Predictors of Making a Referral to Child Protective Services Prior to Expert Consultation

Research electronic data capture (REDCap) is a novel workflow methodology and software solution designed for rapid development and deployment of electronic data capture tools to support clinical and translational research. We present: (1) a brief description of the REDCap metadata-driven software toolset; (2) detail concerning the capture and use of study-related metadata from scientific research teams; (3) measures of impact for REDCap; (4) details concerning a consortium network of domestic and international institutions collaborating on the project; and (5) strengths and limitations of the REDCap system. REDCap is currently supporting 286 translational research projects in a growing collaborative network including 27 active partner institutions.

The Research Electronic Data Capture (REDCap) data management platform was developed in 2004 to address an institutional need at Vanderbilt University, then shared with a limited number of adopting sites beginning in 2006. Given bi-directional benefit in early sharing experiments, we created a broader consortium sharing and support model for any academic, non-profit, or government partner wishing to adopt the software. Our sharing framework and consortium-based support model have evolved over time along with the size of the consortium (currently more than 3200 REDCap partners across 128 countries). While the "REDCap Consortium" model represents only one example of how to build and disseminate a software platform, lessons learned from our approach may assist other research institutions seeking to build and disseminate innovative technologies.

We review three common methods to estimate predicted probabilities following confounder-adjusted logistic regression: marginal standardization (predicted probabilities summed to a weighted average reflecting the confounder distribution in the target population); prediction at the modes (conditional predicted probabilities calculated by setting each confounder to its modal value); and prediction at the means (predicted probabilities calculated by setting each confounder to its mean value). That each method corresponds to a different target population is underappreciated in practice. Specifically, prediction at the means is often incorrectly interpreted as estimating average probabilities for the overall study population, and furthermore yields nonsensical estimates in the presence of dichotomous confounders. Default commands in popular statistical software packages often lead to inadvertent misapplication of prediction at the means. Using an applied example, we demonstrate discrepancies in predicted probabilities across these methods, discuss implications for interpretation and provide syntax for SAS and Stata. Marginal standardization allows inference to the total population from which data are drawn. Prediction at the modes or means allows inference only to the relevant stratum of observations. With dichotomous confounders, prediction at the means corresponds to a stratum that does not include any real-life observations. Marginal standardization is the appropriate method when making inference to the overall population. Other methods should be used with caution, and prediction at the means should not be used with binary confounders. Stata, but not SAS, incorporates simple methods for marginal standardization. © The Author 2014; all rights reserved. Published by Oxford University Press on behalf of the International Epidemiological Association.