BACKGROUND: Podoconiosis (also known as endemic, non-filarial elephantiasis) affects about 4 million subsistence farmers in tropical Africa. Poor awareness of the condition and inadequate evidence for the efficacy of treatment mean that no government in an endemic country yet offers lymphoedema management for patients with podoconiosis. Among patients with filarial lymphoedema, trials suggest that limb care is effective in reducing the most disabling sequelae: episodes of acute dermatolymphangioadenitis. We aimed to test the hypothesis that a simple, inexpensive lymphoedema management package would reduce the incidence of acute dermatolymphangioadenitis in adult patients with podoconiosis in northern Ethiopia. METHODS: We did a pragmatic randomised controlled trial at health posts and health centres in 18 sub-districts of Aneded woreda (district) in Amhara, northern Ethiopia. Participants were adults aged 18 years and older, had a diagnosis of at least stage 2 podoconiosis (persistent lymphoedema) and a negative antigen test for filariasis, and intended to remain within Aneded woreda for the duration of the trial. Patients were randomly assigned (1:1) to either receive a package containing instructions for foot hygiene, skin care, bandaging, exercises, and use of socks and shoes, with support by lay Community Podoconiosis Agents at monthly meetings (intervention group) or to receive no intervention (control group). Participants were aware of their group assignment, but researchers doing all analyses were masked to treatment group. The primary outcome was incidence of acute dermatolymphangioadenitis episodes in the total period of observation of each participant, measured by use of validated patient self-reported pictorial diaries. This trial was registered with the International Standard Randomised Controlled Trials Number Register, number ISRCTN67805210. FINDINGS: Between Dec 1, 2014, and June 30, 2015, 1339 patients were screened, and 696 patients were enrolled and randomly allocated to treatment groups. We allocated 350 patients to the intervention group and 346 patients to the control group. 321 (92%) patients from the intervention group and 329 (95%) patients from the control group provided follow-up results at 12 months. During the 12 months of follow-up, 16 550 new episodes of acute dermatolymphangioadenitis occurred during 765·2 person-years. The incidence of acute dermatolymphangioadenitis was 19·4 episodes per person-year (95% CI 18·9-19·9) in the intervention group and 23·9 episodes per person-year (23·4-24·4) in the control group. The ratio of incidence rate in the intervention group to that of the control group was 0·81 (0·74 to 0·89; p<0·0001), with a rate difference of -4·5 (-5·1 to -3·8) episodes per person-year. No serious adverse events related to the intervention were reported. INTERPRETATION: A simple, inexpensive package of lymphoedema self-care is effective in reducing the frequency and duration of acute dermatolymphangioadenitis. We recommend its implementation by the governments of endemic countries. FUNDING: Joint Global Health Trials scheme (from the Wellcome Trust, the UK Medical Research Council, and UK Aid).