OR WAIT null SECS
© 2023 MJH Life Sciences™ and Applied Clinical Trials Online. All rights reserved.
Applied Clinical Trials
Practical guidance for effectively integrating PDAs and other handheld devices into CTs
Recent estimates of the extent of adoption of electronic data capture (EDC) systems in clinical trials vary. Published adoption rates range from 24% of all Phase I to Phase IV trials using EDC,1 with 15% of all newly initiated trials using some form of EDC,2 20% of Phase II and Phase III studies using EDC,3 and no more than 20% of all current clinical trials using EDC.4 Therefore, one can say that around 20% of clinical trials are currently using some form of EDC. What is also clear is that adoption is increasing.1,5
The potential benefits of EDC are plenty: rapid access to trial data, more efficient monitoring, higher quality data due to real-time edit checks and validation logic, elimination of the need for double data entry, reduced queries, less paper, instantaneous alerts, and faster time to data lock.6,7 Evidence indicates that EDC can reduce overall trial costs, improve patient recruitment, and improve data quality.7-9
Current practice with EDC technology is to have a user working on a fixed personal computer for data entry and reporting. This is most likely through a Web interface. However, this approach can interrupt the workflow of a trial. For example, a fixed computer may not be in the same room where the patient visits will take place. This results in data being collected on paper CRFs and then entered into the EDC at a later time.
Furthermore, not all visits are at a health care facility. For instance, interviews may be conducted at a patient's residence or place of work, or a coordinator may have to travel to multiple sites to collect data. In such scenarios, data is collected on paper and entered into the EDC back at the office. In fact, in many cases the advent of EDC has not eliminated paper in clinical trials.
Adding mobile technology to an EDC system can help reduce the reliance on paper and streamline the workflow even further. Some types of mobile devices are already in common use in health care settings. A recent systematic review10 estimates that the adoption rate of personal digital assistants (PDAs) in health care settings ranges from 45% to 85%, with adoption currently at its steepest point of increase. Therefore, many health care professionals are already accustomed to using PDAs personally and professionally.
Figure 1. In a typical configuration there is a central database with Web as well as mobile access. Each of the devices may be used at a different site.
A systematic review of randomized controlled trials comparing handheld (mobile) devices to pencil and paper methods for data capture concluded that handheld devices improve the timeliness of data handling and are preferred by the subjects, which could result in improved adherence to protocols.11 Improved data quality was also reported under certain conditions. Therefore, there is evidence that the general benefits of EDC extend to mobile devices as well.
The following is an explanation of what mobile technology is, how it can be used in clinical trials, and presents some of the practical issues that need to be considered when deploying it.
Mobile technology means two things: 1) a mobile device is used to enter and view data, and 2) the device can operate when disconnected from the network (e.g., a user can still do data entry and reporting). All mobile devices tend to be relatively small and can be carried around easily. Examples of mobile devices are PDAs (such as Palm and Pocket PC devices), smart phones, Tablet PCs, and laptops.
Mobile devices can be used during patient visits to collect data directly onto the device, and they can be carried around when visiting multiple sites or locations. This can, in many instances, eliminate the need for having to complete paper CRFs and then enter them into an EDC later.
Operation in a disconnected mode is sometimes necessary because network connectivity is not guaranteed in many institutions. It is surprising how often networks will go down, either because of internal issues within the sites themselves or due to denial-of-service attacks on particular segments of the Internet. In these malicious attacks, a large number of requests are sent to a single computer address and the computer becomes overwhelmed or simply crashes.
The ability to operate in a disconnected mode insulates the clinical trial from disruptions caused by network outages. Also, some institutions have slow connections. For example, in one North American organization a new image archival system was installed without adequate capacity increases for the internal network. Soon after deployment, the whole internal network became painfully slow because the same network lines were being used to transfer very large image files. This also slowed down all eCRF completion and submission at the affected sites. During the "slow-down" period, use of a Web-based EDC was very frustrating.
A standard configuration for the use of mobile devices in a clinical trial is shown in Figure 1. All mobile devices used for data capture in clinical trials will need to have a local database on it to hold the data. At specific points, the user will have to synchronize the data on the device with a central database. Data synchronization is a two-way process: copies of all the data are sent to the central database through a secure communication channel and updates to the databases on the mobile device are performed as well.
Data synchronization may not remove all of the data from the device since it is not used only for data entry; the device must be capable of holding data for updating, reporting, and reviewing by users. For example, the concomitant medication forms for a subject are constantly being updated and therefore need to remain on the mobile device for the duration of the study.
There is typically a central database with multiple devices communicating concurrently with the same central database. Typical basic infrastructures can handle thousands of communicating devices, and that can usually be expanded by adding more machines.
It is important to have Web as well as mobile access. Similar to a Web-based EDC, investigators must be held accountable for the data that is entered on mobile devices. It may be impractical for them to review and sign off on all such data on a mobile device, depending on its screen size. Therefore, a Web-based capability also allows investigators to review and sign off on data that has been entered on mobile devices.
When an EDC system is migrated to a mobile device, many of the same functions that were available on the Web are still available through the mobile device. Below is a summary of the main capabilities that you can expect to see.
It is always possible to create new patients, enable new visits, and enter data in eCRFs on a mobile device. While the rendering of the forms and their layout will look different on a mobile device, the basic functionality, navigation, and embedded validation logic is transferable to the device. This also applies to "continuous" CRFs such as concomitant medications, which need to be updated across multiple visits.
Embedding annotations and comments within a mobile eCRF is easy to do. Some solutions even allow the inclusion of images. The video capability on newer PDAs, for example, provides a remarkable improvement in the quality of displayed images.
It is not always practical to keep a patient "master list" on a mobile device. For example, if a site has hundreds of subjects it may be awkward to browse or search for the right patient before each visit. The mobile device should allow the selective removal of patient data so that only data on the most relevant set of patients (e.g., those scheduled for a visit this week) are on the device.
It is also important that a mobile device allows monitors to create new queries and view and update query status. This will allow monitors to perform most of their tasks on the mobile device during site visits.
In disconnected mode, reports will not be in real-time (i.e., they will not necessarily reflect the data in the central database) but will reflect the snapshot from the last synchronization with the central database. However, basic metrics on the progress of a trial (e.g., recruitment actual vs. target, query status, and adverse event summary stratified by site, age or gender) as well as trends should be available on the mobile device. The mobile devices have a local database, therefore they can store the metrics or calculate them on demand based on the local data.
Additional supporting functionality that is integrated with the EDC should also be available on the mobile device. This includes contact lists for the sites, reference materials, copies of protocols, newsletters, and SOPs.
1. D. Borfitz, "Conspiring Forces Behind EDC Adoption," CenterWatch, 10 (2) (2003).
2. "New Drivers of eClinical Technology Adoption," Tufts Center for the Study of Drug Development, Tufts University (2005).
3. "The Waife & Associates EDC Report," Waife & Associates (2005).
4. L. Ramos, "The Promise of Next-gen eClinical Trial Software: Market Overview," Forrester (2006).
5. J. Paul, R. Seib, T. Prescott, "The Internet and Clinical Trials: Background, Online Resources, Examples and Issues," Journal of Medical Internet Research, 7 (1) p. e5 (2005).
6. J. Mitchel et al., "Internet-based Clinical Trials: Practical Considerations," Pharmaceutical Development and Regulation, 1 (1) 29–39 (2003).
7. J. Mitchel et al., "Paper vs. Web: A Tale of Three Trials," Applied Clinical Trials, 2001, 34–35.
8. J. Mitchel et al., "Clinical Trial Data Integrity," Applied Clinical Trials, May 2003.
9. T. Bart, "Comparison of Electronic Data Capture with Paper Data Collection—Is There Really An Advantage?" Business Briefing, Pharmatech, 2003, 1–4.
10. C. Garritty and K. El Emam, "Who's Using PDAs? Estimates of PDA Use by Health Care Providers: A Systematic Review of Surveys," Journal of Medical Internet Research, 8 (2) p. e7 (2006).
11. S. Lane et al., "A Review of Randomized Controlled Trials Comparing the Effectiveness of Handheld Computers with Paper Methods for Data Collection," BMC Medical Informatics and Decision Making, 6 (23) (2006).
Khaled El Emam, PhD, is chief scientist with TrialStat Corporation, 44 ByWard Market, Suite 270, Ottawa, ON, Canada K1N 7A2, email: firstname.lastname@example.org as well as an associate professor and Canada Research Chair at the University of Ottawa.