Answers to clinical questions occurring during healthcare practitioner/patient interaction can be often found in National Library of Medicine's (NLM) databases. The recent advances in wireless handheld computers promise to make them a widely used tool to deliver needed information to the practitioner at the point of service. This paper addresses challenges in organizing and presenting information obtained from NLM's MEDLINE database of indexed citations in a way that will help practitioners reduce literature search time on handheld computers. We study two clustering algorithms and two methods of labeling document clusters.