Improving return rates for health-care outcome

The objective was to test two methods of reducing loss of data in follow-up by abbreviating the questionnaire and offering a token monetary incentive. Primary data were collected between September 14, 1998 and July 23, 1999 from an adult psychiatric population, representing about 50% of all patients, all of whom agreed to participate in a quality-assurance monitoring program. Briefer components of the SF-36 Health Status Survey were expected to yield higher return rates than would the SF-36 plus a form assessing amount of treatment received. Small monetary incentives were expected to result in higher return rates also. Patients were randomly assigned to the groups (n = 73 each). Abbreviation of the questionnaire and use of a monetary incentive each reduced attrition by 10%. A diminishing returns effect for both methods was observed. Further study of a possible interaction between the questionnaire's length and monetary incentive is recommended.