With growing demand for health information and rapid development of information technology, health information websites are emerging as the most effective media to meet the public's needs for health information. This article is intended to offer a technical view on the design and operations of health information websites. Along this line, employed here is the Technology Acceptance Model (TAM), which has been widely used to predict user acceptance based on Perceived Ease-of-Use (PEOU) and Perceived Usefulness (PU). We extend the original TAM by including some exogenous variables since it is necessary to understand the role of the antecedents of acceptance constructs when designing an effective health information website for improving user satisfaction. This study focuses on identifying the core functional factors in designing and operating health information websites. Conducted are some multivariate statistical analyses based on data from an extensive survey. The results from the structural equation analysis suggest that functional characteristics should be categorized into three groups: one affecting PU and PEOU, another affecting only PEOU, and the other having no direct effect on either PU or PEOU. In particular, 'usage support' and 'customization' are two key functional characteristics in the extended TAM framework for health information websites. Contrary to expectations, however, the direct effect of PEOU on usage support is hardly observed, which differentiates health information websites from other commercial websites like online shopping malls. As a result, understanding the antecedents of PU takes on more significance.