Electronic reminders did not improve postal questionnaire response rates or response times: a randomized controlled trial Mei-See Man*, Helen E. Tilbrook, Shalmini Jayakody, Catherine E. Hewitt, Helen Cox, Ben Cross, David J. Torgerson Department of Health Sciences, York Trials Unit, The University of York, Heslington, York YO10 5DD, UK Accepted 17 October 2010 Abstract Objective: We aim to evaluate the effectiveness of electronic reminders (ERs) to improve the response rates and time to response of postal questionnaires in a health research setting. Study Design and Setting: This pragmatic randomized controlled trial (RCT) was nested within a multicenter RCT of yoga for lower back pain. Participants who provided an electronic mail address and/or mobile phone number were randomized to receive an ER or no reminder (controls) on the day they were due to receive a follow-up questionnaire. Results: One hundred twenty-five participants (32 males and 93 females) mean age 46 (standard deviation: 11, range: 20e65) were randomized to ER (n 5 62) or controls (n 5 63). Overall 85.6% of participants returned postal questionnaires (87.1% ER group and 84.1% from controls). No significant differences were found between the two groups for response rate (difference between groups 5 3.0%, 95% confidence interval [CI] 5 10, 16; P 5 0.64) or time to response after adjusting for age, gender, and treatment allocation (c 2 [3df] 5 7.10; P 5 0.07). Conclusion: In the present RCT, we found little evidence for the effectiveness of ERs to increase response rates or time to respond for the return of questionnaires in this study population group. Ó 2011 Elsevier Inc. All rights reserved. Keywords: Reminder system; Data collection; Randomized controlled trial; Research methodology; Short messenger service (sms); Electronic mail (e-mail) 1. Introduction A poor response rate is a perpetual concern for health researchers collecting data using self-administered postal questionnaires. Non-return of questionnaires can have sev- eral confounding effects, including the introduction of bias and reducing the effective sample size, which can poten- tially invalidate the study [1]. Despite this, postal question- naires are commonly used because of being relatively inexpensive, efficient to administer, and can reach a wide- ranging geographical region. Reminders are an effective way of improving question- naire response rates [2], and the widespread ownership and common usage of mobile telephone text messaging (short message service [SMS]) and electronic mail (e-mail) provides a potential alternative method for reminding study participants. A recent Cochrane review [3] found a single randomized controlled trial (RCT) [4] comparing SMS with postcard reminders. SMS reminders improved survey response rates in persistent nonresponders in rural Finland, although the time to response was not investigated. A recent study from our group [5] assessed whether elec- tronic reminders (ERs) improved questionnaire response rate and time to response within the context of an established RCT of food elimination diet for the prevention of migraine. From a total of 148 participants, ER significantly reduced the time to response by approximately 4 days (P 5 0.02), and improved response rates by 5.4% (95% CI 5 4.6, 15.4%), but this difference was not statistically significant. Given that there are few trials in this area, the present study aims to repeat the latter study [5] in a different patient population, clarify any improvements in response rates, and assess the hypothesis that ER reduces time to response. Funding: This study was funded by the York Trial Unit at the Univer- sity of York. Ethics approval: The current trial was nested within the multi-centred randomised trial of yoga for chronic low back pain, which had received ethical approval from the Leeds East Research Ethics Committee, UK. (reference 07/Q1206/35). Provenance and peer review: Not commissioned. . Competing interest statement: none. * Corresponding author. Tel.: þ44-1904-321-664; fax: þ44-1904-321- 387. E-mail address: mm714@york.ac.uk (M.S. Man). 0895-4356/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi: 10.1016/j.jclinepi.2010.10.013 Journal of Clinical Epidemiology 64 (2011) 1001e1004