Can Patient Self-Management Explain the Health Gradient? Goldman and Smith (2002) Revisited

Using the Health and Retirement Study, Goldman and Smith (2002) [“Can patient self-management help explain the SES health gradient?”, PNAS, 99(16), 10929-34] find that poor self-management of health has significant effects on health deterioration and that such poor behavior is less likely to occur among the more educated. They conclude that self-management of health is an explanation for the much-documented positive correlation between education and health outcomes—the education-health gradient. In this paper, I show, using the same data, that controlling for poor self-management behavior has little impact on the education-health gradient, raising doubt about whether such behavior can be said to explain the gradient at all. I also show that Goldman and Smith's results—both the effect of adherence on health and that of education on adherence—depend on how poor self-maintenance behavior is defined. Lastly, using another component of socioeconomic status, viz. resources, I find that a gradient in resources and health does exist, but this too is not impacted much by self-maintenance. These findings cast doubt on whether self-management can explain the gradient in health and socioeconomic status, when the latter is represented by education or resources.