Challenges using online surveys in a Danish population of people with type 2 diabetes

This is the authors' abstract. We don't add key points for this paper.

OBJECTIVES: To investigate response rates for online and paper versions of an identical questionnaire and the differences between respondents to each and between respondents and non-respondents among a population with type 2 diabetes. METHODS: We mailed letters containing an invitation to complete an online questionnaire to 2045 individuals, followed by two reminders; the second included a paper version of the questionnaire. RESULTS: In total, 1081 people responded to either version of the questionnaire, yielding a response rate of 54%. Compared to total respondents, respondents completing the online version were more likely to be male, better educated, and younger, and have had diabetes for a shorter period of time. Compared to non-respondents, respondents were more likely to be male and have a lower hemoglobin A1c level. CONCLUSION: Web-based surveys are capable of delivering a substantial number of responses cost-effectively. However, disadvantages related to selection bias should be taken into account, and mixed-mode methods should be considered when surveying populations with type 2 diabetes.