Most investigators experienced in performing surveys in the medically ill believe that questionnaire packets should be as brief as possible to reduce the refusal rate for participation in the study, the drop out rate among those who initially agree to participate, and the rate of missing data. There is no empirical basis for deciding the optimal length, and a variety of factors may influence the feasibility of longer or shorter questionnaire packets.

Table11.1: Factors that Influence Decisions about the Optimal Length of a Questionnaire Packet Briefer Longer No access to dedicated data collector Questionnaire must be completed in one sitting On-site data collection only Patients have severe medical or psychiatric comorbidities Trained, dedicated data collector May be completed over 2-3 contacts Questions may be administered by interview, in person or by telephone if desired Patients have mild medical or psychiatric comorbidities