In Experiment 1, undergraduates (N = 87) wrote either about trivial events or about a secret while imagining (a) an accepting confidant, (b) a non-accepting confidant, or (c) no confidant. Unlike the no-confidant group, the accepting group reported fewer illnesses at 8-week follow-up than did the non-accepting and trivial groups, especially to the extent that this group found their confidants to be accepting and discreet. Experiment 2 (N = 74) used the same design, except that the confidant manipulation came after the writing. Eight weeks later, the accepting group' having imagined reactions that were more accepting and less judgmental than the non-accepting group' again reported fewer illnesses than did the non-accepting and trivial groups. The author suggests that when people keep personal secrets, they often do so because they fear being ostracized. Revealing to an accepting confidant can reduce distress associated with not belonging and, therefore, can lead to health benefits.