OBJECTIVES: The majority of care for those with Alzheimer's disease and other age-related dementias is provided in the home by family members. To date, there is no consistently effective intervention for reducing the significant stress burden of many family caregivers. The present pilot randomized controlled trial tested the efficacy of an adapted, eight-week mindfulness-based stress reduction (MBSR) program, relative to a near structurally equivalent, standard social support (SS) control condition for reducing caregiver stress and enhancing the care giver-recipient relationship. METHOD: Thirty-eight family caregivers were randomized to MBSR or SS, with measures of diurnal salivary cortisol, and perceived stress, mental health, experiential avoidance, caregiver burden, and relationship quality collected pre- and post-intervention and at three-month follow-up. RESULTS: MBSR participants reported significantly lower levels of perceived stress and mood disturbance at post-intervention relative to SS participants. At three-month follow-up, participants in both treatment conditions reported improvements on several psychosocial outcomes. At follow-up, there were no condition differences on these outcomes, nor did MBSR and SS participants differ in diurnal cortisol response change over the course of the study. CONCLUSION: Both MBSR and SS showed stress reduction effects, and MBSR showed no sustained neuroendocrine and psychosocial advantages over SS. The lack of treatment condition differences could be attributable to active ingredients in both interventions, and to population-specific and design factors.