Objective: To evaluate the approach of endocrinologists in the setting of nondiagnostic thyroid fine-needle aspiration (FNA) biopsies. Methods: In 2002, we surveyed physicians attending the national annual meetings of the American Association of Clinical Endocrinologists and the Endocrine Society of North America, using a 13-item questionnaire. The responses were tallied and analyzed. Results: Of the 143 respondents, 139 were endocrinologists, with a male:female ratio of 2.5:1. Most respondents were involved in a medical practice in North America, but Europe, Asia, New Zealand, and Australia were also represented. Of those performing thyroid FNA biopsy, 31% used thyroid ultrasound guidance. Among the survey respondents, 16%, 49%, 20%, and 15% performed less than 2, 2 to 5, 6 to 10, and more than 10 thyroid FNA biopsies per month, respectively. Among the respondents, 13.5%, 44%, 28.5%, 10%, and 4% had nondiagnostic rates of less than 5%, 5 to 10%, 11 to 20%, 21 to 30%, and more than 30%, respectively. The approach of the respondents to an initially nondiagnostic FNA was repeated FNA biopsy in 87%, observation in 7%, levothyroxine suppression in 4%, and thyroid scintigraphy in 2%. Respondents believed that the most cost-effective approach in a patient with nondiagnostic FNA was repeated biopsy (82%), monitoring the size of the thyroid nodule (17%), and surgical referral (<1%). No one was willing to repeat the thyroid biopsy more than three times. Conclusion: On the basis of findings in our survey, most endocrinologists repeat thyroid FNA at least once when confronted with a nondiagnostic result. No published studies have demonstrated the cost-effectiveness of this approach versus proceeding to surgical intervention or observation. We hope that this survey will encourage further studies on this issue.