To study the hip muscle strength and cross sectional area (CSA) in men with hip osteoarthritis (OA) compared to age and sex matched healthy controls.Based on the American College of Rheumatology criteria regarding classification of hip OA, 27 men (aged 47-64 yrs) with unilateral or bilateral hip OA and 30 age matched randomly selected healthy male controls were studied. The maximal isometric hip abductor, adductor, flexor, and extensor strength (Nm) at 0 degree of hip flexion in the supine position was determined with a dynamometer. The isokinetic hip flexion and extension strength (peak torque, Nm) was determined using angular velocities of 60 degrees /s and 120 degrees /s. The subjective severity of hip pain was rated by visual analog scale prior to the muscle strength test. CSA of the pelvic and thigh muscles was measured from magnetic resonance images.The reliability of intraclass correlation coefficients for repeated measures of muscle strength varied from 0.70 to 0.94 in controls and from 0.84 to 0.98 in subjects with OA. Hip isometric adductor and abductor strength was 25% and 31% lower (p < 0.001) in OA subjects than in controls, respectively. The hip isometric and isokinetic flexion strength was 18-22% lower (p < 0.01) in OA subjects than in controls, but extension strength did not differ between groups. In OA subjects, the hip flexion and extension isometric and isokinetic strength values were 13-22% lower (p < 0.05) on the more deteriorated side compared to the better side. CSA of the pelvic and thigh muscles did not differ between the groups. However, in OA subjects, the CSA of the pelvic and thigh muscles was 6-13% less (p < 0.05 to < 0.001) on the more severely affected hip compared to the better hip.Men with hip OA have significantly lower abduction, adduction, and flexion muscle strength than controls. The decrease of muscle size and hip pain may contribute to the decrease of muscle strength in hip OA. Other possible underlying causes of the muscle weakness need to be studied.