Does pre-notification increase questionnaire response rates: a randomised controlled trial nested within a systematic review

BMC Medical Research Methodology

Benjamin Woolf¹  Phil Edwards² 

[1] Department of Psychological Science, University of Bristol, 5 Priory Road, Bristol, UK;Medical Research Council Integrative Epidemiology Unit, University of Bristol, Bristol, UK;Faculty of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, London, UK;Faculty of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, London, UK;

关键词: Pre-notification;    Randomised controlled trials;    Questionnaire response;   

DOI  :  10.1186/s12874-021-01462-z

来源: Springer

【 摘 要 】

BackgroundMissing outcome data can lead to bias in the results of systematic reviews. One way to address missing outcome data is by requesting the data from the trial authors, but non-response is common. One way to potentially improve response rates is by sending study participants advance communication. During the update of a systematic review examining the effect of pre-notification on response rates, study authors needed to be contacted for further information. This study was nested within the systematic review by randomising authors to receive a notification of the upcoming request for information. The objective was to test if pre-notification increased response rates.MethodsThe participants were study authors included in the systematic review, whose studies were at unclear risk of bias. The intervention was a pre-notification of the request for further information, sent 1 day before the request. The outcome was defined as the proportion of authors who responded to the request for information. Authors were randomised by simple randomisation. Thirty three authors were randomised to the pre-notification arm, and 42 were randomised to the control arm. Authors were blinded to the possibility of an alternative condition.ResultsAll authors randomised were analysed. 14/33 (42.4%) authors in the pre-notification arm had returned responses to the questionnaire, and 18/42 (42.9%) in the control arm. There was no evidence of a difference between these groups (absolute difference = − 0.5, 95% CI (− 23.4 to 22.5%), p = 1). We received no complaints about receiving the pre-notification.ConclusionsThis study’s results do not support the hypothesis that pre-notification increases response from study authors being contacted for a request for more information. However, the study has a low power, and the results may not generalise to other contexts, methods of administering a pre-notification, or study populations.Trial registrationRegistration and protocol:This trial is not registered with any trial registry. However, the protocol was posted in advance on the Open Science Framework website and is available on the Open Science Framework website: DOI: https://doi.org/10.17605/OSF.IO/MSV2W or https://osf.io/msv2w/

【 授权许可】

CC BY