The aim of the study was to present a method enabling a reduction of the number of measurements when quantifying isokinetic muscle strength (IMS), and to describe an expression for IMS which is independent of measuring conditions. IMS is a measure of muscle strength used to categorise patients and to assess changes over time. IMS measurements involve one or several measures of peak torque at different chosen angular velocities. Comparisons between studies are difficult and an expression of IMS independent of measuring conditions is needed. Thirty-six women with rheumatoid arthritis were followed for 1 year. IMS was measured on five occasions on both knees at seven angular velocities, giving 10 sets of peak-torque data for each patient. The log[peak-torque]-angular-velocity diagram showed a linear relationship. From this, the peak-torque value at the chosen angular velocity of 30 degrees/s (IMS30) was estimated for each test and the uncertainties involved were quantified. IMS30 is a reliable parameter when comparing effort between knees, visits and patients, and may be based on three sets of data only. IMS may be expressed by a single value, IMS30, representing the fitted line in a log[peak-torque]-angular-velocity diagram. The regression line may be fitted by three or more angular velocities. In principle, IMS30 is independent of the chosen angular velocity, which makes it possible to use an estimated value as IMS30 to compare results, studies and centres in between.