To review the literature on breaking bad news while highlighting its limitations and describing a theoretical model from which the bad news process can be understood and studied. Sources were obtained through the MEDLINE database, using "bad news" as the primary descriptor and limiting the sources to English-language articles published since 1985. All articles dealing specifically with bad news were examined. These works included letters, opinions, reviews, and empirical studies. Recommendations from these articles were examined, sorted into discrete categories, and summarized. The 13 most consistently mentioned recommendations (eg, delivering the news at the patient's pace, conveying some hope, and giving the news with empathy) were examined. Although much has been written on the topic of breaking bad news, the literature is in need of empirical work. Research should begin with the simple question of whether how the news is conveyed accounts for variance in adjustment before moving to more specific questions about which aspects of conveying bad news are most beneficial. It is suggested that the bad news process can be understood from the transactional approach to stress and coping.