on behaviour change is easier to achieve (although relevant disease-specific and validated measures of behaviour do not currently exist) and would mean more in terms of likely clinical benefit than the demonstration of increased knowledge.
However, the induction of behaviour change is itself of no value unless it is proved that it improves clinical well-being.
There are two main clinically relevant end points: reduction in ulcer incidence and improved outcome of ulcers (or other foot problems) that arise.
Measures of patient well-being, emotion, function and satisfaction are appropriate secondary end points, as are measures of cost.
Sample Size Ragnarson Tennvall and Apelqvist53 have calculated that if any intervention is to be cost- effective, it should reduce incidence by at least 25%.
The number of people needed for study is determined by the outcome measure chosen.
The size of the population is also determined by the level of risk and the size of the hoped-for benefit.
Studies of specific behaviour change can be relatively small, whereas studies with outcomes of direct clinical relevance have to be very much larger.
Since the incidence of new ulceration is only of the order of 2% per year (even when the younger population is excluded), the size of studies that use either ulcer incidence or ulcer outcome as their primary end point may require such large populations that they are either logistically impossible or prohibitively expensive , even when very simple interventions are used.
Any study designed to assess the effect of education in ulcer prevention would require something of an order of 10 000 relatively older subjects in order to have 200 incident events in 1 year.
However, the incidence of new ulcers is higher in those who are known to be at increased risk (e.g. those with neuropathy) and the study size in this group would be commensurately (approximately three times) smaller.
The highest incidence is observed in those who have had ulcers that have only recently healed, and if this population was chosen for study, the population needed may be only of the order of 200.
CONCLUSION There needs to be a systematic approach to foot care education that ensures that appropriate foot care advice and information is available to all people with diabetes.
Those identified as being at increased risk of developing ulcers should receive education designed to change behaviour with the result that there is reduced ulcer incidence and improved ulcer outcome.
The programme adopted should address the needs of the individuals being targeted, and those delivering it require special training.
None of this can, however, be properly implemented until further research has been undertaken.
This research is needed to define with greater precision the extent to which any educational strategy is accompanied by behaviour change, the extent to which this change in behaviour achieves the desired clinical effect and whether the magnitude of the benefit would justify the cost of implementation.