In England in 2013, a survey of 4,000 sheep flocks indicated that the geometric mean flock prevalence of lameness in sheep fell from 5.4% in 2004 to 3.4% in 2013. The reduction was associated with changes in management of lameness. Most lameness is due to footrot (caused by Dichelobacter nodosus), but contagious ovine digital dermatitis is increasing in prevalence. In 2015, 162 farmers, a subset of respondents to the 2013 survey, completed a questionnaire and the geometric mean flock prevalence of lameness was calculated. A multivariable quasi-Poisson regression model was used to identify managements associated with the prevalence of lameness in ewes and the population attributable fractions (PAF) of these were calculated. The geometric mean prevalence of lameness in ewes in 2015 had increased to 4.2%. Managements associated with a higher prevalence of lameness were routine foot trimming when ≥5% of sheep bled (compared with not routine foot trimming) and mixing sheep with other flocks. Managements associated with a decreased prevalence of lameness were treating sheep within three days of becoming lame and annual vaccination with FootvaxTM for >5 years compared with not using the vaccine. The model explained 65.3% of the variation in prevalence of lameness. The PAFs were: 34.7% for not vaccinating for >5 years, 25.3% for >3 days to treat lame sheep, 2.9% for causing ≥5% of sheep to bleed during routine foot trimming and 2.4% for mixing with other flocks. The percentage of farmers treating lame sheep within three days and practising routine foot trimming decreased between 2013 and 2015 and vaccination use increased. We conclude that whilst annual vaccination contributes to reducing the percentage of lameness, this takes >5 years to observe an effect whereas rapid treatment of lame sheep would lead to an almost immediate effect on reducing the prevalence of lameness in England.