“…The self-designed questions were evaluated and the results showed excellent or good agreement for most questions, demonstrating good stability and reliability ( table 4 ). However, the MG-specific questions regarding Robertson et al [12] 1998 UK standardized questionnaire Paul et al [13] 2001 USA HR-QoL, SF-36, MOS Padua et al [14] 2001 Italy HR-QoL, SF-36 Paul et al [15] 2002 USA demographic questionnaire, CDMI, MFI Padua et al [16] 2002 Italy disease-specific 25-item MG-QoL Oopik et al [17] 2003 Estonia questionnaire about onset and course of MG, comorbidity and treatment Rostedt et al [18] 2005 Sweden disease-specific 25-item MG-QoL Padua et al [19] 2005 Italy disease-specific 25-item MG-QoL Rostedt et al [20] 2006 Sweden disease-specific 25-item MG-QoL, SF-36 de Freitas Fregonezi et al [21] 2006 Spain HR-QoL, SF-36 Rostedt et al [22] 2006 Sweden disease-specific 25-item MG-QoL Bilinska and Sitek [23] 2007 Poland HR-QoL, acceptance-of-illness scale, LOT-R, BDI, STAI, MG disability scale Mullins et al [24] 2008 USA 60-item MG-QoL Burns et al [25] 2008 clinical characteristics and some potential MG-modifying factors showed a larger range with excellent-to-poor agreement scores indicating less stability, but still satisfactory test-retest results compared with other similar studies [30,31] . The MG clinical characteristics and severity were analyzed separately by regional domains, as shown in figure 3 a.…”