Source Data Verification in a Critical Care Academic Observational Study
CCCF ePoster library. Topp-Nguyen N. Oct 28, 2015; 117350; P115 Disclosure(s): This work was funded by a Strategic Grant from Hamilton Health Sciences to AFR
Nam Topp-Nguyen
Nam Topp-Nguyen
Login now to access Regular content available to all registered users.

You may also access this content "anytime, anywhere" with the Free MULTILEARNING App for iOS and Android
Rate & Comment (0)

Topic: Other

Source Data Verification in a Critical Care Academic Observational Study

Nam Topp-Nguyen, S. Rotella, C. Wardell, D. Jain, M. Xu, A. Fox-Robichaud, E. McDonald

Life Sciences, University of Toronto, Toronto, Canada | Critical Care, Hamilton Health Sciences, Hamilton, Canada | Life Sciences, McMaster University, Hamilton, Canada | Michael Degroote School of Medicine, McMaster University, Hamilton, Canada | Health Research Methodology Program, McMaster University, Hamilton, Canada | Medicine, McMaster University, Hamilton, Canada | Medicine, McMaster University, Hamilton, Canada


Source data verification (SDV) is the process of reviewing and confirming that reported research data is true. Industry sponsored studies usually perform on-site monitoring for all data resulting in large costs1. In academia where resources are scarce more cost-effective and innovative alternatives for monitoring are required. There are currently no guidelines for this process.

Objectives: To use a limited central SDV method to verify that crucial data collected in our observational ICU study is of the highest quality possible thus ensuring the generalizability of our findings.


We retrospectively reviewed a convenience sample of 10 patients (20% of the planned sample size) enrolled in an ICU pathogen identification pilot observational study. We identified 6 clinically relevant data categories for monitoring: Inclusion Criteria, Exclusion Criteria, Past Medical History, ICU and Hospital Discharge Dates, Routine and Research Specimens (sputum and blood) Collection Dates and Final Status (dead or alive). In order to ensure a representative and unbiased sample was collected we used a computerized randomization tool to determine the patient selection. The Study Research Coordinator abstracted and entered the data into REDCap. Using a standard operation procedure for data capture, a trained data collection monitor, compared each patient’s source data, hospital records with corresponding REDCap study database entry. Source data and study data were compared and we calculated the percentage agreement, a kappa statistic for each value. Discrepancies in data were detailed to the Study Research Coordinator for correction.

Results: Of the 64 study data points examined for our 10 patients, 5 data points were missing and excluded from the analysis. Past Medical History was least accurate with 80% agreement. Final status and inclusion criteria were both 100% accurate. Overall, the SDV was 90% accurate across the sample of 10 patients and the 6 data categories. (Table 1)

Conclusion: We have shown that SDV techniques can be applied to observational studies. SDV should be considered in academic studies to assure accurate reporting of critical data. This requires further development before it can be widely utilized.

References: 1. De S: Hybrid approaches to clinical trial monitoring: Practical alternatives to 100% source data verification. Perspectives in clinical research 2011, 2:100-104.
    This eLearning portal is powered by:
    This eLearning portal is powered by MULTIEPORTAL
Anonymous User Privacy Preferences

Strictly Necessary Cookies (Always Active)

MULTILEARNING platforms and tools hereinafter referred as “MLG SOFTWARE” are provided to you as pure educational platforms/services requiring cookies to operate. In the case of the MLG SOFTWARE, cookies are essential for the Platform to function properly for the provision of education. If these cookies are disabled, a large subset of the functionality provided by the Platform will either be unavailable or cease to work as expected. The MLG SOFTWARE do not capture non-essential activities such as menu items and listings you click on or pages viewed.

Performance Cookies

Performance cookies are used to analyse how visitors use a website in order to provide a better user experience.

Save Settings