Collecting saliva samples by mail.

Collecting saliva samples by mail can serve numerous purposes in epidemiologic research. The objectives of this study were to assess what proportion of participants in a mail survey would provide a saliva sample and whether incentives could improve participation. In 1995, 2,994 students, faculty, and staff members of Geneva University, Geneva, Switzerland, were randomized to receive, together with a mailed questionnaire about smoking, a saliva vial, a ballpoint pen, the offer of a lottery, or any combination of these. After one mailing and a reminder letter, response rates were 52% among those who had been requested to provide saliva and 63% among controls (p < 0.001). In the former group, most respondents (98%) provided a saliva sample. Incentives improved participation only among those who were asked to provide saliva (lottery: +11% response, p = 0.003; pen: +6% response, p = 0.1). The final participation, after up to three reminders, was 76% overall. The authors conclude that while the collection of saliva samples by mail is feasible it tends to decrease response rates.