Clinical guidelines have been increasingly used in medicine. They represent a system of recommendations for the conduction of specific procedures used in fields from public health to different diagnostic and therapeutic procedures in clinical medicine. Guidelines are designed to facilitate to medical practitioners the adoption, evaluation and application of an increasing body of evidence and arising number of expert opinions regarding the presently best treatment and to help in delivering proper decision for the management of a patient or condition. Clinical guidelines represent a part of complementary activity by which research is implemented into praxis, standards are defined and clinical excellence is promoted in all health care fields. There are specific conditions which quality guidelines should meet. First of all, they need to be founded on comprehensive literature review, apart from clinical studies and trials in the target field. Also, there are more systems for analyzing and grading the strength of clinical evidence and the level of recommendation emerging from it. Algorithms are used to organize and summarize guidelines. The algorithm itself has a form of an informatic record and a logical flow. Algorithms, especially in case of clinical uncertainty, must be used for the improvement of health care, increasing it's availability and integration of the newest scientific knowledge. They should have an important role in the health care rationalisation, fight against non-rational diagnostics manifested as diagnostic procedures with no clinical indications, it's unnecessary repetition and wrong sequence. Several diagnostic algorithms used in the field of thyroid diseases are presented, since they have been proved to be of great use.