Programme no. 138-OP
SESAMe: Opening doors to improved digital data collection, enabling large randomised controlled trials in general practice. An inductive software development process.
Trygve Skonnord*1, Arne Fetveit2, Mette Brekke3, Finn Steen4, Holgeir Skjeie5, Atle Klovning6
1Department of General Practice,University of Oslo,Oslo,Norway;Centre for Medical Web Research (cmwr.org),Oslo,Norway, 2Department of General Practice,University of Oslo,Oslo,Norway, 3Department of General Practice,University of Oslo,Oslo,Norway, 4Centre for Medical Web Research (cmwr.org),Hong Kong,China, 5Department of General Practice,University of Oslo,Oslo,Norway, 6Department of General Practice,University of Oslo,Oslo,Norway;Centre for Medical Web Research (cmwr.org),Oslo,Norway
* = Presenting author
Objectives: We aimed to develop software that would automate the sending of emails with links to electronic surveys, and thereby improve data collection.
A common problem in clinical research in general practice is small sample sizes. Logistics limit the feasibility of conducting large randomised controlled trials (RCTs). Interventions take place in small and busy practices, and researchers often need to organise the study themselves. Funding is often not sufficient to employ research assistants who can do phone interviews or send out reminders.
Electronic questionnaires greatly ease data collection. When planning our RCT – the Acuback trial, we struggled to find existing software that could help automate email distribution. The trial plans to consecutively include 270 patients with acute low back pain, and collect data by electronic questionnaires at 19 defined time-points from before treatment to 1 year after. Consequently, we need to send out separate emails for all questionnaires, totalling 5130 emails, a process that obviously needs to be automated.
Results: We developed an online software application that sends automated emails with links to electronic surveys. We also developed a study monitoring function, detecting missing responses. The software now can send reminders by either email or SMS, thus reducing the incidence of missing data. SESAMe is an acronym for “Survey Email Scheduling And Monitoring in eRCTs”.
The first version of the software was tested in a pilot study in November 2013. It had limitations, and was unable to deliver multilevel and online access to the software. This was solved in the next version of the program, which now schedules and sends the required emails in the main study that started in March 2014.
The system architecture evolved through an inductive process between the project study leader and the software programmer.
Conclusion: The SESAMe software application eases the study logistics by automating and monitoring data collection. This hopefully opens doors to conducting larger RCTs with sufficient sample sizes, enabling researchers to conduct higher quality clinical trials in general practice.
Points for discussion:
Pitfalls in data collection for RCTs.
How to improve the quality of conducting clinical trials in general practice.