Race and hypertension: science and nescience.

Abstract

Although race is widely used in hypertension research as a marker of increased risk, its meaning as an etiologic quantity is obscure.1 2 3 4 5 Because of its importance as a social category in American life, separating what we “know” about race through socialization from what can be known on the basis of scientific inquiry is difficult.6 This dilemma represents a specific example of how social influences impact the conduct of science. While the process by which evidence is gathered and evaluated is usually constrained by an existing theoretical model (the “hypothetico-deductive process”), hypotheses arise more informally from the interplay of “inspiration” and “intuition.”7 Although hypothesis generation is at the core of scientific activity, it is peculiarly vulnerable to the influences of ideology. Contemporary perspectives on science as a practical human activity thus acknowledge that inspiration and intuition are the products of imagination and, as such, are derived from the investigator’s experience—both inside and outside the laboratory.8 While it is necessary to take account of the assumptions about race that are generated by our participation in a racially stratified society, it is also necessary to define exactly what we can know about race through the conduct of science. In this discussion, we are interested in moving beyond concerns about social influences on hypothesis generation and will attempt to address the logic of studies that use race comparisons to examine causal relationships. Specifically, we ask the question, “Can available statistical adjustment methods lead to valid inferences when race is regarded as an etiologic quantity, rather than a broad indicator of risk factor status?” Despite the popularity of racial hypotheses, we argue that comparative studies based on observations of broad phenotypes—such as high blood pressure, obesity, or glucose intolerance—have little potential to contribute new knowledge. The attempt …