Sensitivity reviews of test content are commonly advocated techniques for reducing bias and enhancing fairness in employment and educational testing. However, few descriptions or empirical investigations of these techniques exist. The present paper presents a study documenting common sensitivity review practices and the extent to which expert reviewers agree in their judgments of item sensitivity. Results indicated that reviewers do not always receive training or adequate guidance and most frequently encounter subtle forms of insensitive item content. Further, only modest agreement in expert ratings of item sensitivity was found. Implications for improving sensitivity review practices are presented.