Government efforts to protect the rights and safety of participants in clinical trials are focusing more attention on the activities and roles of institutional review boards. Increased funding of research by the National Institutes of Health and by the pharmaceutical industry continues to expand IRB work loads. New government policies are adding to these assignments. Such developments raise concerns about the ability of these panels to provide timely and informed review of research programs.

The Department of Health and Human Services (HHS) is responding with more guidance to improve institutional review board (IRB) operations. The Food and Drug Administration has proposed new policies to help IRBs do their jobs. The HHS Office of the Inspector General (OIG) is keeping its eye on IRB policies following its major 1998 report that was highly critical of IRB performance. The OIG acknowledges that HHS and the research community have implemented many of its recommendations for improving the medical research process, but notes that more remains to be done.


IRBs have become such a key element for ensuring research integrity that policy makers are wrestling over the need for more intense regulatory oversight of these entities. Senator Edward Kennedy (Democrat, Massachusetts), who chairs the Senate Health Committee, is considering legislation that would mandate accreditation of all IRBs. Last spring, Kennedy circulated a proposal requiring IRBs accredited by the HHS Office of Human Research Protection (OHRP) to approve all human subject research regulated or funded by the federal government. Such a mandate would involve establishing more detailed national standards for human subject research.

HHS officials, including OHRP director Greg Koski, prefer a voluntary accreditation program for IRBs and research institutions. Similarly, a legislative proposal in the U.S. House of Representatives advocates voluntary accreditation by private organizations, an effort now getting under way. To build support for the voluntary approach, OHRP is working with FDA, accrediting organizations, and the research community to improve and expand OHRPs quality improvement program for all institutions involved in human research protection.

One OHRP initiative is to help IRBs improve operations. OHRP issued a guidance in July advising IRBs how to develop written procedures to better inform outside observers about the way they handle research plan reviews. HHS requires IRBs to establish written procedures for their oversight activities, including how to hold board meetings, assemble materials needed to conduct a review, conduct reviews in emergency situations, and maintain records and meeting minutes.

Another OHRP guidance describes appropriate IRB procedures for carrying out continuing review of long-term clinical research programs. The guidance defines what constitutes substantive and meaningful continuing review, how to set a continuing review schedule, continuing review of multicenter trials, expedited continuing review procedures, and dealing with lapses in continuing review.


FDA also is encouraging accreditation of IRBs that oversee clinical trials regulated by the agency, a move that reflects the steadily expanding role of IRBs in ensuring research integrity. A new FDA guidance indicates that failure of an investigator to obtain IRB approval of significant protocol changes is one reason for the agency to put a clinical trial on hold. An August 2002 Draft Guidance for Industry and Clinical Investigators on the Use of Clinical Holds Following Clinical Investigator Misconduct (www.fda.gov/OHRMS/DOCKETS/98f/ 082702f.htm) says that if FDA finds that an investigators actions pose a significant risk to study subjects and should be halted immediately, it will go the clinical hold route. FDA emphasizes that it wont close down a study merely for record-keeping violations or failure of an investigator to disclose certain types of financial interests. But agency officials say that shutting down a trial is warranted in serious cases because it usually takes months or years for the agency to bring charges to disqualify an unethical investigator from involvement in FDA-regulated studies (see sidebar on Grounds for Disqualification).

Those significant violations most likely to prompt FDA to suspend a trial include: 

	failure to report serious or life-threatening adverse events.

	serious protocol violations, such as enrolling ineligible subjects in a study that could put them at risk.

	repeat or deliberate failure to obtain adequate informed consent, including falsification of consent forms or failure to disclose serious risks of the investigational drug.

	failure to obtain IRB review and approval for significant protocol changes.

FDA says it will meet with a sponsor to discuss such problems before imposing a clinical hold, unless the agency finds that subjects are being exposed to immediate and serious risk. The agency also indicates that it may lift a clinical hold if a sponsor replaces the violative investigator or hires a monitor to oversee the trial. FDA is accepting comments on the clinical hold proposal until the end of November.


Another FDA proposal requires sponsors and clinical investigators to inform IRBs of any prior reviews of a protocol by another IRB. The aim of this draft proposal, issued March 2002, is to discourage IRB shopping by sponsors who run into opposition to a research plan from one review board and then look for another panel likely to regard their plan more favorably. The FDA proposal responds to the OIG 1998 report that cited IRB shopping as a problem for adequately protecting clinical trial participants.

In June, PhRMA (Pharmaceutical Research and Manufacturers of America) filed comments objecting to the requirement as likely to delay IRB decision making, particularly for trials at multiple sites or in foreign countries. Sponsors fear that the requirement will add to the heavy volume of information already overwhelming IRBs and would require researchers to track the status of different site studies more closely. Some sponsors and research institutions indicated, though, that mandatory disclosure of important safety issues raised by another IRB might be justified.


 In addition to these and other new requirements, IRBs and privacy boards will play a central role in a vast new program for ensuring that biomedical researchers take steps to protect the privacy of participants in clinical and health care studies. HHS issued its final Standards for Privacy of Individually Identifiable Health Information on 14 August 2002, establishing a federal policy to protect the privacy of individual health information, as required by the Health Insurance Portability and Accountability Act of 1996 (HIPAA). The privacy rule regulates how health plans and providers may use and disclose identifiable health information to treat patients, pay for medical services, conduct health research, and other purposes.

Researchers repeatedly have raised concerns about the privacy rule because its requirements are likely to make it difficult for them to access important patient health information (see ACT View from Washington, June 2002). The final rule addresses some of these objections, but not all. One important revision is to permit more patient identifiers, including zip code and birthdate, in the prescribed limited data set that would be available to researchers, health care providers, and public health officials. Researchers must agree to keep the information secure and use it only for permitted purposes, a requirement that many fear will be difficult to carry out and may delay or kill certain research projects.

The final rule also streamlines procedures for researchers to obtain individual consent to access personal medical information. However, considerable uncertainty remains about what criteria are needed to obtain a waiver from an IRB to use patient data for research without individual authorization. HHS says it will issue a guidance on waiver criteria to address this difficulty, including how an IRB should determine that the research use of the information poses no more than a minimal risk to an individuals privacy. A guidance will further define what steps a research entity should take to ensure privacy protection, how long a researcher may retain individual data, and whether restrictions on data retention would compromise data reanalysis and other later studies.

The highly complex rule poses even greater difficulties for health plans and providers. Many health care organizations continue to press HHS for further delay in full implementation of the policy, but it now is scheduled to become effective for most entities April 2003.

SIDEBAR: Clinical Trial Control and Conduct


In June, PhRMA approved a set of principles for the conduct of clinical trials in June that calls for IRBs or ethics committees to review all clinical study proposals before the studies begin. The code also recommends that sponsors conduct all clinical trials in accordance with applicable laws and regulations and follow good clinical practice (GCP). While these policies raise little controversy, the code also addresses more tricky issues involving control and publication of research results. Here is a summary of some of the more interesting statements:

	Study results involving marketed products should be reported in an objective and timely manner. However, studies that are exploratory in nature or involve a product that is not marketed may be highly proprietary. Sponsors do not commit to publish the results of every exploratory study or to make clinical trial protocol designs publicly available at inception. 

	Sponsors are the owners of a study database and are responsible for receipt and verification of data from all study sites. Sponsors thus have discretion to determine who will have access to the database. Individual investigators in multisite trials will have access to data on their own research participants as well as a summary of study results from the sponsor. Investigators also should be able to review relevant statistical tables and to examine reports on the entire study at the sponsors facilities or other agreeable location. 

	Only individuals who are involved in the conception or design of a study, in data acquisition and interpretation, and in writing the manuscript should be recognized as authors or contributors to a manuscript. All authors should be given the relevant statistical tables, figures, and reports needed to support the planned publication. Companies may have staff help analyze and interpret data and produce manuscripts and presentations, and their contributions should be recognized appropriately.

	Sponsors have the right to review any manuscripts, presentations, or abstracts originating from their studies prior to publication. Sponsors will review such material in a timely manner and will not suppress or veto publication, although publication delay may be necessary in some cases to protect intellectual property. 

	While clinical investigators should be independent, it is reasonable to compensate researchers for their work. However, such compensation should not be stock in the sponsor company, and payments should not be tied to the outcome of a trial. 
JW


FDA may take enforcement action to disqualify an investigator from conducting future studies regulated by the agency if the individual commits offenses likely to put a study on clinical hold, according to its draft guidance on clinical holds issued in August. This process begins when an FDA Center issues a Notice of Initiation of Disqualification Proceedings and Opportunity to Explain (NIDPOE) letter. In addition to taking enforcement action for the above violations, FDA may send such a letter to an investigator who repeatedly and deliberately:

	fails to obtain or document informed consent, which may involve inadequate description of serious risks associated with an experimental therapy or failing to provide informed consent in language understandable to the subject.

	administers the investigational product to ineligible individuals. 

	fails to comply with conditions placed on the study by the IRB, sponsor, or FDA or fails to follow the signed investigator statement or protocol.

	fails to maintain accurate study records or submit required adverse event reports to the sponsor.

	falsifies or conceals study records, possibly by substituting erroneous biological samples or by fabricating subjects. 
 JW            

Articles

View from Washington

Federal agencies seek to enhance IRB operations as new policies further expand board responsibilities.

IRBs in the Spotlight

Since the bursting of the dotcom bubble, recreational technology users have had a lot less fun. Our favorite Web sites have disappeared, news magazines are no longer filled with technology toys that we want to buy, and our attention is focused on technologys practical business uses, which demonstrate return on investment. This column will change all that and more. Lets examine a relatively new bit of personal technology for your home that is relatively inexpensive, practical, time-saving, fun to install and use (at least for techies like me!), and very important for protecting the security of your datathe home network. Let me demystify home networks and convince you of their value and their ease of use. Along the way, I hope you learn a bit about networks, firewalls, security, virtual private networks, and configuring a clinical site for optimal data security. 


Before we get to home networks, we have to discuss a bit about broadband Internet connections. Broadband Internet connectivity is rapidly replacing dial-up connections in many homes. Internet connections through broadband most commonly include standard cable lines (the same ones from which you get your cable TV) and cable modems or digital subscriber line (DSL) connections through specialized telephone lines. The data transmission speeds for downloads in a broadband connection typically range from 128 kilobits per second (kbps) up to, theoretically, 6100 kbps for DSL and 384 kbps to 2000 kbps for cable modems, compared with the typical 33 kbps to 52 kbps for a dial-up modem. Information moves 10 to 100 times faster with a broadband connection compared with a dial-up. For those who havent experienced it, such a connection is breathtakingly fast. A 10 MB file download that takes an hour on a dial-up connection can happen in less than a minute, and Web site access is more or less instantaneous.

Internet connections through broadband bring many advantages. Your home Internet connection will be as fast, or faster, than your work Internet connection. You can use a virtual private network (VPN) to access your work computers at the same speed that you do at work, giving you complete access to your work computer network from home. This makes it possible to telecommute without any data disadvantage. Broadband connections are usually $39 to $75 per month and currently are available in a very large percentage of the area of the United States. Balance this against the cost of an Internet service provider (ISP; from $10 to $20 typically) and a second phone line, both of which are unnecessary with broadband connectivity. Furthermore, with broadband Internet access your computer can be always onno log-ins or dial-ups are needed, so you can view a Web page at a moments notice, or check your email regularly. With instant Web access, family members who couldnt be bothered dialing in will soon be shopping, looking up phone numbers, or paying bills.


Now that you are sold on broadband connectivity and have your computer online, you need to know the bad news. You are now part of a large wide area network (WAN) and completely vulnerable to having all the data on your computerpersonal or work-relatedexamined, copied, erased, altered, or otherwise vandalized. Many people are shocked to learn this. They blissfully called their cable company, added broadband connectivity, and went about their business. As did I, until I heard about my friend David (the name has been changed to protect the innocent).

 David is a home-based consultant who had a broadband Internet connection to his home computer through his cable company. One evening, his neighbor down the street called him and began relaying his knowledge of personal details concerning finances. His neighbor was also a cable Internet subscriber and had clicked on his Network Neighborhood icon and discovered Davids computer right there, next to his. So, the neighbor clicked on the hard disk and found that Davids hard disk was now a folder on the neighbors desktop. Any and all of Davids files were viewable, editable, and deletable. Although this scary scenario is now less likely than in the last few years, a hacker can gain access to your computer with little more effort than that taken by Davids neighbor.

 Perhaps you are not worried about hackers looking at your home computeryou figure that it would be too much bother for them to find you and figure out how to break in. Unfortunately, there are many high school and college-age hackers with powerful tools (downloaded for free from the Internet) that allow them to automatically search for vulnerable computers and hack into them for fun. The programs systematically ping a range of Internet Protocol (IP) addresses by sending a packet of data and then waiting for the automatic response that all computers are programmed to send out. When they get a response, these programs log the address of the responding computer for the hacker to examine later. In fact, these ping sweeps happen constantly, and your computer has probably been pinged in the last 24 to 72 hours if it is online. After finding a vulnerable home computer, a hacker often can enter without any passwords at all.

Do the security issues of broadband keep me from having my home computer on 24/7/365? No, I have a home network with appropriate security set up. Countless variations on home networks are available. The purpose of this article is not to go through the spectrum of such networks. You can, for example, find a great amount of basic information and home network product reviews at www.practicallynetworked.com.


I will summarize an easy-to-set-up, wireless home network that can cost less than $200, allows you to legally (if concerned, consult your local broadband provider) share your broadband connection, and provides effective security for your data.

Remember, there is no such thing as absolute security. If someone wants to break into your house, it doesnt matter how many alarms, armed guards, and locks you haveif they want to bad enough and have the resources, they can break in. The same applies to any computer network.

The core of the home network is a network router, firewall, and Dynamic Host Configuration Protocol (DHCP) server (often termed broadband router). Dont be put off by the technical wordsthey are simple to understand. Physically, these three things come together as a small box the size of a paperback book that costs from $89 to $179. They are manufactured by companies such as SMC, 3Com, Orinoco (Lucent), and LinkSys among many others. You have probably seen them and walked right past them in your local computer store. When you install one of these boxes, the cable wire comes into your house, connects to a modem, and the output line of your modem connects to the broadband router. We will now discuss each of these three basic functions in turn.

 The first function is that of a DHCP server. For those of you who currently have cable modems without a broadband router, your computer is hooked directly to the cable modem and is automatically assigned an IP address by your cable company or ISP. The IP address is a simple number that looks like this: 1.160.10.240. This address uniquely identifies your computer on the Internet, much as your home address uniquely identifies your home. Without an IP address, you cannot access the Internet or be a part of a network. IP addresses sometimes change from week to weekbut that is beyond this discussion. In any case, the IP address is the unique address that can tell hackers where your computer is and also serves as your front door. When you install the broadband router, you replace your computer with this small box. The box itself is assigned the IP address and communicates back and forth with your ISP. Installing this device can now be done in a few minutes, with little or no knowledge of computers, through new wizards that come packed with the boxes.

So, you may be asking, where does my computer fit into all this? Quite simply, one of the outbound ports on your router (typically there are 3 to 12 wired ports) is plugged directly into your computer with an Ethernet wirethe same one that you used to plug your computer to the cable modem. All that is left is getting your computer an IP address, a necessary requirement to being on the Internet or part of a network. This is done automatically by the DHCP server. This part of the box creates a set of private internal IP addresses. Not only are these addresses not visible to the outside world, but they are a special type of address that will only work in a local area network. You cant use them to reach a remote computer across the Internet even if you knew what the private IP address was. The DHCP server leases and assigns an IP address to each computer that gets hooked up to one of its ports and requests an IP address to be assigned. Once this internal IP address is assigned and automatically installed on the computer, the computer can be a part of the network. For the sake of argument, lets say that you hook up three computers to the router/firewall/DHCP server, and that each is assigned an internal IP address.

 The next function is the router function. A router is a device that connects at the gateway between two networks. The router receives packets of data from one computer, determines the next point where they should be forwarded, and forwards them. The router acts like a postal sorting machine, which reads zip codes on letters and automatically sends them to the next sorting facility or postal carriers bag.

In the case of the home network we are building, the router is located at the gateway between the outside networks or WAN (in this case, the Internet) and the internal network or local area network (LAN). If you start to download a file from the Internet, your computer with an internal IP address makes a request in the form of packets of data and sends it to the router. The router understands that the request is intended for a Web server (say www.tucows.com) and forwards the packets to it. When the file begins to download from the server, the router receives the packets of downloaded data, and routes them to your computer alone, and not the other two computers on the network. This type of routing happens all along the path of data, but only your local router is relevant to this discussion. Alternatively, the router can receive a download request from one computer on the internal network to another computer on the same network. It channels these packets and the responses between the two internal computers.

 One of the advantages of having private IP addresses is that these are not known to the outside, and cant be easily addressed by outside IP addresses. Therefore, your internal computer exists in stealth mode, invisible to the outside world. When your router sends packets from your computer to the outside world, it also changes the return IP address on the packet to match its own IP address and not the address of the original computer, using a method called Network Address Translation (NAT). One secure side effect of NAT is that the outside world never knows the actual internal IP address of your computers, making attacks by hackers even more difficult. In addition, your router can be set to ignore pings rather than reply to them as usually done. This simple measure will prevent your network from appearing on a list of ping sweeps by a hacker. Between these two functions, you have come a long, long way in protecting your data and privacy. To go further requires an active firewall.

 While excellent software firewalls exist (one of the best is actually free at www.zonealarm.com), the easiest and most effective firewall is one that is part of a hardware router. Firewalls are programs that examine each packet of data before routing it. They examine incoming and outgoing packets of data. Firewalls follow rules to determine whether to forward a packet or not. For example, a firewall can be set up to disallow any incoming packets that dont represent Web page traffic. Or they can be set up to disallow any packets that might be communicating on the port that allows downloading access to your computer.

Going outbound, a firewall can easily be set up to prevent specific computers on your network from accessing particular Web pages. Particular computers can be blocked from Internet access between 6 p.m. and 9 p.m. on weekdays (homework timeare you getting some ideas?) and can be set up to similarly exclude things like game traffic and Instant Messenger traffic. The usefulness of a firewall is only limited by the creativity and thoroughness of the network administrator (you, on your home network). Be warned, though. It is easy to change the default setting of the firewall, but to do so you will need to first educate yourself to understand some technical jargon.

 Finally, most of the inexpensive routers now come with a built-in wireless access point. This means that the router and DHCP server can also route packets and assign IP addresses to laptop computers and desktop machines that are connected through wireless data transmission that is almost as fast as wired linkups. Operating more or less like a cordless telephone, your router can communicate with a wireless card on each computer for $50 to $80 and connect them to the Internet and your internal network without any wires. The ability to connect to your local network without wires can be a godsend for the home or office that lacks network cabling installed in the walls. Most wireless routers/ DHCP servers can handle over 200 wireless computers. If you are going the wireless route, you will need to use data encryption that is built into these devices.


To set up an internal network you will also need to make a few setting adjustments on your individual computers so that they can communicate with one another. These are easily done, but are beyond the scope of this article. Information and resources can be found at various sites, including the Practically Networked site mentioned earlier.

Finally, if you decide to be a real network pro, you can buy a slightly more expensive broadband router that can act as a VPN endpoint. Using software built into the operating system, you can create a completely encrypted, secure pipeline between any computer (at work, a laptop in hotel, and so on) and your computer at home. This VPN gives you complete access to all the files on your home network (as long as the computers are turned on) with great security.

The principles discussed in this article are completely applicable to a physician group practice or clinical trial site. With a minimum of fuss you can set up a network for your clinical trial site, and have wireless laptops available in your exam rooms for collecting subject data. The future is yours. Carpe diem!            

Technology Viewpoint

Paul Bleicher examines the pros and cons of broadband Internet technology and explains how to build a home network and make it secure.

Building a Home Network

Doubts are nearing crescendo level on the merits of the European Unions regulation of clinical trialsand not only within the European pharmaceutical industry. Investigators, ethics committees, and the regulatory authorities themselves are starting to question just how feasible the impending changes are as the EU buckles on the final implementation details of its 2001 directive on clinical trials (pharmacos. eudra.org/F2/eudralex/vol-1/new_v1/ dir2001-20_en.pdf), with its formidable range of new data requirements.


It wont be so easy to implement it in the current system, 

 was told at the start of September by no lesser figure than Daniel Brasseur, MD and chairman of the Committee for Proprietary Medicinal Products (CPMP) at the European Agency for the Evaluation of Medicinal Products (EMEA). Brasseur, speaking both as a leading academic clinician and in his formal role as the head of Europes principal scientific body for the evaluation of medicines, openly admitted that problems exist both with the current functioning of the clinical trials rules, and with the prospect of more than a dozen new detailed guidelines and subsidiary directives due to come into force next year.

This body of rules may lead to the generation of large quantities of clinical trials data, according to Brasseur. But there is no guarantee that it will be of benefit, he said.

Brasseurs remarks suggest a lack of overall strategy in the approach to medicines regulation in Europe. The EMEA is currently undergoing administrative reorganization, he pointed out. The major review of the entire body of European pharmaceutical rules now underway leaves many of the established procedures and practices for drug registration and review open to question. Enlargement of the EU is just around the corner, with the prospect of membership increasing from the current 15 member states to 25 by 2004. On top of that, the new clinical trials rules agreed to in 2001, which emphasize good clinical practice (GCP), are now in the final stages of implementationincluding this large body of highly detailed extra guidelines. Handling all the details simultaneously and successfully is a major challenge, Brasseur indicatedparticularly since the new guidelines will not necessarily fill all the gaps.

Making the new rules work will be one thing. But making intelligent use of the material generated by the new rules will be quite another, he went on. It will be a major administrative burden principally for the EMEA, rather than the CPMP itself, and I have some sympathy with them for the workload they will have to cope with, he said.

Brasseurs concerns relate not merely to the size of the task, but also to its utility. The question is, who will be interested in all this data? he asked. What is an ethics committee to make of all this information? They are interested in ethical questions. But this is the sort of material for an institutional review on a scientific basis, which an ethics committee is neither equipped nor mandated to carry outand does not wish to. Its role is to make ethical assessments.


Heavy criticism of the new rules is coming in from all sides. Brasseur spoke to 

 at the end of a meeting he attended in Brussels of the European Forum for Good Clinical Practice (EFGCP), at which a long sequence of diverse experts pointed out concerns over the dozen or so texts that the European Commission is now finalizing as its detailed implementation rules for the European Commissions proposed List of implementing texts of Directive 2001/20/EC. These texts provide detailed guidance on specific areas of GCP, including principles, submissions to ethics committees and to competent regulatory authorities, databases, adverse reactions, inspections, and good manufacturing practice compliance (many of these proposals have been summarized or highlighted in this column over recent months). They will determine responsibilities on the part of sponsors, investigators, ethics committees, and regulatory authorities in the practice of conducting clinical trials in Europe.

Comments came from many people, including representatives from ethics committees, patient organizations, regulatory agencies, industry, academia, and specific research areas such as oncology and pediatrics. EFGCP chairman Roger Bickerstaffe, of Solvay Pharmaceuticals, summarized the discussion, noting that because of overlap, things are described differently from one document to another, with much imprecision in the way theyre described in addition to a great deal of redundancy and uncertainty. Youre required to collect so much data that the question arises as to how to derive a signal from it, Bickerstaffe said, adding: Industry itself will find it difficult, and regulatory authorities simply arent in a position to do so.

Doubts and questions about guidelines and guidance dogged the meeting.

GCP and medicinal products for human use.

 On the Detailed guidelines on the principles of good clinical practice in the conduct in the EU of clinical trials on medicinal products for human use, questions were raised about whether the responsibilities of the various GCP actors are set out sufficiently clearly, and about how consistent these requirements are with ICH GCP.

 On the Detailed guidance on the application format and documentation to be submitted in an application for an ethics committee opinion on a clinical trial on a medicinal product for human use, the questions covered the extent to which this guidance meets the needs and expectations of ethics committees, researchers, and sponsors, and whether the guidance is sufficient. Doubts were expressed about the extent to which it will contribute to efficiency in review, to improved subject protection, to reducing the workload for ethics committees, and to increased speed of review.

 On the Detailed guidance for the request for authorisation of a clinical trial on a medicinal product for human use to the competent authorities in the European Union, notification of substantial amendments and declaration of the end of a clinical trial, and the Detailed guidelines on the trial master file and archiving to implement the directive on Clinical Trials on medicinal products for human use, anxieties related to how far the guidance provides for a harmonized EU approach to the commencement and conduct of clinical trials, whether it provides for sufficient protection of clinical trials subjects within and across member states and in non-EU countries, and whether the requirements are sufficient or excessive.

 Participants expressed worries over the definitions and procedures for reporting suspected unexpected serious adverse reactions (SUSAR) when the meeting examined the Detailed guidance on the collection, verification and presentation of adverse reaction reports arising from clinical trials on medicinal products for human use and the Detailed guidance on the European database of Suspected Unexpected Serious Adverse Reactions (EudravigilanceClinical Trial Module). Participants offered conflicting views over the extent to which the guidance provides standards for uniformity and efficiency, and to what extent research subjects in clinical trials will be better protected.

 On the Detailed guidance on the European clinical trials database (EUDRACT Database) and Detailed guidance on the European database of Suspected Unexpected Serious Adverse Reactions (EudravigilanceClinical Trial Module), queries were raised over what oversight these databases actually provide the European regulatory authorities. Uncertainties remain on how data will be received, codified, and entered into the databases, and over who will have access, how confidentiality will be guaranteed, and how the databases will be linked.

 When the meeting discussed the Detailed guidelines on inspection procedures for the verification of GCP compliance to implement the directive on Clinical Trials on medicinal products for human use and the Detailed guidelines on the qualifications of inspectors who should verify compliance in clinical trials with the provisions of Good Clinical Practice for an investigational medicinal product to implement the directive on Clinical Trials on medicinal products for human use, questions were raised over what will be the inspection process in the European Union, how inspections will be coordinated within and across member states and with non-EU countries, what standards will be used for inspecting clinical trials, how inspection results will be communicated, and to whom they will be available.

Real questions arise over the extent to which these guidance documents will improve the European clinical trials environment for the benefit of researchers, research organizations, and above all, subjects, EFGCP states in its meeting documentation. The past 12 years of debate on European GCP and the need for a general European clinical trials framework is very much at stake in the specifics of what this set of guidance proposes, the EFGCP document states, adding that the general consensus achieved by the 2001 Directive is under threat, because of specific concerns about the way these leftover issues are being handled in terms of detailed guidance. The broad consensus among the critics is that the texts go beyond current European Union GCP and ICH GCP, adding new guidance based on the legal framework provided by Directive 2001/20/EC.


Much of the criticism leveled at the texts focused on the inconsistencies and redundancies but, above all, on the extensiveness and the provision for member states to add further rules of their own. A Turkish participant at the EFGCP meeting expressed bewilderment and disappointment at what had emerged. In Turkey, we want to be more involved in European trials, and we are anxious to upgrade our understanding so we can do that. I came here expecting to learn how the European Union wants us to do things. But I find that there are contradictions even in what Europe wants. A Wellcome Trust delegate warned that the catchall nature of the new rules, with their heavy requirements on data generation, would imperil clinical trials in the noncommercial sector.

Roger Bickerstaffe spoke of a disappointed dream, noting that when the initial concept of an EU regime for clinical trials was floated in the early 1990s, the aim was to make things easier for everyonefrom sponsors and investigators to subjectsthrough a process of harmonization. But, he told 

 this objective has been lost. Instead, what has emerged is an accumulation of all the national requirements. 

As Jean-Pierre Tassignon, an EFGCP member, put it in his concluding remarks to the meeting: The dream was of a single European regulatory environment for clinical trials and an improved competitive environment for EU. The dream was of a solution to the real losses occurring in Europe because the EU was too slow, partly because of slow clinical trial authorization, particularly through ethics committeebut this has changed. Increased focus on safety, and decreased capacity by the European Commission to impose itself on the EU member states, has robbed the exercise of its original objective. On one of the central questionstrial authorizationthe new documents threaten to extend controls so that it will no longer be just a matter of submitting data and getting an approval number and presenting a final report. It is being turned into a system of constant reporting in minor detail, he said.

 afterwards: The tendency of member states is to go further all the time. And the Commission is protecting the prerogatives of the member states. For them, he said, safeguarding their right to regulate at a national levelparticularly on authorizations, inspections, and ethics committeesis the principal concern. For member states this is a matter of life and deathand we are caught in the middle, he said. Too much detail exists, for instance, about the requirements on misconduct or the prior details of subjects per center. Our dream as patients, ethics committees, investigators, and sponsors is to have a rethink of all these documents in an integrated way to eliminate inconsistencies, Tassignon concluded.

Members of the EFGCP are very worried. The final shape of these guidance documents will determine the EU regulatory framework for clinical trials in the future, the organization states. And EFGCP members arent happy with the shape that is emerging.


1.	European Parliament, Directive 2001/20/EC of the European Parliament and of the Council of 4 April 2001 (European Parliament, Brussels, Belgium, 2001).            

View from Brussels

Leaders in the European regulatory community examine the implications and merits of the EU?s new clinical trials regulations?and find themselves buried in minor details.

A Dream Unfulfilled: European Clinical Trials Rules Are Not Working

Trainers are relegated near the bottom of the pecking order in the corporate environment. Training is a dirty word. Training departments are looked upon as the holding pen for under-performing line staff and has-been managers. Training programs are considered to be an onerous burden on an employees schedule. And not surprisingly, training is always last on the budget list and first to be cut. When implementing change in the clinical research process, especially when introducing new concepts or technology applications, this attitude is fatal.

Maybe we need a new word for teaching people how to use new processes. We are bored by the idea of training. Training can be passive, repetitious, artificial, and divorced from reality. None of these adjectives need be true about the learning process, of course, but too often they are. So lets use the word learning for now, and talk about how essential learning is to process improvement.


Process improvement is a learning process: we learn by analyzing the root causes of delay, dysfunction, or inefficiency; we learn by mapping how work actually gets done; we learn from the metrics we use to gauge where we have been and where we are getting to.

The phrase lessons learned is an essential component of new technology introduction: if your company pilots new software without formally and proactively documenting your experiences, you willfrom your earliest attempts to use the softwaresimply repeat the mistakes and missteps of those who have gone before you. The more you use the software the more you learn, and the more you will lose if someone doesnt keep track. Even so, few companies have a formal process for capturing their experiences. Well-established procedures for instituting such processes existindeed, you may have colleagues in another area of your company who have already drawn up a set of procedures that you, too, can use.

Being a learning organization or building a knowledge management system are two of the most fashionable phrases in process improvement today. The logic of these concepts applies well to clinical operations, but do not let the textbooks on these subjects intimidate you. Do not think that these concepts are beyond the capability of clinical operations staff to exploit. Instead, let your efforts at learning be reinforced by these fashions.


As integral as learning is to process improvement, technology adoption, and knowledge management, training is rarely budgeted for. No dollars, no learningat least no formal, proactive, and comprehensive learning. This universal lack of funding for training is one of corporate lifes most self-defeating characteristics. 

The reasons are obvious. Which is more efficient? To teach someone how to do a monitoring site visit according to company SOPs, or to let them do a few visits and then explain to them what they did wrong? To teach them the contingency plans for software user problems, or to wait until you realize you arent getting the benefits from the software you expected?

And yet, if you put a training line item in your regional monitoring program budget, or your EDC implementation budget, it will almost always be at the bottom of the spreadsheet and the first to go when your boss, or your contracting officer, wants to prove how great at cost-cutting he or she can be.


How do we get by as well as we do if training budgets are always cut? We do it through the creativity, the conscien-tiousness, and the goodwill of our staff. We, as corporations, learn through serendipitythrough the initiative of individuals to mentor their juniors, share information with their peers, and record their mistakes and their insights. Think of how much more productive we would all be if this valuable insight was captured and delivered in a conscious and professional process of learning.

Whats the problem with serendipitous learning? Today most process improvement projects, or technology adoption initiatives, are so complex and line staff so busy with their everyday work that the learning required to make these innovations succeed must not depend on happenstance. You cannot rely on your colleague in the next cubicle to tell you what he learned when he tried to use that clinical trial management system (CTMS) for the first time. You cannot rely on your fellow monitor in the next state to tell you how to avoid getting your expense report kicked back by accounts payable. Companies invest millions of dollars to design process innovation or roll out new enabling software. To not spend the money to go that last mile of learning makes a mockery of those investments.


At least two key points about institutional learning are highly relevant to clinical operations.

 First and foremost, once is never enough. For busy, expert professionals used to doing work one way (and doing it well), you cannot expect them to absorb the learning required to implement the new process correctly in one sitting. We always recommend a three-part program. The trainer provides an overview of the process or software, during which he or she hopes to plow the field and ready it for planting. Next, the trainer goes over everything again, this time in more detail, planting the seeds. Finally comes the real training, just in time, where the seeds sprout, take root, and grow. At the moment when the students start using the process or software, the plant can begin to blossom.

 Second, when it comes to technology adoption in particular, you cannot abdicate the learning process to the software vendor. What needs to be learned in a software implementation situation is one-fifth screen navigation and four-fifths business rules. Business rules comprise the hundreds of micro decisions that a company needs to make about how yougiven your particular organizational chart, staff skills, strengths and weaknesses, monetary and time resources, and so onwill actually use the software. For instance, a CTMS may enable your staff to write a site visit report (SVR) within the application. How does this fit with your SOPs? What information contained in the SVR should be communicated to other departments? Can this communication be automated? Should it be? Do you want changes to SVRs to be recorded and auditable? The vendor cannot decide these things for youonly you can make these choices, and once made, they must be taught.


Some companies pat themselves on the back for taking care of training by spending money on an innovative learning technology; eLearning, for example, is another heavily promoted fashion. Many techniques of training delivery can enhance and make learning more efficient (none more powerful than a good teacher, something no technology can replace). But delivery is secondary to the contentsecondary to having content, to making sure that your company does not cut training out of the budget, to taking the time and expense to document lessons learned, to making the choices among business rules, and to devoting the staff time to learn them.

Weve all heard the aphorism, Those who can, do; those who cant, teach. I offer a new aphorism for clinical research improvements. Those who dont learn wont succeed; those who do, will.            

Process Improvement

Putting staff training at the bottom of your list of important tasks is a self-defeating attitude in the business world.

Learn Before You Leap

This month we interview a faithful subject who has stuck with the Womens Health Initiative trial of strategies to improve the health of postmenopausal women despite adverse effects, a 500-mile move, and the early stopping of part of the trial. 

 A little over five and a half years ago, I received a brochure about the study in the mail. At that time I was working as a certified nursing assistant (CNA). I thought it was a good idea, because womens health issues had not really been addressed. Im 66 years old, and I thought that the study might be important down the road, although I didnt think at the time that it was going to be of any benefit to me. I talked about it to several people I knew who had received the brochures, but they didnt go ahead and contact Womens Health Initiative. I did. 

 What were some of the reasons that your friends were hesitant to participate?

 They didnt say. You know how people say, Oh, gee, thats an excellent ideamaybe next week? 

I was drawn to this for various reasons. My mother lived to be 93. She had never had a Pap smear until she was in her late seventies. She had never had a mammogram. Thinking how my parents lived and what we know now, I just thought maybe a study like this would get information down in black and white and help some people. I have a lot of friends who have had hysterectomies. I thought that this study was important for womens health issues. Also, I had had [some gynecological problems] in the past, and so I thought maybe it would be a good idea to find out about estrogen and some of the other things the Womens Health Initiative was addressing. 

So I went ahead and contacted them and went in and had an interview. They explained what was going on, and I said, Yes, I would like to go ahead and be involved.

 Had you already been taking hormone replacement therapy? 

 I had, but I had stopped because of my age. When I hit my early forties, I had hot flashes that were so intense I couldnt sleep nights. So I contacted my gynecologist, and they put me on the hormones. That was a while ago, but as I remember, I said, Well, if this alleviates my problems, how long will I be on? And he said, How long are you going to live? That was 20-some years ago.

 So you took them and they helped you, and then you tapered off using them? 

 In this trial, what were your clinic visits like? 

 They were very professional. Once a year, we would have a full exam, and a mammogram and a Pap smear were necessary to remain in the study. Every six months, you would fill out a questionnaire to see if particular events had occurred. They would check to see what kinds of over-the-counter medications or supplements you were taking, such as vitamins, besides the test pills. 

Between visits, they would mail us updates on what was going on with the study, some pertinent things that had come outfor example, that there was a higher tendency for stroke. And then, because people dont always check their mail, they would verify when I would go in that I had read it: Did you see this, and were you aware of that? 

If you had any leftover medication, you took it in at those times so they could count it. They asked if you took them every day, or how many times you missed. 

I missed one appointment in five years. I had a March appointment, and we had snow in Nevada, and I couldnt come in in person because of the condition of the roads. So they took my information over the phone, I mailed in my unused medication, and they mailed me my new supply.

 Did they give you any kind of a symptom recording sheet or diary to keep track of your symptoms? 

 Yes, it was calendars on which you made notations for spotting or anything that was pertinent. But we were also told to call a toll-free number if anything happened, and the nurses were excellent in calling back. 

 Did you have any medical problems during the study? 

 When I signed on, I didnt know what I was going to be takingthe real McCoy or a placebo. I was working as a CNA, and theres a lot of heavy lifting. Sometimes I would pull a double shift. Anyway, I started spotting [bleeding]. At that time, being that I had not been taking anything else, I knew that I was taking estrogen. Of course, youre supposed to call if you have any problems. I called and told them, and then they put me on the progestin, and that kind of corrected it. I went off of that during the study a couple of times because they figure that the body will stabilize, and even to the end I was still spotting through even though I had quit work. Women who do more manual labor sometimes have more problems with their periods. I dont know if they published that or not, but when I went in, they said, We are finding that women who do more manual labor instead of sitting at a desk seem to have a few more problems.

 During the study, you moved to a different state. 

 I thought the study was important enough that I asked if I could remain in the study when I moved, and they said, Of course. They like to keep people in it as long as they can. 

 But now the closest site for you is 100 miles away. 

 Yes. Although when I did go in and have a sonogram to check my innards because of my bleeding, they set up an appointment for me in [a city 50 miles away]. So that was nice; it saved me a little extra drive. But I loved working with the people in both places. 

 So you were never tempted to drop out of the study? 

 No, it never entered my mind. Thats not because Im dumb, either. [laughter] I had been instructed [about the risks], but I just thought that as a widow, and at my age, maybe it would help my grandkids or something down the line. I had no idea that the findings were going to get to the point where they decided it was not good to proceed [with the estrogen part of the trial] for women with uteruses.

 How did they notify you that the estrogen part of the trial was being stopped? 

 I called the nurse practitioner for some other reason, and she said, Have you seen the national news? She said, The story is going to be breaking tomorrow in local news. We are in the process of calling all the people, and we will be contacting everyone afterwards to inform you if you were on the placebo or the study medication. I reminded her that I already knew what I was taking, but somebody else on the staff did call and verify that I had received all the information.

 How did you know whether you were receiving the drug or the placebo? 

 Well, I had to know, because I was bleeding. When I called in [about my bleeding] and they sent out the progestin, it was pretty obvious that I was getting estrogen. They did perform other tests to verify there were no further problems.

 Is there anything that you think the study personnel should have done differently? 

 Ive never been in a trial like this before, but I had worked in the health field for a few years. And I was quite impressed with all of the personnel. They treated you like you were the most important one out of the 18,000.

It seems to me like it was a very well-organized thing. If it had been kind of fly-by-night, I probably would have stopped, but I was quite impressed. It didnt make any difference who I spoke to or which nurse examined methey were always willing to answer questions. If they didnt know an answer when you talked to them on the phone, you got a call back after they pulled your file. 

 So if you had an opportunity to participate in another clinical trial, would you do it? 

 Yes, I would probably go ahead if I thought it would be a benefit. Not only for womens health, but I was drawn to this because most of the tests that had been going on when I was younger all concerned men, and womens health issues had not been addressed.

 Has your experience changed anything about your own health habits? 

 I dont think that a pill is the way to correct every problem. Like, you know, people taking pills to lose weightI dont think so! But now the medical field has the results of this test, and that should wake a lot of people up. But if my doctor were to say, I think you should be taking some estrogen because of this or that, I would probably do it. They know better. I know that there are alternatives, but Ive known that all along. I took some of those in the last 20-some years. As a matter of fact, I take calcium plus the test pills now. [Faithful Subject is still participating in the calcium and vitamin D supplementation part of the study.]

The Womens Health Initiative was also doing a dietary study. Probably because of the work I was doing and my husbands health, I was eating properly and I couldnt join that study. [They were looking for] people who were overweight or not exercising, and they said, You dont meet the specifications for what we are looking for. But I would have joined it. I just thought that it was an important study.            

Both personal health considerations and altruism helped keep this subject in a long-term study.

A Faithful Subject

Feature Article

Integrating IVRS with in-house clinical trial management solutions can provide a powerful project management tool.

Using IVRS in Clinical Trial Management

Meet the EAB

Bert Spilker has been involved in clinical trials for 30 years. He?s written books, founded companies, and helped steer regulatory agencies. Now he wants to change how drugs are developed.

Bert Spilker

Some companies prefer live training for PIs; some are moving toward virtual training. Two Lilly authors advocate a combination of the two methods.

The Virtual Training Process

Careful attention to geography can help sites enlarge their catchment areas and increase subject accrual rates without negative consequences.

Applied Clinical Trials-10-01-2002