This study was an attempt to confirm the results of an earlier experiment in which the benefit of an education programme was assessed in a group of people with diabetes complicated by ulceration of the foot. Foot ulcers are the source of considerable suffering and cost and carry a high risk of amputation: they are difficult to heal and approximately 40% recur in the first 12 months. The earlier experiment (published by Malone and colleagues in 1989) indicated that a single hour-long education session appeared to lead to a three-fold reduction in the numbers of ulcers which recurred after successful treatment. It was not possible to confirm these findings in the present study in which patients from three specialist clinics in Nottingham and Derby, UK, were allocated either to receive a one-to-one, individually targeted, education programme in the own home (and reinforced after one month by a telephone call), or to receive usual care. The group who received the education reported better recommended foot care behaviour (intended to minimise the risk of injury) at 12 months but despite this, there was no difference between the two groups in the percentage who suffered either a new ulcer (41% education versus 41% usual care) or amputation (10% and 11%, respectively). While the benefit of education is undeniable in general, it was not possible to show that this particular teaching session had an impact on the occurrence of new disease in this group of patients.

This was a randomised controlled trial of the effect over 12 months of a one-to-one educational package designed to improve foot care behaviour in a group of patients at the highest risk of new foot disease: those with a recently healed ulcer. The primary outcome was the incidence of new ulceration at 12 months. Secondary outcomes included self-reported footcare behaviour, new ulceration at 6 months, amputation (major or minor) at 12 and 6 months, and measures of well-being (SF36; Diabetic Foot Ulcer Scale, DFS short-form). The educational intervention was based on the content of the International Consensus on the Diabetic Foot, 2003, was modified following discussion with a focus group of former ulcer sufferers, and included graphic images of different forms of foot disease. Although structured, the information was targeted in discussion to the individual's circumstances, and was delivered one-to-one in the patient's own home by a single health care professional. It was reinforced after one month by a semi-structured phone conversation with the same educator. Outcomes were determined by postal questionnaire, backed up with reminders and phone calls when necessary, as well questionnaires sent to GP staff and by searching hospital records. Foot care behaviour was assessed at 12 months using a newly devised and validated measure of footcare function (NAFF) which was also delivered by post. All outcomes were assessed by researchers blind to allocation group. Mean age was 64.9 (SD 10.9) years, and there were no differences between groups at baseline in terms of demographics, diabetes type and complications. The incidence (%) of new ulcers at 6 and 12 months was 26 (30%) and 36 (41%) in the intervention group and 18 (21%) and 35 (41%) in the controls, RR 0.997 (CI 0.776-1.280). There were no differences in any of the other outcome measures although the intervention group reported a higher score in recommended foot care behaviour when compared with controls (42.0 versus 38.7, p=0.03).

Inclusion Criteria:

• Patients with diabetes and an ulcer of the foot which had been managed at one of the three participating centres and which had healed within the preceding 3 months, leaving them ulcer-free for at least 28 days

Exclusion Criteria:

• Living in institutional care
• Documented history of dementia
• Other serious medical problems
• Non-English speaking and without an English-speaking carer
• Lived more than 50 miles from the centre
• Involved in another study