The aim of this study was to evaluate the effectiveness of three knowledge-transfer intervention trial types (postal, group, one-to-one) to promote best practice to treat sheep with footrot. Further aims were to investigate whether farmer behaviour (i.e. management of lameness) before the trial was associated with uptake of best practice and whether the benefits of best practice framed positively or negatively influenced change in behaviour. The intervention was a message developed from evidence and expert opinion. It was entitled "Six steps to sound sheep" and promoted (1) catch sheep within three days of becoming lame, (2) inspect feet without foot trimming, (3) correctly diagnose the cause, (4) treat sheep lame with footrot or interdigital dermatitis with antibiotic injection and spray without foot trimming, (5) record the identity of treated sheep, (6) cull repeatedly lame sheep. In 2013, 4000 randomly-selected English sheep farmers were sent a questionnaire, those who responded were recruited to the postal (1081 farmers) or one-to-one intervention (32 farmers) trials. A random sample of 400 farmers were invited to join the group trial; 78 farmers participated. A follow-up questionnaire was sent to all participants in summer 2014. There were 72%, 65% and 91% useable responses for the postal, group and one-to-one trials respectively. In both 2013 and 2014, the prevalence of lameness was lower in flocks managed by LC1 farmers than LC2 and LC3 farmers. Between 2013 and 2014, the reduction in geometric mean (95% CI) period prevalence of lameness, proportional between flock reduction in lameness and within flock reduction in lameness was greatest in the one-to-one (7.6% (7.1-8.2%) to 4.3% (3.6-5.0%), 35%, 72%) followed by the group (4.5% (3.9-5.0%) to 3.1% (2.4-3.7%), 27%, 55%) and then the postal trial (from 3.5% (3.3-3.7%) to 3.2% (3.1-3.4%), 21%, 43%). There was a marginally greater reduction in lameness in farmers using most of Six steps but slow to treat lame sheep pre-trial than those not using Six steps at all. There was no significant effect of message framing. The greatest behavioural change was a reduction in therapeutic and routine foot trimming and the greatest attitude change was an increase in negative attitudes towards foot trimming. We conclude that all three intervention trial approaches were effective to promote best practice to treat sheep with footrot with one-to-one facilitation more effective than group and postal intervention trials. Results suggest that farmers' behaviour change was greater among those practising aspects of the intervention message before the trial began than those not practising any aspect.