Introduction: Hand-held dynamometry (HHD) is a technique used for evaluating muscle strength. The reproducibility and reliability of this instrument have been little reported. Method: Healthy individuals aged >18 years were recruited. Isometric strength was tested for ten upper limb muscle groups. The fixed method was performed using a system with suction cups connected to the dynamometer through an inelastic belt. However, during the non-fixed method, the examiner supported the device with the hand. The reproducibility and reliability were calculated using the intraclass correlation coefficient (ICC). Result: A total of 25 right-handed volunteers participated. The reliability ICC values (0.89-0.99) of the non-fixed method were higher than those of the fixed method (0.43-0.85). The reproducibility of the non-fixed method was also superior to that of the fixed method. Discussion: The comparison between the HHD methods suggested the non-fixed method has greater reproducibility and reliability.