Theses Doctoral

Good Evidence, Bad Evidence: Science, Ethics, and the Politics of Making and Unmaking Public Health Policies

Johns, David Merritt

This dissertation examines the recent historical period of emphasis on optimizing the use of scientific evidence in policymaking, the nature of the challenge to existing sources of authority in public health, medicine, and public policy initiated by the evidence-based movement as it developed and unfolded over fifty years, and its effect on the making and unmaking of public health policies in the United States. It engages in a broad study of how the concept of evidence has been mobilized in public health policymaking in the late twentieth and early twenty-first centuries, how and why concepts of what should count as good evidence have changed over time, and differences in the role of scientific evidence in the policymaking versus policy-unmaking processes.

The dissertation does this through an intellectual history of the evidence-based movement and three historical case studies. The three cases are: 1) the development and implementation of the low-fat campaign and its subsequent destabilization, incremental modification, and partial replacement with policies aimed at reducing population-level intake of sugar-sweetened beverages; 2) the development and adoption of salt reduction policies, and subsequent efforts by health officials to buttress those policies amid changes in the science that threatened to destabilize the policy paradigm; 3) the development and implementation of policies early in the AIDS epidemic requiring that risk-reduction counseling always be provided both before and after administration of the HIV test, and the struggles of health officials to discontinue those counseling programs when doubts emerged about their efficacy. The thesis concludes with a critique of the concept of “evidence-based policy” through the example of the US Preventive Services Task Force, a pioneer organization involved in the conduct of evidence-based analysis, which has struggled to maintain its exclusive focus on the data in the face of new policy responsibilities under the Affordable Care Act. In each case study I focus on the networks of researchers, advocates, journalists, industry professionals, and public health decision makers whose collective negotiations shape policy outcomes. I draw upon extensive documentary evidence gathered in public and private archives, emails obtained through the Freedom of Information Act, and key informant interviews.

The dissertation shows how the evidence-based movement has roots in the social sciences and the contentious politics of the War on Poverty and took shape in the 1970s in the context of a pivot to disease prevention and health promotion at moment when the efficacy of many clinical interventions had come under question. The case studies show that in situations of scientific uncertainty public health interventions must sometimes be implemented before obtaining evidence of efficacy, that institutional arrangements and historical context can powerfully shape interpretations of the available research, how pragmatic considerations such as feasibility contribute to decisions to implement interventions, the stark challenge that can be posed by institutional inertia and resistance to efforts to de-implement existing programs, and the ways in which public health actors can selectively invoke and distort the past in the service of contemporary initiatives to organize for policy change. The dissertation suggests the “evidence-based” mantra masks a complex interplay of politics, values, cultural trends, and other extra-scientific factors that often better explain the policy process than do shifts in the evidence.


This item is currently under embargo. It will be available starting 2029-03-26.

More About This Work

Academic Units
Sociomedical Sciences
Thesis Advisors
Bayer, Ronald
Ph.D., Columbia University
Published Here
March 29, 2019