Buy this Article for $10.95

Have a coupon or promotional code? Enter it here:

When you buy this you'll get access to the ePub version, a downloadable PDF, and the ability to print the full article.


hospitals, intention to return, patient satisfaction, quality of care



  1. Otani, Koichiro
  2. Kurz, Richard S.
  3. Burroughs, Thomas E.
  4. Waterman, Brian


This article considers several models of how patients integrate their reactions to hospital attributes and how these reactions impact their overall satisfaction and behavioral intentions. It finds that patients combine their reactions to the attributes by means of noncompensatory and nonlinear models to form their overall satisfaction or behavioral intentions.


Both academicians and managers are interested in discovering how patients form their satisfaction levels and behavioral intentions. Consequently, the number of patient satisfaction studies is rapidly increasing.1-5 Many of these studies use multiple regression and assume that there are linear and compensatory relationships between reactions to health care attributes and patient satisfaction or behavioral intentions. No study to date has investigated a possible noncompensatory and nonlinear process of patient satisfaction, even though the psychological literature consistently indicates that human decision processes are neither compensatory nor linear.6-11


The study on which this article is based considers noncompensatory and nonlinear relationships with regard to the effect of hospital-discharged patients' reactions to health care attributes (admission process, nursing care, physician care, compassion to family/friends, pleasantness of surroundings, and discharge process) on their overall satisfaction with the quality of care and service (overall satisfaction) and their behavioral intentions to recommend and return to the hospital for future care. The first aim of the study is to determine whether patients simply average out their reactions to multiple attributes mathematically (compensatory response) or whether their evaluation is overly affected by either highly positive or highly negative reactions to health care attributes (noncompensatory response). The second aim is to focus on the reactions to individual attributes in relation to overall patient satisfaction and behavioral intentions. Given that an attribute is strongly related to overall patient satisfaction or behavioral intentions, is the relationship linear or curvilinear? If it is curvilinear, does it display diminishing marginal utility or increasing marginal utility or does an interaction occur with an additional variable?


For example, assume that a patient is reflecting on her experience at Hospital A. She believes that three attributes are equally important to her, namely, physician care, nursing care, and the environment, including the food quality and other amenities. She notices that when she had a bad experience with nursing care, her overall satisfaction level went down even though physician care and the environment exceeded her expectations. Based on her reflections she decides not to return to Hospital A. This patient does not seem to average out the three attributes mathematically, but responds differently. This scenario illustrates that a negative experience with one attribute might influence her overall satisfaction disproportionately. In other words, a patient's negative experience with one service attribute might doom her satisfaction with the stay and influence her decision not to return to the facility.


These issues are important for health services researchers and hospital administrators who are seeking to improve patient outcomes and organizational performance. By addressing these analytic issues, it becomes possible to find where and how administrators should invest their limited resources to improve specific service attributes to increase patient satisfaction and intentions to return or recommend a facility. Our results suggest a more focused approach by managers.