Associating radioiodine therapy in hyperthyroidism with cancer mortality: robust or random results of a statistical analysis?