President Donald Trump’s proposed budget for the 2026 fiscal year has made drastic cuts to the National Institute of Health’s (NIH) budget, sparking alarm among many. While the level of proposed cuts is unprecedented, calls for efficiency are nothing new. In fact, they echo decades-old efforts to make publicly funded science more accountable. What has gone largely unnoticed, however, is how these reforms reshaped how NIH research is managed—as well as the very definition of what counts as rigorous, worthwhile health research in the first place.[time-brightcove not-tgx=”true”]Even as the NIH’s budget soared over the past half-century, much of that growth came at a price: a narrowing of NIH’s scientific imagination. Driven by bureaucratic reforms and the need to demonstrate fiscal responsibility, the agency gradually shifted away from large, community-based, longitudinal studies aimed at understanding what keeps people healthy. Instead, it prioritized smaller, faster studies with statistical significance and quantifiable data, but far less explanatory power about how to stay healthy. In the late 1950s, the NIH was beginning to expand its mission to address chronic ailments like heart disease and cancer. These growing health threats required a fundamentally different kind of science—slower, more complex, and deeply embedded in communities. Early NIH leaders, such as James Shannon, embraced this challenge with a bold vision: government-led, multi-site observational studies tracking large populations over decades. The Framingham Heart Study, launched in 1948, embodied this approach. It aimed to enroll over 5,000 healthy residents of Framingham, Mass., and follow them for at least 20 years to understand how lifestyle factors and social context shaped long-term health outcomes. Read More: RFK Jr. Says Ultra-Processed Foods Are ‘Poison’—But That He Won’t Ban ThemOver the next decade, the NIH became the de facto institution for carrying out this sort of bold population-based investigation into health and disease.But as the 1960s progressed, this vision for the NIH ran afoul of a growing government-wide push for budgetary control. Reforms like Planning, Programming, Budgeting, and Execution and Zero-Base Budgeting demanded that all federal agencies and initiatives define outcomes in advance and justify expenses with quantifiable projections. Large-scale observational studies—by their very nature, exploratory, slow, and expensive—were easy targets for government watchdogs obsessed with efficiency. For example, the Wooldridge Committee, a task force appointed by President Lyndon B. Johnson’s Office of Science and Technology to review the federal research enterprise, sharply criticized the NIH in 1965 for failing to provide adequate oversight of its biggest studiesThe committee warned that scientific freedom could no longer excuse a lack of fiscal discipline.The NIH responded, not by defending the long arc of discovery required for understanding the causes of chronic disease, but by adapting. Researchers were asked to project statistical returns on investment. Studies were re-evaluated not just for scientific merit, but for how likely they were to generate measurable results within a budget cycle. Framingham, once a flagship of public health research, was deemed too open-ended. By 1970, it had lost its privileged status and instead had to compete for grants like any university-based project. This shift marked an institutional pivot away from NIH-led, community-grounded studies and towards a more manageable model of research. During this time, the NIH also shelved several other large, prospective population-based studies of health and disease, including the Diet-Heart Study—an ambitious effort to definitively test the role of high-fat diets in causing heart disease.In their place, a new framework for investigating chronic diseases emerged, one built around smaller, investigator-initiated grants awarded to outside researchers. These grants, and the peer-review process that governed their approval, increasingly relied on the tools of biostatistics to demonstrate methodological rigor and fiscal discipline. From an administrative perspective, these outside projects were easier to justify: they were shorter in duration, cleaner in design, and more narrowly focused. Politically, they were appealing too—distributed across universities in different congressional districts, they helped spread NIH funding across the country. By encouraging investigators to design studies with tightly defined objectives, measurable outcomes, and clear statistical models, the NIH was able to present its growing budget as aligned with the broader federal push for transparency and accountability. In effect, the agency avoided deeper scrutiny by embedding oversight expectations into the very structure of scientific inquiry. By doing so, it created the conditions for its outside grant program to flourish.Yet, this shift also produced a subtle, but profound, change in the kinds of questions NIH research was designed to answer. Rather than pursuing the fundamental causes of health and disease, the types of population-based investigations that received NIH grants looked at discrete, isolated lifestyle factors and their relative impact on specific conditions — what have come to be known as risk factor epidemiology. In the case of heart disease, this involved studies on the impact of certain foods on conditions commonly associated with heart disease, especially high cholesterol, high blood pressure, and elevated body mass index. And while these types of investigations yielded a flood of peer-reviewed publications and some effective interventions at the individual level, they also left crucial questions unanswered. After decades of risk factor research, for example, we still do not fully understand the causes of heart disease—or how best to prevent it.Read More: NIH Budget Cuts Are the ‘Apocalypse of American Science,’ Experts SayIn the decades since, many investigators and commentators have criticized the dominance of “risk factor epidemiology.” Critics include Gary Taubes, a science journalist known for his writing on nutrition science and the history of dietary guidelines, and John Ioannidis, a Stanford researcher who has long argued that most epidemiological studies of nutrition are limited in scope and contradictory. They and others contend that risk factor–driven research has led to public health guidance built on fragile associations and patterns in data that do not reflect causality. These critics often point to the decades-long emphasis on reducing dietary fat to lower cholesterol and prevent heart disease as problematic. This advice led many Americans to adopt low-fat, high-carbohydrate diets—a diet that is now linked to obesity, diabetes, and ironically, heart disease. Today, many health experts and institutions have reversed course, encouraging the consumption of healthy fats and warning against excess sugar and refined carbohydrates. The result has been public confusion, eroded trust in nutrition science, and a generation of health advice that, in retrospect, may have done more harm than good.These studies have flourished since the 1970s not because they promised definitive answers on how to stay healthy, but because they appeared to offer a clear return on investment. Their study designs were statistically rigorous and focused on narrowly defined variables and outcome measures, which enabled these projects to routinely yield statistically significant results for the questions they were designed to answer. That gave policymakers and funders the impression that public dollars were driving scientific progress, even as it provided few answers to the biggest scientific questions. Ironically, it was the promotion of this particular style of research—narrow in scope, statistically precise, and managerially friendly—that helped the NIH expand its budget and reach. But the accumulation of these rigorous, but smaller-in-scope, findings rarely translated into an applicable understanding of the complex, long-term, and interconnected forces that truly shape health. Today, as the NIH again faces oversight and budget pressures, the American scientific establishment has a chance to course-correct. The current administration has emphasized health promotion and the importance of diet. But if those goals are to be more than talking points, President Trump, Congress, and the NIH must be willing to invest in the kind of science that can actually reveal what keeps us well. That means returning to community-based, long-term observational studies—even if they are expensive, even if they take decades, and even if they do not fit neatly into the bureaucratic logic of annual performance metrics. Sejal Patel-Tolksdorf is a health policy analyst and former chief research historian at the National Institutes of Health. Her work focuses on the politics and policy of American health research. Made by History takes readers beyond the headlines with articles written and edited by professional historians. Learn more about Made by History at TIME here. Opinions expressed do not necessarily reflect the views of TIME editors.