@Article{info:doi/10.2196/jmir.3898,
author="Turnbull, Alison E
and O'Connor, Cristi L
and Lau, Bryan
and Halpern, Scott D
and Needham, Dale M",
title="Allowing Physicians to Choose the Value of Compensation for Participation in a Web-Based Survey: Randomized Controlled Trial",
journal="J Med Internet Res",
year="2015",
month="Jul",
day="29",
volume="17",
number="7",
pages="e189",
keywords="data collection; monetary incentives; cash; physicians; electronic questionnaire; survey design; response rate",
abstract="Background: Survey response rates among physicians are declining, and determining an appropriate level of compensation to motivate participation poses a major challenge. Objective: To estimate the effect of permitting intensive care physicians to select their preferred level of compensation for completing a short Web-based survey on physician (1) response rate, (2) survey completion rate, (3) time to response, and (4) time spent completing the survey. Methods: A total of 1850 US intensivists from an existing database were randomized to receive a survey invitation email with or without an Amazon.com incentive available to the first 100 respondents.  The incentive could be instantly redeemed for an amount chosen by the respondent, up to a maximum of US {\$}50. Results: The overall response rate was 35.90{\%} (630/1755).  Among the 35.4{\%} (111/314) of eligible participants choosing the incentive, 80.2{\%} (89/111) selected the maximum value.  Among intensivists offered an incentive, the response was 6.0{\%} higher (95{\%} CI 1.5-10.5, P=.01), survey completion was marginally greater (807/859, 94.0{\%} vs 892/991, 90.0{\%}; P=.06), and the median number of days to survey response was shorter (0.8, interquartile range [IQR] 0.2-14.4 vs 6.6, IQR 0.3-22.3; P=.001), with no difference in time spent completing the survey. Conclusions: Permitting intensive care physicians to determine compensation level for completing a short Web-based survey modestly increased response rate and substantially decreased response time without decreasing the time spent on survey completion. ",
issn="1438-8871",
doi="10.2196/jmir.3898",
url="http://www.jmir.org/2015/7/e189/",
url="https://doi.org/10.2196/jmir.3898",
url="http://www.ncbi.nlm.nih.gov/pubmed/26223821"
}