Optimism and the Global Health Growth Curve.

Abstract

David Hume, the brilliant and skeptical Scottish philosopher, was known for criticizing the human tendency to extrapolate—to assume that things would continue to motor along at the same speed on the same trajectory. He encouraged us to not make assumptions or predictions based merely on the “constant conjunction” (correlation) of past events, but to carefully examine apparent relationships for a causal link. Much of what happens in our world is linear and predictable, like clockwork in some cases. In other cases, nonlinear and startling trajectories best describe the behavior of economies, political movements, or the various factors in what we might call “global health.” Global health progress has been dramatic in some areas, but discouraging and slow in others. What trends should we anticipate in the coming decades? Is there reason for optimism? Simon Kuznets was a Nobel laureate economist who charted social inequality against national per capita income in a classic article in 1955. He used data from a variety of countries to plot a parabolic “inverted U” curve, predicting that inequality might increase as farmers moved to cities and industry, but would later shrink as prosperity reached broadly into a society. Kuznets understood the fragility of his data and some of its limitations, but others have extrapolated from his outcome to dubious conclusions. Will the trajectory of global health programs follow a Kuznets’ curve? Climate change or another major disruption could affect improvements in global health indices, and even turn back some successes. The growth of global health might follow the nonlinear shape of Kuznets’ curve, but I doubt it. The growth curve for global health might also take a “J” shape. Author and political scientist Ian Bremmer writes a compelling narrative for this trajectory in his 2006 book, The J Curve: A New Way to Understand Why Nations Rise and Fall. He describes a nonlinear relationship of stability to progress in social freedoms, in which stability decreases initially as social freedoms increase; when freedom continues to increase, the brief dip in stability is followed by a long upward linear improvement. We see this behavior in the prevalence of polio immunity, i.e., when violence or deceitful propaganda are allowed to intrude on immunization campaigns. But I do not believe the “J” curve will be the most common trajectory for global health programs in the coming decades. Many physical processes, like chemistry “saturation” curves, the growth curve of populations, and improvements in longevity over time, follow a sigmoid (S shaped) curve. Input in the early stages may not show much response or return on investment, until reaching the cusp of an exponential rise, when return vastly outruns input. Later, the output stabilizes and reaches a plateau, where “saturation” or “carrying capacity” is achieved, and additional input produces diminishing returns. Even the exponential improvement in computing capability predicted by Gordon Moore in 1965 may finally be approaching a plateau, and showing a sigmoid shape. I believe that future growth in most global health indices will describe a sigmoid curve. Samuel Preston first demonstrated this behavior in his description of the relationship of longevity to gross national product (GNP) per capita. At relatively low GNP per capita values ($2,000), average national longevity rises dramatically with each small increase in economic activity. Many of the developing world nations are poised at the base of the steep upsloping part of Preston’s curve. At about $5,000 GNP per capita, the spectacular longevity gains flatten out, and only inch up as GNP increases to about $20,000 per capita. (There are few data points beyond this, including the United States, and longevity remains stable or may even decrease slightly as GNP continues to rise. Other developed countries, struggling to reach the same economic level, may not want to go there!) Preston’s curve does not demonstrate the sigmoid shape at low GNP value. A truly sigmoid function should have a flat “tail” at low values, before beginning the exponential rise. I suspect most human societies have been on that part of the curve throughout history, only in recent centuries reaching the upslope as a group. Some of the poorest nations may still be mired in this phase today, when investments in Department of Preventive Medicine and Biometrics, Uniformed Services University, 4301 Jones Bridge Road, Bethesda, MD 20814. The views expressed are those of the authors and do not reflect the official policy or position of the Uniformed Services University of the Health Sciences, the Department of Defense, or the U.S. Government. doi: 10.7205/MILMED-D-16-00020