Data extraction and quality assessment
Data was extracted by one author (JU) and checked by another (BR). Data was entered into a predefined protocol and then inputted into Excel (Microsoft; Redmond, WA, USA). Evidentiary tables were constructed to summaries the studies. We decided to address the quality of papers in different phases, due to the nature of the studies and the phase of assessment. Firstly, all retrieved studies were independently graded by all authors in the task force (JU, TI, BR) according to the Centre of Evidence Based Medicine (CEBM) diagnosis criteria [23]. Discrepancies in grading were resolved through discussion. Quality ratings ranged from 1 (strongest evidence, for instance reports of clinical decision rules and high quality validation studies) to 5 (weakest evidence, often expert opinion). Studies receiving CEBM scores of 5 were excluded.