Use of the data system for field management of a clinical study conducted in Kolkata, India

Running a clinical study smoothly like ours is challenging due to its complex and long duration visit schedule. In our study, all 372 participants were successfully recruited according to the planned schedule. The drop-out rates were only 8 % and protocol deviations stood at 3 %. The reasons for these protocol deviations included: visits occurring outside the window period (44 %), missed specimen (34 %), missed visits (11 %), specimen collected outside the window period (10 %), and missed procedures (1 %). 99 % blood samples, 95 % stool samples, 98 % urine samples, and 100 % breast milk samples were successfully collected. Such an excellent performance would have not been possible without having the support provided by the data system.

In addition to manage data activities and generate data reports, our data system provides support for management of the field activities such as preparing visit schedule, alerting next day visit schedule and assessing requirement of the logistics including number of doses of the vaccines, etc. for the next visit schedule. It also ensures a good clinical practice, a legal requirement to conduct clinical trials in many countries. The other advantage in our system is that it generates nutritional status report of the participants at each time point so as to take immediate actions against severely malnourished infants, such as referring to a specialized clinic where proper treatment can be ensured. In the system, the data interoperabilities are limited to Acrobat Reader and Excel, as required by us. However, we have kept provision for the data interoperability in any other standard data format.

Our data system is specific to our study, which is unlikely to match exactly with any other studies. Three-month time of a skilled system designer and programmer was required to design and develop the system. Unlike open source software where many developers have the opportunity to scrutinize the system, our system was scrutinized in-house.

Since many developers scrutinized the open source application, it is hard for bugs to hide in that application. Also, since codes of the open source software are shared among numerous parties, it is typically well structured, which cannot be ensured a single company develops like us [10].

However, our system passes through different courses of action over the time, and we were able to fix the bugs in the system. More importantly, we were able to tailor it specific to the project activities, which may not have been possible in an open source software solution.

The limitation in our system is that it was installed on standalone computers. Therefore, we had to upload of the updated database to the field office computer on a regular basis. Similarly, the data generated at the lab were uploaded manually into the system. This offline mechanism could have been avoided if the computers at the clinic and lab were connected to the data office through a network. However, we could not make it possible in our setting due to limitation of resources.

Still a large part of clinical centers uses their own developed solution or a single solution [11] because clinical data managements are very heterogeneous, and the open source solutions do not play a major role in clinical trial data management [12]. However, because of flexibility, increased innovation, shorter development times, and faster procurement processes, open source software may be attracted by an organization. Also by using open source software, an organization will not be locked into using a proprietary software program. One disadvantage in an open source solution is that, due to the web-based nature of this system, it may pose a challenge for real time data entry because in many developing country settings internet connectivity is a problem. One may also face sluggish response times and system timeouts. Additionally, XML rules may adversely affect the application’s response time and therefore other options may need to be explored and used while using open software solution [3].