Prevalence of harm in mindfulness-based stress reduction.

Abstract

Mindfulness meditation has become a common method for reducing stress, stress-related psychopathology and some physical symptoms. As mindfulness programs become ubiquitous, concerns have been raised about their unknown potential for harm. We estimate multiple indices of harm following Mindfulness-Based Stress Reduction (MBSR) on two primary outcomes: global psychological and physical symptoms. In secondary analyses, we estimate multiple indices of harm on anxiety and depressive symptoms, discomfort in interpersonal relations, paranoid ideation and psychoticism. Intent-to-treat analyses with multiple imputations for missing data were used on pre- and post-test data from a large, observational dataset (n = 2155) of community health clinic MBSR classes and from MBSR (n = 156) and waitlist control (n = 118) participants from three randomized controlled trials conducted contemporaneous to community classes in the same city by the same health clinic MBSR teachers. We estimate the change in symptoms, proportion of participants with increased symptoms, proportion of participants reporting greater than a 35% increase in symptoms, and for global psychological symptoms, clinically significant harm. We find no evidence that MBSR leads to higher rates of harm relative to waitlist control on any primary or secondary outcome. On many indices of harm across multiple outcomes, community MBSR was significantly preventative of harm. Engagement in MBSR is not predictive of increased rates of harm relative to no treatment. Rather, MBSR may be protective against multiple indices of harm. Research characterizing the relatively small proportion of MBSR participants that experience harm remains important.