Health statistics

Tensions betwen individual and aggregate perspectives are not new to this field. The problems of translating between clinical and epidemiological views are a long-standing aspect of professional debates, with clinical perspectives generally dominating.

One ideological bridge between the two that has been gaining ground for the epidemiological perspective is Evidence Based Medicine (EBM). Central to the medical discourse of EBM is the measurement, interpretation, and communication of risk and of risk reductions achieved by various medical interventions. This places statistics at the core of modern (especially preventive) medicine.

One instance where statistics play an even stronger role is the case of osteoporosis. In 1994 an expert panel appointed by the WHO proposed a definition of female osteoporosis as "bone mineral density more than 2.5 standard deviations below the young adult mean." This definition raises a number of issues at aggregate and individual levels. Since bone density declines with age, using young adults as the reference population renders osteoporosis typical amongst the elderly. This may lead to medicalization and (over-)medication, which carries their own burdens of risk. Furthermore, there is the question of which young popoulation is the relevant reference point. Defining populations geographically implieas a naturalization and racialization of health differences that may have a socio-economic basis. In an undernourished populace, average bone densities may be low. Any thereby increased incidence of fractures would, as an artifact of the definition, be hidden as "normal" for that population. Conversely, a well-fed population with fewer fractures would nevertheless have its definition-created proportion of osteoporosis patients representing a demand for treatment.

In "For Whom the Bell Curves" we will address the following questions: Why did the medical community adopt their current understanding of osteoporosis? How has this statistical definition changed the perception of osteoporosis? How has the definition of osteoporosis as a "treatable risk condition" been transmitted to lay and medical audiences? What consequences appear, at individual and aggregate levels, of practices based on the statistical definition of osteoporosis? We will also be using data from the HUNT database (containing data from several county-wide health checks, including bone density measurements) to model consequences of various interpretations and applications of the WHO definition.