A RANDOMIZED CONTROLLED TRIAL OF DIETARY-FAT REDUCTION - THE RETENTION OF SUBJECTS AND CHARACTERISTICS OF DROP OUTS

We have examined the feasibility of carrying out a randomized controlled trial of dietary fat reduction in women at increased risk for breast cancer. The randomization was either to a control group who were taught the principles of balanced nutrition, but were not counselled to change their fat intake, or to an intervention group who were taught to reduce their dietary fat intake to 15% of total calories from a baseline average of 35% of calories. Potentially eligible subjects were women attending a breast diagnostic clinic who had the mammographic pattern of dysplasia. Subjects were recruited by letter from their referring surgeon followed by a telephone call. Subjects interested in participating in the study then entered by one of two phases. In Phase I, the study was explained, informed consent sought and willing subjects randomized to the intervention or control group. Using this procedure 227 subjects were randomized and 48 (21%) dropped out of the study in the 12 months following randomization. (A drop out was defined as a subject who persistently failed to keep appointments and provide nutrient data.) Of these drop outs, 30 (63%) occurred at or soon after randomization. A modified procedure of entry was then adopted in which subjects interested in the study were first taught the procedures involved, including keeping food records and clinic appointments, and were then asked to provide consent and randomized. Two hundred and eighty subjects were enrolled using this modified procedure and 25 (9%) have dropped out in the 12 months following randomization. The use of a period in which subjects were taught and practised the procedures required in the trial thus substantially reduced the number who dropped out after randomization.