The study by Kravitz and colleagues [1] is a valuable contribution to our understanding of patients' expectations and therefore is important to the emerging science of demand management. One can imagine, then, our surprise at the authors' suggestion that these findings should decrease interest in demand management. We believe that the authors may have an inaccurate view of demand management and suggest that they consider the following points.

First, demand management does not "discourage health care utilization." The current definition of demand management is "the use of decision support and self-management support to enable and encourage individuals to make appropriate use of medical care" [2]. We cannot over-emphasize that the goal of demand management is concerned with quality, not decreased utilization or cost savings.

Second, the contrast between rational decisions and complex decisions put forth by the authors seems to us to be a false dichotomy. The examples of patient reasoning given in the article do not strike us as irrational, but they do illustrate that patients may reason by using information, perceptions, and attitudes that may be unknown to or considered erroneous by the physician.

Third, the decision support and self-management support systems envisioned by demand management are predicated on the fact that patients' perceptions, attitudes, and preferences are the dominant factors influencing the use of medical care.

Finally, the authors seem to suggest that demand management is completely foreign to physicians and the practice of medicine. We are unaware of anything in the demand management model that requires such separation. Indeed, at least one physician leader has suggested that demand management is an integral and inextricable component of primary care [3].

Confusion about demand management is common, perhaps because of the relentless marketing of "nurse advice lines." Many, if not most, such services do not meet the criteria for demand management and might be better termed "access management," a somewhat newer term for traditional attempts to decrease access to medical services (that is, supply management).

We believe that those of us interested in advancing the scientific basis of demand management and the authors share a common belief: Understanding the perceptions, attitudes, and beliefs that determine patient behavior is essential for improving health care. We hope that the authors will come to the same conclusion.