Frank 2003.

Methods	RCT
Participants	100 people with type 2 diabetes mellitus randomised: intervention 50 vs control 50 Baseline risk for foot ulceration: Current smoking. intervention 12.5% vs control 16.7% Mean HbA1c: intervention 7.44 vs control 7.66, P = 0.559 Mean score neuropathy screening questionnaire (0 to 13): intervention 2.46 vs control 2.46, P = 1.00 Mean number of positive sensations of a monofilament on prespecified locations on the foot (0 to 8): intervention 6.06 vs control 5.38, P = 0.215 Baseline outcome measures: Foot care knowledge scores: intervention 18.65 (SD 2.65) vs control 17.50 (SD 3.14), P = 0.056 Patients' behaviour assessment: checking feet: intervention 5.21 vs control 4.52, P = 0.243 washing feet: intervention 5.17 vs control 5.42, P = 0.572 applying lotion: intervention 4.54 vs control 4.19, P = 0.560 wearing shoes and socks: intervention 3.71 vs control 4.92, P = 0.057 Study setting: primary care (mostly indigent) people with type 2 diabetes visiting a podiatrist in 1 of 2 designated community health centres associated with the Indiana University School of Medicine in Indianapolis, Indiana Inclusion criteria: > 65 years of age, no previous foot or leg amputation, access to a working telephone, able to understand English
Interventions	Intervention group: Lower extremity amputation risk assessment. Content: use of a monofilament Foot care videotape. Content: people demonstrating proper foot care Bag of foot supplies. Content: soap, towel, socks, mirror, toenail clippers, lotion samples, information on smoking cessation and exercise Hand‐out. Content: foot care instructions 30‐ to 40‐minute individualised education session by research nurse. Content: persuasion to perform foot care + demonstration of content of bag of foot supplies Reminder checklist. Content: instructions for daily foot care Weekly reminder telephone calls. Content: persuasion to perform foot care Care as usual by a podiatrist Control group: Lower extremity amputation risk assessment. Content: use of a monofilament Weekly telephone calls. Content: only outcome assessment Care as usual by a podiatrist Adherence: no data provided, but likely that all intervention group people received the single brief educational session directly after randomisation
Outcomes	Primary outcomes: not reported Secondary outcomes: foot care knowledge scores, patients' self‐reported foot care behaviour scores
Duration and completion of follow‐up	4 weeks; 96 people completed follow‐up intervention 48 vs control 48
Types of assessment	Foot care knowledge: 26 items, with a ''true'', ''false'' or ''don't know'' answer (range 0 to 26) Behaviour scores: retrospectively self‐reported. Data collection during weekly telephone calls. Foot care behaviours of interest were checking feet daily for injury, washing and drying feet daily, applying lotion to the feet daily, wear socks and shoes, trimming toenails weekly. Results for each item were presented by the mean number of days per week that people adhered to the desired behaviour (range 0 to 7 per item)
Notes	It was originally intended to report changes in 'weekly trimming of toenails', but this was abandoned, as all people were seen by a podiatrist for trimming of their toenails
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Papers with either 'group A' or 'group B' were drawn from an envelope
Allocation concealment (selection bias)	High risk	The envelope was not sealed
Blinding (performance bias and detection bias) Blinding of outcome assessors	High risk	Outcomes were assessed by the research nurse, who also performed the educational intervention
Incomplete outcome data (attrition bias) All outcomes	High risk	96 of 100 people completed 4 weeks' follow‐up (intervention 48 vs control 48) No ITT analyses were undertaken
Selective reporting (reporting bias)	Low risk	No study protocol available, but the trial report lists the outcomes of interest in both the methods and the results section
Other bias	Unclear risk	Baseline risk for foot ulceration: Current smoking. intervention 12.5% vs control 16.7% Mean HbA1c: intervention 7.44 vs control 7.66, P = 0.559 Mean score neuropathy screening questionnaire (0 to 13): intervention 2.46 vs control 2.46, P = 1.00 Mean number of positive sensations of a monofilament on prespecified locations on the foot (0 to 8): intervention 6.06 vs control 5.38, P = 0.215 Co‐interventions were not described Adherence: no data provided, but likely that all intervention group people received the single brief educational session directly after randomisation