Medical Registry Data Collection Efficiency: A Crossover Study Comparing Web-Based Electronic Data Capture and a Standard Spreadsheet

Abstract

BackgroundElectronic medical records and electronic data capture (EDC) have changed data collection in clinical and translational research. However, spreadsheet programs, such as Microsoft Excel, are still used as data repository to record and organize patient data for research.ObjectiveThe objective of this study is to assess the efficiency of EDC as against a standard spreadsheet in regards to time to collect data and data accuracy, measured in number of errors after adjudication.MethodsThis was a crossover study comparing the time to collect data in minutes between EDC and a spreadsheet. The EDC tool used was Research Electronic Data Capture (REDCap), whereas the spreadsheet was Microsoft Excel. The data collected was part of a registry of patients who underwent coronary computed tomography angiography in the emergency setting. Two data collectors with the same experience went over the same patients and collected relevant data on a case report form identical to the one used in our Emergency Department (ED) registry. Data collection tool was switched after the patient that represented half the cohort. For this, the patient cohort was exactly 30 days of our ED coronary Computed Tomography Angiography registry and the point of crossover was determined beforehand to be 15 days. We measured the number of patients admitted, and time to collect data. Accuracy was defined as absence of blank fields and errors, and was assessed by comparing data between data collectors and counting every time the data differed. Statistical analysis was made using paired t -test.ResultsThe study included 61 patients (122 observations) and 55 variables. The crossover occurred after the 30th patient. Mean time to collect data using EDC in minutes was 6.2±2.3, whereas using Excel was 8.0±2.0 (P <.001), a difference of 1.8 minutes between both means (22%). The cohort was evenly distributed with 3 admissions in the first half of the crossover and 4 in the second half. We saw 2 (<0.1%) continuous variable typos in the spreadsheet that a single data collector made. There were no blank fields. The data collection tools showed no differences in accuracy of data on comparison.ConclusionsData collection for our registry with an EDC tool was faster than using a spreadsheet, which in turn allowed more efficient follow-up of cases.

Barriers and motivators to participation and retention in HIV/HCV cohort studies among people who inject drugs: a community consultation in Iran

“…using iPads or tablets) and had no objection against the application of such tools for data collection or storage. Over the past decade, using electronic devices to collect data has become very popular in many epidemiological [ 74 , 75 ] and clinical studies [ 76 ]. Applying electronic devices have been shown to improve the quality of data, make staff training easier and more effective, and reduce the time-lag between data collection, data entry and analysis and use of data [ 77 ].…”

Using Digital Technologies in Clinical HIV Research: Real-World Applications and Considerations for Future Work

“…For example, if a participant indicates sexual encounters in the past week, more questions appear to gather additional details about those encounters, but these questions are omitted if the participant answers that there was no sexual activity [ 6 , 22 ]. Forced item response (when participants are required to answer a question before they can move to the next question)—offering a “don’t want to answer” option to avoid coercion—helps reduce or eliminate missing data, substantially contributing to data quality [ 6 , 23 , 24 ]. Advances in technologies for measurement of biomedical variables can help improve integration of biomedical measures with psychosocial measures in a single dataset [ 25 ].…”

Barriers and motivators to participation and retention in HIV/HCV cohort studies among people who inject drugs: A community consultation in Iran

“…using iPads or tablets) and had no objection against the application of such tools for data collection or storage. Over the past decade, using electronic devices to collect data has become very popular in many epidemiological (75,76) and clinical studies (77). Applying electronic devices have been shown to improve the quality of data, make staff training easier and more effective, and reduce the timelag between data collection, data entry and analysis and use of data (78).…”