Contemporary Clinical Trials 28 (2007) 329 – 336

www.elsevier.com/locate/conclintrial

Discussion
Implementation of electronic data capture systems:
Barriers and solutions
James A. Welker ⁎
University of Maryland School of Medicine, Director, Franklin Square Clinical Research Center, Franklin Square Hospital Center,
Baltimore MD 21237, United States
Received 28 April 2006; accepted 4 January 2007

Abstract

Although increasing in pace, the conversion to Electronic Data Capture (EDC) has been a slow progression. The use of EDC
systems should confer improved data integrity, cost savings and a shorter time to study database closure. This will reduce the time
to market and cost of new medications. With the current sentiment of the industry suggesting the cost analysis has been accepted to
be in favor of EDC, the likely limitation to disseminated use is an inability to implement these systems. If the leadership at the
sponsor, clinical research organization and investigator site is cognizant of the barriers to implementation, they can anticipate and
mitigate them prior to the users becoming disgruntled and resistant to the new method of data capture. Once understood, barriers
such as user input, technical support, user motivation, regulatory requirements, communication with users, timing of
implementation, software installation, graphical user interface, identification of bridgers, patient participation, availability of
technology, and costs can be better addressed at the beginning of the implementation process and successfully averted. This review
discusses these barriers and potential solutions that can assist the clinical trial industry in achieving more wide-spread EDC use and
the resulting improvement in operating efficiencies.
© 2007 Elsevier Inc. All rights reserved.

Keywords: Electronic Data Capture; Implementation; Barriers; User input; Technical support; User motivation; Competing projects;
Communication with users; Timing of implementation; Software installation; Graphical user interface; Identification of bridgers

1. Introduction

Quality and financial benefits are increasing the rate at which the clinical trial community incorporates Electronic
Data Capture (EDC). These benefits include direct data entry at the investigator site leading to greater accuracy, fewer
queries, decreased paper record storage, and timelier population of the study database. The study monitor's increased
database access leads to more timely feedback to the investigator site and fewer on-site monitoring visits. The central
storage of data and ubiquitous user access allows the inclusion of intelligence that can remind individual users to
perform required tasks; i.e. remind the investigator site when an enrolled patient is due for a follow-up visit or alert the

⁎ Hospitalist Section, Department of Internal Medicine, Franklin Square Hospital, 9000 Franklin Square Dr., Baltimore, MD 21237, United States.
Tel.: +1 443 777 7706; fax: +1 443 777 7159.
E-mail address: jimwelker@hotmail.com.

1551-7144/$ - see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.cct.2007.01.001
330 J.A. Welker / Contemporary Clinical Trials 28 (2007) 329–336

study monitor when new data has been entered and is ready for review. Additional benefits are generated through the
use of a comprehensive EDC system that incorporates patient randomization, medication dose adjustment and
dispensing, data collection and real-time data monitoring. Electronic prescribing has been found to reduce clinical trial
errors associated with illegible handwriting, inappropriate dosing, and inappropriate medication choice [1].
Direct data capture, the next step in the automation of EDC systems, is a type of electronic data capture in which
the digital data collected by a specific device is directly transferred into the study database thereby eliminating the
need for human entry [2]. Direct data capture optimizes the accuracy and reduces the costs of EDC and electronic
medical record systems [3]. This is seen in the form of the sponsor supplying an instrument such as a computerized
ECG machine or pulmonary function machine and asking the investigator's site to routinely connect the machine to
a phone line and transfer the data to the central database. Less conspicuous is the direct data transfer that takes place
when the central laboratory or Interactive Voice Response System (IVRS) directly interfaces with the study database
[4]. The frontier of direct data capture offers possibly the greatest advancement in EDC systems, the elimination of
the manual entry of data collected locally at the investigator's institution. Due to incompatibility of systems,
currently patient data collected at the site of patient care must be extracted from the hospital database and manually
entered into the EDC system. During the summer of 2006, Siemens (Erlangen, Germany) achieved successful
integration of an EDC system and patient care database for the purpose of research [5]. Though the elimination of
duplicate testing, this offers to streamline patient lab testing and allow the direct transfer of ever increasing amounts
of research data in the future.
The conversion to EDC has been a slow progression so far [6]. Only one-third of clinical trials are performed using
EDC with two of the top five identified reasons for adoption delays being directly associated with the failure to
implement. Of respondents to an international survey, 46% identified inertia or concern with changing current
processes, and 40% identified resistance from investigative sites as the major causes for adoption delays [7]. The
dissemination of new technologies into an established working environment is commonly slow. Factors that delay this
dissemination include the initial expense along with the lack of recognition of eventual cost savings, and the inability to
implement the new system. While the timelier database population and greater accuracy are worthy of an additional
expense, analyst also suggest an actual cost savings due to the decreased time to market for medication, and increased
employee productivity created by decreased data entry time requirements, decreased redundancy in the data recording
and easier access to the database. Regardless of the premise, the current sentiment of the industry suggests the cost
analysis has been accepted to be in favor of EDC. This sentiment is portrayed in a recent article noting the importance
of EDC-specific skill sets and enhanced technical knowledge for Clinical Research Associates to be successful in the
future [8].
This growing use of EDC suggests all parties, the sponsor, clinical research organizations (CRO) and
investigator sites will have to comply with implementation. While EDC systems have technical differences:
application service provider, institutionally hosted, custom developed software, local PC applications using file-
transfer-protocol and direct data capture, and have functional differences: study database, electronic source
documents, inclusion of enrollment, randomization and medication prescribing and dispensing, all of these system
require a successful implementation to offer the expected benefits to the clinical trial. Implementation of new data
systems can be expensive, requires experienced leadership, and frequently fails requiring the institution to
withdrawal from using the selected software. In fact, even at large clinical institutions implementation failure of
technology initiatives is all too common an occurrence. This is due to inexperienced staffs' inability to recognize
the challenges of implementation.
Despite the challenge, institutions small and large need not be resistant to new technologies. If the leadership at
the sponsor, CRO and investigator sites is cognizant of the barriers to implementation, they can anticipate and
mitigate them prior to the users becoming disgruntled and resistant to the new method of data capture. These
barriers include problems with user input, technical support, user motivation, regulatory requirements,
communication with users, timing of implementation, software installation, graphical user interface (GUI),
identification of bridgers, patient participation, availability of technology, and costs. While there exists some
published evidence on which to review these barriers, the techniques of implementing new technologies are neither
well published nor does there exist a generous supply of well designed research from which to make decisions.
Most of the knowledge in this area is acquired from experience in managing implementations with EDC and other
software applications. Herein, we describe these potential barriers and offer advice on how to successfully
overcome them based on experience and the available literature.
 J.A. Welker / Contemporary Clinical Trials 28 (2007) 329–336 331

2. Implementation barriers and solutions

2.1. User input

All too frequently input is only obtained from a small user group that is technology oriented. This has resulted in
implementation failure due to a discrepancy between organization cultural presumptions embedded into a new
information system design and the actual organization's culture, and can be avoided by obtaining appropriate user input
[9]. Prior to selecting a new product a large user group that is process oriented rather than technology oriented should
be surveyed regarding their needs, and suggested solutions. This group should have members from all participating
parties, including the sponsor, the CRO and the investigator site. This ensures worker productivity is addressed and
engages users in the new solution prior to its selection [10]. In addition to the direct attainment of user input, published
data can offer information regarding users' opinions. Data reflecting opinions of patients and investigator sites are
discussed in specific areas of this review.

2.2. Technical support

The Information Technology (IT) department is too frequently a source of delays. Typical of many non-revenue
producing departments, IT departments are frequently understaffed in comparison to the demands for their services and
don't have revenue oriented incentives in place. This is consistent with the absence of revenue for the IT department in
clinical trial contracts, and can make it particularly challenging to receive timely service from the department. While
some suggest the future may see the addition of revenue for the IT department in clinical trial contracts, this
development would require a change in philosophy by both the sponsors who are currently not offering money for these
services, and the IT department which is composed of employees who are not receiving a revenue incentive. While the
future may see these hurdles overcome, presently clinical trial leaders must know other methods of motivating the IT
department.
In order to maximize the quality of service received, clinicians leading clinical informatics initiatives must maintain
a close relationship with their IT department. One should be aware of the IT agenda and effectively communicate your
needs with them, thereby allowing your projects to be properly positioned on their priority list. This can be a
particularly challenging step if the institution is going to host the application. Hosting requires the identification or
setup of the appropriate server and the partitioning of space by your local IT department. Alternatively, when web-
based systems are chosen this server maintenance can be performed by the software company that hosts the EDC
application. The application service provider model can reduce local IT support to focus on basic equipment needs such
as PCs, Internet and phone line access and peripheral equipment such as scanners.

2.3. User motivation

New technologies should be incorporated into a system that increases the institutions overall productivity and
profitability. In order to have the data entry user embrace the product, they need to be educated how this change will
improve the entire system. If their workload is going to increase, they need to be educated as to how they are going to
benefit, or what they are going to receive in exchange for their added efforts. Many affectionately refer to these items as
the WIIMS; an acronym for the plural of “What's In It for Me”. Considering users that are disinterested prior to
implementation remain disinterested during the implementation process, an education program that garners user
motivation is ideally provided prior to the implementation process [11]. Complicating matters further is the prevailing
management style of inspiration and persuasion that is common at investigator sites. As opposed to centers with a
hierarchical management style, healthcare provider institutions commonly lack a central authority figure that can
command a subordinate to begin using a new data collection system. This makes the attainment of user motivation
though education prior to implementation especially important at investigator sites.
The transition of data entry and review to the clinical research coordinators and clinical research associates requires
the CRO and investigator sites spend additional time educating their staff. In order to engage these users, they should be
educated that their increased initial time commitment will result in time savings at a later date resulting from a more
accurate database, fewer queries and an earlier closure of the study database. Methods that effectively demonstrate the
value of the EDC system along with the practical aspects of how the user works with it include lectures during
332 J.A. Welker / Contemporary Clinical Trials 28 (2007) 329–336

investigator meetings, lectures via the Internet, computer based readings and computer based testing that is required
prior to site initiation.

2.4. Regulatory requirements

Although designed to protect patients, Federal Regulations by their nature set challenging expectations of EDC
systems. Regardless of the challenge, regulations created by 21 CFR 11 must be met by the designers of the EDC
system [12]. Typical of governmental guidelines, the statements are broad and require expert interpretation prior to
implementation. In addition to governmental guidelines, private industry groups have formed with the goal of
developing global data standards that enable information system interoperability improving the electronic acquisition,
exchange, submission, and archiving for clinical trial data. One such group, the Clinical Data Interchange Standards
Consortium (CDISC) has created the Study Data Tabulation Model (SDTM). Currently, the FDA plans to adhere to this
SDTM format for its data warehouse to allow it to be more efficient with its reviews and to permit cross-trial analyses
[7]. When choosing an EDC system one should ensure the system is compliant with federal regulations and industry
standards. If those choosing the EDC system are uncertain or they are developing their own EDC system, they should
seek advice from a consulting firm with experience interpreting the guidelines and recommended standards.

2.5. Communicating with users

Although paramount to the success of an implementation, ongoing user communication is often a challenging
obstacle to overcome. Users are busy employees who gravitate towards the most efficient method of accomplishing
their work. They understandably avoid interference of their daily tasks created by on-going conversations regarding a
new method of performing their job. However, this reluctance must be overcome to enable the user to be informed of
the implementation progress and changes to the EDC system. As well, the user must have a convenient feedback
mechanism that allows their concerns to be addressed. While this can become rather time consuming for leadership, an
Internet-based communication system has been shown to result in a cost and personnel effort reduction of 85–89%
[13]. Indeed web-based EDC systems have seen a transition from a simple entry screen, to a more appealing web page
format that includes notices pertaining to software updates and allows the submission of questions by the user.

2.6. Timing of implementation

As with many opportunities, timing can be the difference between success and failure of any technology project. The
goal is to find a time when the staffs are less busy than average and engaged in learning a new project or techniques.
Fortunately, the general start and stop nature of clinical trials offers ample opportunity to implement a new system at the
beginning of a clinical trial. When beginning a new trial the staff is engaged in learning new information and has a
lighter work load than average due to the pending onset of enrollment. This is an excellent time to implement a new
data collection system. Considering the generally short duration of each individual clinical trial, to attempt a change in
data collection systems and an implementation after the project has begun would be rather counter-productive and
inappropriate.

2.7. Software installation

Software installation requires time consuming server maintenance, customization and error correction. Now that the
Internet has become more accessible and reliable, web-based solutions have become the ideal method for deployment
of an EDC system. The use of open source software has made Internet-based applications available and affordable for
both small and large organizations [14]. By allowing multiple users to simultaneously work on the same EDC system
along with the inclusion of tools that enhance site training, study documentation, subject recruitment, randomization,
medication dispensing, and trial management procedures, web-based solutions offer the greatest productivity for all
users and increase data integrity [15]. As well, user acceptance has been demonstrated in a survey of investigator sites
evaluating their experience with a comprehensive, Internet-based EDC system that included patient randomization,
data collection and query resolution [16]. The software installation efficiencies realized by having one EDC system to
setup and maintain makes web-based solutions preferable to the model of local EDC system installation at the
 J.A. Welker / Contemporary Clinical Trials 28 (2007) 329–336 333

investigator sites combined with periodic file transfers using a File Transfer Protocol (FTP) to the master database at the
sponsor. Finally, allowing the software company to act as the application service provider and maintain the EDC on
their own servers eliminates delays in server access and software installation that are common when the purchasing
company desires to host the software themselves.
Whether maintained by the sponsor or an application service provider, adequate time for the IT department to create
server space and install and customize the EDC system's software must be allocated. In addition, individual sites may
be required to install software components to their PCs. When necessary, on-site installation of EDC system
components should be done with applications that are functional on all common operating systems by self-executable
files. This enables individual PC application loading to be done with ease by the end-user over the Internet.

2.8. Graphical user interface

The Graphical User Interface (GUI) of the EDC system is the equivalent of the Case Report Form (CRF) used in the
paper-based model. A well designed GUI should realize reductions in anomalies and queries previously seen with user
friendly improvements in the design of the paper CRF [17]. The GUI has the additional responsibility of enabling users
to perform their tasks in the EDC system. Users want to approach the PC and know how to create new patients and
enter data without spending time in training courses or reading the help section. The chosen software product must be
designed with language and visual cues that make the user interface intuitive.

2.9. Identification of bridgers

Early adopters who become the support person for other members of the research team are known as super-users or
bridgers [18]. A bridger is distinguished from other early adopters by their willingness and ability to educate and
support their co-workers and their steadfast devotion to the successful implementation of the new technology. A single
bridger alone, without other assistance can make implementation into a small workforce a success. This makes their
identification at the investigator site tremendously valuable. At larger sites such as the sponsor or the CRO, time should
be taken to identify numerous bridgers. These bridgers are ideally identified ahead of time, most typically from
experience with other technology projects. Bridgers can be assigned specific tasks during the implementation process,
thereby giving them early visibility and encouraging the users to gravitate to them when they need assistance.

2.10. Patient participation

A truly comprehensive EDC system would include direct data entry for both investigators and patients. The value of
patient data entry continues to increase as Patient Reported Outcomes (PRO) are more commonly collected.
Investigators have compared the correlation of the data submitted via EDC and paper models, and assessed patient
preference. With intraclass correlation coefficients ranging from 0.82–0.96, excellent correlation between the two
methods of data collection has been demonstrated [19,20]. With 88–98% of patients submitting data via each the paper
and the EDC systems when asked to use both, the two methods appear to achieve equal patient compliance [21,22].
However, the EDC system has the added benefit of populating the study database sooner. Finally, patients appear
sensitive to the method of EDC used to collect PRO. Surveys demonstrate an 85% preference for the paper system
versus 4% for EDC, when the telephone/Interactive Voice Response System (IVRS) is used [22]. However, patients
have a 47% preference for hand-held computer data collection tool versus 23% for paper [19], and when PCs are used
the patient preference increases to 69–77% for the EDC versus 9–17% for paper [20]. In addition, a slightly higher
preference for EDC is seen among children than adults. This is likely due to children's equal familiarity with PCs and
paper, and suggests that as PCs continue to become integrated into our daily lives, EDC preference among patients will
continue to grow.
The EDC system could offer each patient their own login whereby they could perform direct entry of diary information
and answer survey questions. Achievement requires patients who have home PCs with high-speed Internet access and
reliably enter data. Unfortunately, these hurdles offer severe limitations outside the control of the research team. Therefore,
patient access to EDC systems are generally not being employed and patients are still completing paper diaries. However,
progress is seen in the form of patient performance of surveys during a clinic visit directly into the EDC system. In this
scenario, the patient is asked to answer a survey directly on the PC while in the research clinic for their research visit.
334 J.A. Welker / Contemporary Clinical Trials 28 (2007) 329–336

Table 1
Cost comparison of EDC versus paper-based model
Clinical trial Actual EDC budget Estimated paper model costs Cost after internalization of EDC
Phase I $59,640 $68,440 $6500
Phase II $319,776 $851,468 $47,952
Phase III $745,986 $3,714,000 $243,000
Phase IIIb $1,643,000 $7,051,125 $693,000
All costs in US dollars.

When the patient is asked to perform direct data entry, the EDC system must ensure that the same information is
maintained in the investigator's source documents [23]. Currently, investigators are commonly maintaining a paper
version of source documents when using an EDC system. This offers the reasonable solution of the EDC system
automatically generating and faxing a report of the patient entered data to the patient's investigator who can add it to the
source documents. Many expect the future to include the maintenance of electronic source documents. In this scenario,
the patient's can enter PRO into the source document database, which can then be electronically transferred to the study
database. Supplying adequate technology to the patient's home is another challenge discussed elsewhere in this review.

2.11. Availability of technology

The implementation of an EDC system requires the ubiquitous availability of technology among its users. Since the
software is typically being provided by the sponsor, the required technology for the investigators and CRO includes
adequate PC availability and Internet accessibility. The current standard is that investigator sites supply themselves
with the required technology at their own expense. Prior to the last few years, sponsors were not collecting information
regarding the presence of the required equipment prior to site selection. This resulted in situations where the study
monitors were required to allow the study site to perform data entry during the monitoring visit in order for it to be
entered into the EDC system. These occurrences were terribly inefficient and those sites without adequate hardware
were removed from the trial. Subsequently, sponsors began inquiring as to the presence of PCs and Internet access
during pre-study site selection questionnaires and excluding those without adequate hardware.
In addition to PC availability, Internet connectivity is equally important. As we are moving toward Internet-based
systems, high-speed Internet access becomes a requirement [24]. Dial-up Internet access does not offer adequate speed
to access these Internet-based applications. The dissemination of broadband has made high-speed Internet access easily
affordable and available is the USA, Europe and Japan. However, clinical trials are increasingly infiltrating countries
with fewer economic resources. Once again due to it affordability, broadband is becoming available in the poorest of
countries. Countries such as Bangladesh have been noted to disseminate wireless broadband such that their urban areas
have ubiquitous Internet access [25]. Companies such as Green Wifi (San Francisco, CA) are addressing one of the
largest barriers to Internet access: the lack of reliable electricity in developing areas required to power the broadband
network. Green WiFi has developed a low cost, solar-powered, standardized WiFi access solution that runs out-of-the-
box with no systems integration or power requirements. All that is required is a single source of broadband access.
Green WiFi nodes can then be deployed on rooftops to form a self-healing network that hops the source signal over a
virtual 802.11 b/g grid. Because these nodes require no fixed installation or power tie-ins, these nodes can form an
unplanned, mobile grid that can grow or be relocated as needed. In combination with inexpensive hand-crank powered
PCs, the future promises the ability to have an EDC system functional in any country.

2.12. Costs

Cost is a barrier to all aspects of research. While research is necessary to develop products that ultimately generate
revenue, research itself doesn't generate revenue for the sponsor. In this setting, the organization focuses on minimizing
costs. Therefore, advocates of EDC must demonstrate how the EDC system reduces overall research costs.
Substantial cost reductions have been demonstrated when compared to traditional paper-based data capture
methods. These reductions have been shown in all phases of research, and increases with the trial size and associated
expenditures. In addition, large organizations with IT departments can realize further cost reductions by internalizing
the software and performing their own CRF and report customization. Cost data from four different clinical trials have
 J.A. Welker / Contemporary Clinical Trials 28 (2007) 329–336 335

Table 2
Impact of built-in error checking on data quality for 10 Phase III trials
EDC Paper
Invalid subjects enrolled 7.50% 15%
Cost of raising and resolving a query $10 $60
Number of queries/subject 0.25–1 5–20
Data requiring correction 0.05–0.1% 1–2%
Queries caused by missing data 0% 48%
Queries caused by inconsistent data 5% 35%
Queries caused by out-of-range data 0.10% 8%
Queries requesting clarification 0% 6%
Queries due to invalid data 0.05% 0.10%

been analyzed, including phases I, II, III, IIIb and ranging in size from one site for the phase I trial to 192 sites for the
phase IIIb trial. The data collected for this analysis includes actual approved EDC budgets, estimated costs for a paper
model implementation from four panel experts, and actual EDC costs applied when the a client internalizes the
software. The result is the potential decrease in cost from $7 million to $700,000 for a phase IIIb study at a large
organization. (Table 1) [26] Despite the direct costs of the EDC system, the reduction in overall project costs are the
result of increased employee productivity created by less time required for data entry and less redundancy in the data
recording. The EDC has built-in error checking that results in fewer invalid patients enrolled in the trial and fewer
queries. This was demonstrated in data from 10 phase III trials involving 6700 patients, half of which used EDC and the
others had paper systems. In addition, easier access to the database lowers the cost of query resolution from $60 for the
paper system to $10 for the EDC system.(Table 2) [26] The combination of electronic tools and reports along with
manual ad-hoc query by study monitors has been found to enhance the ability to clean, analyze and report more
accurate data [27]. Finally, a reduction in the clinical trial duration of 30% and the time to database closure of 43% is
seen with EDC compared to paper data collection models [26]. These improvements in the efficiency by which clinical
trials are performed offer a shorter time to market and earlier revenue return for the investigational products.

3. Discussion

In order to overcome these barriers, support must be generated from each organization: the sponsor, CRO and
investigator site. All of these groups need support from each other and possibly outside groups to overcome these
challenges. While the sponsors and CROs tend to be large organizations and likely have well versed implementation
managers, the investigator sites tend not to be so well endowed. In particular, the investigator sites may not have
adequate access to experienced implementation managers to be successful. The necessary support could come from the
sponsor or CRO through the extension of their own implementation team to the investigator sites.
Alternatively, all of these groups may benefit from external implementation support. Software companies are
cognizant that the initial sale of a software package is only the beginning of what is hopefully a long-term relationship
which has yearly growth in sales. This sales growth is the result of the purchase of additional user licenses and other
supportive services that occur when the software is successfully implemented in the customer's organization.
Recognizing that software cannot simply be sold to the user, in an effort to ensure its success large software companies
offer comprehensive, on-site implementation support. Small software companies are not adequately staffed to offer
onsite implementation support, particularly on the national or global scale. These small companies may successfully
deploy their software to small user group in the presence of a strong super-user. However, when attempting software
implementation to a large company, even the best designed EDC application will likely fail without a good
implementation team guiding the process. Considering many of our best software products are created by small
companies, the clinical trial industry would benefit from the ability to implement this software by obtaining support
elsewhere. In this setting, organizations are well served to hire a local technology firm to guide the implementation. The
local technology company investigates the software functionality and institution's current processes, then creates and
executes an implementation plan that ensures the EDC system is successful at the organization.
The use of an EDC should confer improved data integrity, cost savings and a shorter time to study database closure.
This will reduce the time to market and cost of new medications. Dissemination of EDC usage will ensure these
336 J.A. Welker / Contemporary Clinical Trials 28 (2007) 329–336

benefits are realized and can be achieved through an increased understanding of implementation barriers. Once
understood, barriers such as user input, technical support, user motivation, regulatory requirements, communication
with users, timing of implementation, software installation, graphical user interface, identification of bridgers, patient
participation, availability of technology, and costs can be better addressed at the beginning of the implementation
process and successfully averted.

References

[1] Cooper-Dehoff R, Handberg E, Heissenberg C, Johnson K. Electronic prescribing via the Internet for a coronary artery disease and hypertension
megatrial. Clin Cardiol 2001;24(11 suppl):V14-6.
[2] Hyde AW. The changing face of electronic data capture — from remote data entry to direct data capture. Drug Inf J 1998;32:1089–92.
[3] Turner D. Digital data capture for electronic patient record systems. J AHIMA 1995;66(3):63–5.
[4] Byrom B. Using IVRS in clinical trial management. Appl Clin Trials 2002;11(10):36–42.
[5] Siemens helps streamline clinical trial process: pilot study demonstrates new, scalable method of automatically transferring data from EMR to
EDC system, reducing lag time and increasing efficiency. Siemens News Release. Sept. 19, 2006.
[6] Marks RG. Validating electronic source data in clinical trials. Control Clin Trials 2004;25:437–46.
[7] Kush RD. The future for electronic data capture. Drug Dev 2006:46–7.
[8] Brandon S. Positioning yourself in a changing environment: the new world of clinical trials. The Monitor 2006;20(2):71–2.
[9] Pliskin N, et al. Presumed versus actual organizational culture: managerial implication for implementation of information systems.
Comput J 1993;36(2):143–52.
[10] Berg M. Implementing information systems in health care organizations: myths and challenges. Int J Med Inf 2001;64:143–56.
[11] van der Meijden MJ, Tange H, Troost J, Hasman A. Development and implementation of an EPR: how to encourage the user. Int J Med Inform
2001;64:173–85.
[12] 21 CFR Part 11 Electronic Records; Electronic Signatures. http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm. (accessed
October 24, 2006).
[13] Fortin AH, Luzzi K, Galaty L, Wong JG, Huot SJ. Developing an Internet-based communication system for residency training programs.
J Gen Intern Med 2002;17(4):278–82.
[14] Scognamillo R, Strozzi C, Vincenzi B, Recchia G. Clinical trial management and remote data entry on the Internet. Drug Inf J 1999;33:1061–5.
[15] Marks R, Bristol H, Conlon M, Pepine CJ. Enhancing clinical trials on the Internet: lessons from INVEST. Clin Cardiol 2001;24(11 suppl):
V17–23.
[16] Lallas CD, et al. Internet based multi-institutional clinical research: a convenient and secure option. J Urol 2004;171(5):1880–5.
[17] Mastro D. Clinical trial data collection — giving case report forms a facelift without radical surgery. Drug Dev 2006:48–52.
[18] Ash JS, Stavri PZ, Dykstra R, Fournier L. Implementing computerized physician order entry: the importance of special people. Int J Med Inform
2003;69:235–50.
[19] Bushnell DM, et al. Validation of electronic data capture of the Irritable Bowel syndrome — quality of life measure, the work productivity and
activity impairment questionnaire for Irritable Bowel syndrome and the EuroQol. Value Health 2006;9(2):98–105.
[20] Bushnell DM, Martin ML, Parasuraman B. Electronic versus paper questionnaires: a further comparison in persons with asthma. J Asthma
2003;40(7):751–62.
[21] Nyholm D, Kowalski J, Aquilonius SM. Wireless real-time electronic data capture for self-assessment of motor function and quality of life in
Parkinson's disease. Mov Disord 2004;19(4):446–51.
[22] Weiler K, Christ AM, Woodworth GG, Weiler RL, Weiler JM. Quality of patient-reported outcome data captured using paper and interactive
voice response diaries in an allergic rhinitis study: is electronic data capture really better? Ann Allergy Asthma Immun 2004;92(3):335–9.
[23] Guidance for Industry: Patient-Reported Outcome Measures: Use in Medical Product Development to Support Labeling Claims. US FDA Draft
5460, February 2006.
[24] Lopez-Carrero C, et al. Internet in clinical research based on a pilot experience. Contemp Clin Trials 2005;26(2):234–43.
[25] Chen N. Wireless mesh networking for developing countries. http://www.convergedigest.com/bp-bbw/bp1.asp?ID=372&ctgy=2006 (accessed
October 21, 2006).
[26] Bart T. Comparison of electronic data capture with paper data collection — is there really an advantage? Business Briefing Pharmatech
2003:1–4.
[27] Trocky NM, Fontinha M. Quality management tools: facilitating clinical research data integrity by utilizing specialized reports with electronic
case report forms. AMIA Annu Symp Proc 2005:1142.