Eliciting relative preferences for two methods of colorectal cancer screening

Two methods of mass population screening for colorectal cancer - faecal occult blood testing and sigmoidoscopy - have been the subject of randomized controlled trials in the UK. A national screening programme is currently under consideration and the choice of screening method remains open. To be successful, a programme will require high levels of uptake, and uptake is likely to depend upon subjects' attitudes towards the screening method introduced. Although a preferred screening method has already been identified from a questionnaire survey, we undertook a further interview study (n = 106), with a view to comparing the results of two different approaches to eliciting public preferences. In comparison with the questionnaire study, a higher proportion of interview subjects stated a preference. Interview subjects were generally more favourably disposed towards sigmoidoscopy, excepting those with previous experience. Compared with the questionnaire survey, the interviews provided richer information on the reasons for preferences offered. Individual preferences were evidently subjective and dependant on attitudes towards a variety of method characteristics, such as discomfort, convenience and perceived sophistication. Characteristics such as age and low income, which had predicted preferences in the questionnaire study, predicted preferences in the interview study also. The difference between the results obtained by the different elicitation techniques can be explained in terms of the differential provision of information and sample selection. Conclusions made about public preferences are likely to depend on the technique employed in eliciting them.