The influence of weekly locomotion scoring and, thus, early detection and treatment of lame cows by a veterinarian on lameness prevalence, incidence, duration of lameness, fertility and milk yield on one dairy farm in Northern Germany. Cows were distributed to two groups. Cows in Group A (n = 99) with a locomotion score (LS) > 1 were examined and treated. In Group B (n = 99), it was solely in the hands of the farmer to detect lame cows and to decide which cows received treatment. Four weeks after the beginning of the experimental period, the prevalence of cows with LS = 1 was higher in Group A compared with Group B. Prevalence of lame cows (LS > 1) increased in Group B (47.6% in Week 2 to 84.0% in Week 40) and decreased in Group A from Week 2 to Week 40 (50% to 14.4%; P < 0.05). Within groups, the monthly lameness incidence did not differ. The average duration of lameness for newly lame cows was 3.7 weeks in Group A and 10.4 weeks in Group B (P < 0.001). There was no effect on fertility and incidence of puerperal disorders. The 100-day milk yield was calculated from cows having their first four Dairy Herd Improvement (DHI) test day results during the experimental period. The mean 100-day milk yield tended to be higher in Group A compared with Group B (3,386 kg vs. 3,359 kg; P = 0.084).