OBJECTIVE To evaluate the feasibility and the reliability of a quantitative method of maximal isometric strength measurement for the assessment of patients with myositis in clinical practice and research. METHODS Three observers independently examined the strength of 13 muscle groups of 7 patients with stable polymyositis and dermatomyositis (PM/DM) using a handheld pull-gauge. Reliability was assessed following a balanced random 3-way cross classification with interactions and using the analysis of variance method. Serial followup strength data for 2 patients were compared to change in creatinine kinase (CK) levels. RESULTS All measurements could be performed without causing pain to the patients. One single test including 13 muscle groups was done in 7 min or less. Both intra and interobserver correlations were strong and significant for all muscle groups. The intraobserver reliability ranged from 0.88 for elbow extensors to 0.98 for knee extensors and cervical spine flexors. The interobserver reliability ranged from 0.81 for elbow flexors to 0.98 for knee extensors and cervical spine flexors. The followup of 2 cases yielded moderate to strong correlations of serial strength measurements with CK levels. CONCLUSION Serial measurement of isometric muscle strength with a handheld pull-gauge is a feasible, inexpensive, time efficient and reliable method and may provide additional quantitative information in the clinical assessment of patients with myositis.