Telephone reminders are a cost effective way to improve responses in postal health surveys

Salim Silva, M
Smith, W T
Bammer, G

¹National Centre for Epidemiology and Population Health, The Australian National University, ACT 0200, Australia
²Federation of Industries of the State of Minas Gerais, Brazil and Secretariat of Health of the State of Minas Gerais, Brazil

Correspondence to:Marluce Salim Silva, National Centre for Epidemiology and Population Health, The Australian National University, ACT 0200, Australia; [email protected]

Accepted 03 July 2001

Marluce Salim Silva is a PhD candidate sponsored by the Federation of Industries and the Secretariat of Health of the State of Minas Gerais, Brazil. Valuable assistance was received from Dr Jim Butler, Dr Keith Dear, Ms Karen Lees, Dr John Corry, Dr Geoffrey Speldewinde and Ms Karen Alarcon.

Journal of Epidemiology & Community Health 56(2):p 115-118, February 2002.

Study objective: To assess the effectiveness of a telephone reminder in increasing responses to postal surveys and to calculate the differential costs per completed questionnaire.

Design: Randomised controlled trial.

Setting: Australian university and rehabilitation medicine practice.

Participants: The trial was conducted in 1999 among the 143 non-respondents to a questionnaire about work related neck and upper body disorders. The questionnaire was sent to two Australian female samples: 200 office workers (Sample A) and 92 former rehabilitation medicine patients (Sample B). A reminder letter, another copy of the questionnaire and a final letter were sent at two week intervals. Half of the non-respondents within each sample were randomly selected to receive a telephone reminder just after the second mailout of the questionnaire. All direct costs were calculated.

Main results: Responses were significantly higher among those who received the telephone reminder intervention (relative risk 2.54, 95% confidence intervals 1.43 to 4.52). Analysed by intention to phone, 47% of non-respondents in Sample A and 38% in Sample B returned a complete questionnaire after the intervention, compared with 21% and 10%, respectively, in the control groups. For the 112 women (combined samples) who returned completed questionnaires before randomisation, the average cost per respondent was AUD14. There was a higher total cost for the intervention groups (AUD851 versus AUD386 for controls), but the significantly higher number of additional completed responses (31 versus 12) resulted in a 15% lower marginal cost per completed questionnaire in those groups.

Conclusion: Telephone reminders are cost effective in improving responses to postal surveys.