Although not specifically discussing the Lightdale olen Klein. BM, Landrinan, and wely(1997 highlighted ueveral health belief tions thought to be important in health promotion in an American Indian ulation. First, it was popi wwgested that individuals must believe theyare at risk before hey would take steps to change behavior, which is consistent with the perceived susceptibility and perceived severity functions of the HBM, American Indians in the Lighdale study significantly underestimated personal risk in 10 of the 11 variables measured, including diabetes Participants in the study also showed significant personal bias when comparing their own risks to that of their peers.other measures taken in ehe study showed that there was no significant correlation betweenestimated personal risks and behavior. For example. those who naled themselves at low risk for being in atraffic accident were no lesslikely to drink and drive, speed, or drive long distances dann tose who estimated their risk for accident as high A study examining regimen adherence of American Indian diabetics used a dnerent health beliel model that examined the influence of"individual's beliefs that other persons important to them think they should perform a specific behavior(Miller e al.. 1987, p. 25) nis study found that the perception of important othen' beliefs was the most significant contributor in a regression oquation that included coping skills and as comparison variables This study also examined personal attitudes toward attitudes components a diabetes management regimen that included medication, stress. activity, smoking, and diet(Miller et al. 1987). This measure ofattitudes had some similarity to the HBM variables perceived benefit and perceived barriers. Participants in the study were divided into two groups, one with good metabolic control af their disease and the