The Problems for Academic and Entrepreneurial Research in the Use of Health Services: the Case of Unstable Structural Relationships*

Multiple regression analysis was used to determine the longitudinal stability of the structural relationships between the predisposing, enabling, and illness-morbidity characteristics of the population-at-risk and measures of health services utilization. The structural coefficients estimated from the 1971, 1972, and 1973 Health Interview Surveys were examined using the Relative Instability Ratio (RIR) presented in this paper. The RIR values indicate that—except for age, family size, occupation, and family income—all the predisposing, enabling, and illness-morbidity characteristics produce unstable structural relationships. In addition, a negative monotonic relationship was found between the temporal ordering of the groups of characteristics and their ability to produce stable structural relationships. These results demonstrate the need for entrepreneurial health manpower projection techniques and abstract academic explanatory models of health services utilization to become dynamic, rather than remaining in their present static states.