in biomedical science—producing a better alignment between the number of entrants and their future opportunities." Their specific proposals for achieving this goal are familiar; many of them were included in the 2012 report from the Biomedical Research Workforce Working Group. A 1998 report from a National Academies committee (chaired by Tilghman) covered much of the same ground.
They propose, for example, altering "the ratio of trainees to staff scientists in research groups." To gain more control over the number of trainees and improve the quality of their training experience, they propose increasing the proportion of graduate students supported by training grants instead of professors' research grants. They recommend providing graduate students with better career information, increasing postdoc pay, making changes in grant review, and changing the "perverse incentives" that encourage universities to expand facilities and hire faculty on soft money. In addition, they call on those who provide the money that supports research, namely Congress and the executive branch, to develop means of "planning for predictable and stable funding of science" rather than continuing the current damaging pattern of lurching unpredictably from funding increases to funding cuts and back again.
Numerous  other  observers  have  been  making  similar  points, and calling for similar reforms, for decades. Indeed, in 1945, presidential science adviser Vannevar Bush wrote in "Science, The Endless Frontier "—the report that outlined the organizational structure that still governs federally supported academic research—that the first of five "fundamental" principles must be "stability of funds over a period of years," a goal that Congress has never met. He also warned against wasting talent by drawing more young people into science than could find productive roles there.
The authors of the PNAS article are generous to the scientific and governmental figures who have long ignored repeated calls for reform and who for so long failed to see (or to acknowledge) that problems even exist. "The strains have been building for some time," the authors write, "but it has been difficult to recognize them in the midst" of the successes that biomedical research has achieved over the past generation.
That is true, no doubt, from the perspective of those who have long benefited from the sacrifices of graduate students and postdocs—among them administrators, policymakers, and well-funded lab chiefs. As a result, as recently as last
October September, Sally Rockey, the National Institutes of Health's (NIH's) deputy director for extramural research, and NIH Director Francis Collins could write, in a comment on a blog post , "there is no definitive evidence that Ph.D. production exceeds current employment opportunities."
Today, however, due to the recent throttling of the nation's research budgets, the pain that early-career scientists have endured for decades is reaching into the ranks of the formerly comfortable. Now that the more privileged members of the scientific community—those likely to chat with top administrators—have begun to fear for their own pocketbooks and careers, more people are taking note.
It is the nature of reform movements that, although grassroots movements can be instrumental, systemic change usually comes only when elite opinion moves to the side of change. The significance of the PNAS article lies not so much in its specific proposals as in the possibility that it signals the start of such a realignment.
Resistance by some of the current system's beneficiaries will continue. A reliable source tells Science Careers that within a high-level group looking into issues affecting early-career scientists, so inoffensive a suggestion as raising postdoc pay has produced controversy. The PNAS article authors do not provide a practical strategy for overcoming the dense tangle of vested interests and perverse incentives that protect the current system.
The authors do warn, however, that "mere discussion will not suffice. Critical action is needed on several fronts by many parties to reform the enterprise." In this they are correct. As a scientist devoted to empirical evidence, Tilghman could have been forgiven for giving up—for concluding from the disappointing results of her previous efforts that meaningful reform is unlikely. Yet, she persists in working for the long-range benefit of the scientific enterprise. Here's hoping that the influence of her current group of prestigious co-authors will help move the wider community toward desperately needed change.