“The research integrity panel continues the ethics discussion by focusing on standards and practices that promote responsible data collection and appropriate authorship byline decisions. The panel will explore issues related to current RCR training for data integrity and authorship as well as consider the impact that different international PhD educational standards can have on data integrity. The panel will conclude with a discussion by an international team who will describe their experiences in negotiating authorship agreements and in building capacity to assure data integrity.” (Workshop Agenda)
David Resnick, Bioethicist and Chair of the National Institute of Environmental Health Sciences (NIEHS) Institutional Review Board at the Department of Health and Human Services, began his talk with perspectives on collaboration and disputes. As the staff bioethicist, Dr. Resnick is often called in to provide advice when ethical issues arise in research. Most of the issues that he sees involve authorship, with some disputes arising about data management or intellectual property. Many of the authorship
1In this section and other sections summarizing presentations, views and opinions are attributed to the presenter unless stated otherwise.
disputes occur because the researchers had different understandings when they began the collaboration.
One way to minimize these disputes is to work out an agreement beforehand with some of the specific details of the collaboration and what is going to happen. These agreements are important for several reasons. Researchers may have different ideas about what an “author” is, different understandings of human research, and so forth. Educational background can play a role. An MD in one country may have very different training than a U.S. MD. Researchers also vary with respect to their knowledge and training on the responsible conduct of research and specific issues such as data management, authorship, publication, and research integrity.
Collaboration agreements should discuss authorship and how it will be determined. For example, the International Conference Medical Journal Editors Guidelines can be helpful. It should cover publications—where and when they are to be submitted. Data management is an especially important issue in large, international collaborations where research records may be kept in different countries, and in different languages. How will the data be stored, analyzed, and shared? Will any data be confidential, and how can this be protected? Management and sharing of research materials (cell lines, tissue samples, and reagents) should be covered as well. Intellectual property and funding expectations should be addressed at the beginning whenever possible.
Dr. Resnick pointed out that conflict of interest is a very serious issue. There are real conflicts of interest and perceived conflicts of interest—both should be dealt with because something one researcher might consider a perceived conflict could ethically tarnish the whole project. Also, concrete expectations regarding roles and responsibilities in the research and timelines should be covered. Finally, regulation is a very important issue, particularly in meeting requirements of different national jurisdictions and addressing any conflicts that do not harmonize.
Entering into a formal agreement implies that some level of trust already exists. So prior to concluding an agreement it is necessary to learn about one’s potential collaborator. Does he or she actually have the qualifications that they claim to have? How do you know that they will fulfill their commitments?
In practice, these sorts of agreements are not utilized very often. One reason is that researchers are not aware that they can develop these sorts of agreements. Another reason is inconvenience. Researchers may not want to take the time to develop an agreement, or want to avoid being legalistic
and work things out informally. As a result, misunderstandings can develop when the project is undertaken.
A written collaboration agreement does not supersede other sorts of agreements that exist in research, such as material transfer agreements, data use agreements, confidentiality agreements, cooperative research and development agreements, and memoranda of understanding. These are all important but none of them addresses some of the key issues discussed above.
Philip Altbach, Director of the Center for International Higher Education at Boston College, emphasized one key takeaway: researchers should know about the universities and academic systems in which their potential overseas partners work.
World higher education has been expanding dramatically in the latter half of the 20th century and will continue to expand. In the rich countries, the rate of expansion has by and large stopped. In fact most of the growth in global higher education in the coming years will come from China and India.
In the United States, most research is performed in doctoral-granting institutions. In much of the world doctoral education is not well-developed. The majority of those individuals teaching in colleges and universities around the world (excluding the top institutions) do not have doctorates. There are essentially two approaches to doctoral education around the world. It is important to understand what they are and how that is developing. One is the American style PhD, or course doctorate, which involves heavy coursework, a dissertation and examinations. In contrast, most doctorates in most of the rest of the world are research doctorates, where students register for a PhD, write a dissertation under a supervisor, and get the degree when it is complete. There is a move globally, in the direction of the American style PhD or a modified American style PhD.
Dr. Altbach explained that the academic culture of a potential collaborator may be an important influence on their priorities and behavior. For example, countries may lack full academic freedom, which may raise issues connected with access to data and controls over the Internet, even in science and engineering. Some research topics may be restricted. University corruption in admissions, examinations, and promotions is also a problem in some countries. Even if one’s collaborator is not corrupt and wants to do the right thing, their context may make it difficult for them.
Other aspects of the academic culture have to do with the incentives and reward structures, such as the degree to which academic promotion is meritocratic or the degree to which faculty members at the institution received all their degrees at the same institution. Are faculty rewarded and recognized for superior research, or are they rewarded for just having been at the school for a long time? The culture of the academic profession and of the university is of central importance in understanding how people approach collaboration and their careers.
Dr. Altbach concluded by pointing to the importance of working conditions. Faculty members in most of the world have very heavy teaching loads, and are still expected to produce research.
More importantly, in most of the developing world, professors are not paid enough to work full time at their academic jobs. The collaborative agreement with an international partner may allow them to spend more time on their academic work. This is a positive aspect of collaboration. Other aspects of pay (compensation) are important to keep in mind as well. Faculty members may receive significant payments for publishing in a prestigious domestic journal or an internationally recognized journal, so receiving authorship credit is very important.
William Blattner, Director and Principal Investigator for the Institute of Human Virology (IHV) HIV Vaccine Trials Unit, University of Maryland; and Aliyu Gambo Gumel, Fogarty International Research Fellow, discussed their research project on Recruiting Acute Cases of HIV in Nigeria, the REACH Study. The sponsor was the Centers for Disease Control, and the research was undertaken collaboratively by the University of Maryland and the Nigerian Federal Ministry of Health.
Dr. Blattner gave an overview of the project. There was a large project team, with a variety of roles. The specific deliverable was the generation of samples in serial follow up from patients with acute infection, meaning that they were viral positive before they were antibody positive, very early in infection. CDC wanted these in order to develop and improve tests. The project had additional goals, such as to explore the magnitude and the correlates of infection, to address the operational issues of getting patients into treatment and care through the PEPFAR (U.S. President’s Emergency Plan for AIDS Relief) Program, and to
understand more about the genetic diversity of the virus and the performance of blood test screening assays.
The project worked with several populations. Some women came into the project through health facilities because they were pregnant and were being screened for prevention of mother-to-child transmission. There were also groups of brothel-based sex workers, street-based sex workers, and the so-called Okada motorcycle taxi drivers who sometimes collect their fee in non-cash payment, meaning sex.
The two-phase study involved consent for testing and then consent for enrollment in prospective follow up. The project team was looking for people who had not been previously tested, and outreach included efforts to promote safer sex among the target populations. About half of the sex workers are HIV positive. There are real challenges in getting services to this population. Although prevalence may be falling for Nigeria as a whole, certain populations are experiencing an increase. Those in most need generally have the least knowledge of what services are available.
Dr. Blattner reviewed the several different types of intellectual work product that resulted from the project. The Nigerian collaborators are coauthors of some papers. Others addressed the Center for Disease Control’s (CDC) need for laboratory test development. One goal was to ensure the Nigerian researchers were integrated at all levels of the program, not just the clinical and epidemiological aspects, but also into the more sophisticated laboratory pieces, and to meet the criteria for authorship credit. Ultimately, the Fogarty Program administered by the National Institutes of Health (NIH) aims to provide international partners with the skills to successfully compete on their own for independent research funding.
Dr. Aliyu discussed the ways in which the project increased the capacity for the Nigerian partners to ensure data integrity, by developing effective methods for data collection, storage, transfer and validation. There were three teams working together: counselors, lab technicians, and data managers. Through training, the team was able to essentially double its capacity for data integrity in the year prior to the launch of the project in May 2005. This capacity continued to increase in the areas of data collection, data capture, and data storage. Data analysis and interpretation was the only area that did not see a consistent, significant increase in capacity.
Several factors contributed to the increase in data integrity capacity. Training was mostly led by IHV (University of Maryland) faculty with some help from local resource people. Training covered issues in data collection, data capturing and storage, and analysis and interpretation. Also, clearly-
written standard operating procedures (SOPs) were very helpful. Having the right equipment and office facilities was also essential.
There were challenges as well. Most of the analysis and interpretation continued to be done in the United States, which is why there were not any papers with Nigerian lead authors. At the start of the project there were sometimes mistakes in data collection, such as multiple screenings of the same person, the same ID number being assigned to different participants, missing forms, and unsigned consent forms. At the beginning, some data had to be discarded, but performance improved over time. Some problems with faulty data entry, missing data, multiple entries, and database design continued to occur.
In the area of data storage and transfer, weak backups like flash drives and CDs sometimes caused problems. The instability of the electricity supply was a continuing issue, and required the project to buy backup generators. Multiple log books (lab, data capture, and counselor) were utilized to facilitate validation. This made it easier to discover cases of missing or incorrect data for participants. Weekly meetings allowed the teams to troubleshoot problems. Weekly conference calls with the IHV team and regular visits provided additional monitoring.
A future task will be to build capacity in data analysis and interpretation. Dr. Aliyu also reflected on his own experience, and on the value of long-term training he had in the United States. At the time of the workshop, he was designing a study to look at the possible relationship between bovine TB in Nigeria and HIV, responding to a surge in the prevalence of bovine TB. He has developed the protocols, the forms, the SOPs, and the database, and also secured ethical approvals.
The goal of building capacity among the participating Nigerian researchers and the broader research enterprise in Nigeria continues to be pursued by the Institute of Human Virology in Nigeria. It is a free standing academic center of excellence affiliated with a number of universities. The West African Bioethics Center in Nigeria has also established a national legal framework for research ethics. Although research funded by U.S. government agencies undertaken abroad will generally follow U.S. ethical conduct procedures, such as the Common Rule for human subjects protection, research collaboration is much easier when partner countries develop their own frameworks for institutional review boards, animal subjects protection, and so forth.
Participants in this breakout session discussed key points from several of the presentations given in the plenary session. Much of the conversation focused on data integrity and authorship issues as important areas that would benefit from additional discussion and understanding of international approaches. The group was especially interested in problems facing the three sectors—government, universities, and industry—and ways they could be addressed.
Individual participants made a number of points during the discussion. This is a non-exhaustive list, and is not intended to represent consensus views of the workshop or the breakout session:
- Possible Issues for Industry. Industry-supported research is moving offshore, particularly clinical trials in the pharmaceutical industry. Cost is an important factor driving this. Also, the incidence of some diseases is overwhelmingly in developing countries. Industry may need to takes steps to avoid the perception of being self-serving in the host country. When there is documentation of biased reporting, who is responsible? Do the researchers in developing countries know what they are supposed to do? There are complexities in managing the roles of contract research organizations (CROs), site management organizations (SMOs), study coordinators and investigators.
- Possible Actions by Industry. Good communication with the public by companies in the host country is helpful. Steps have been taken to increase transparency, such as establishing the clinical trial portal, but industry could better clarify responsibility and accountability in clinical trial roles. Industry could also get more involved in education and training in areas where companies have expertise, such as data integrity.
- Possible Issues for the U.S. Government. U.S. government agencies have regulations related to responsible conduct of research (RCR) training by grantees. Some government regulations related to research may not be applicable to modern research operations. Institutions may not take responsible conduct or research (RCR) seriously and only require some online courses.
- Possible U.S. Government Actions. The government might clarify clinical trial responsibilities and accountability. The RCR training plan and mandates could be put on the front page of the grant along with human subjects, etc. Government might require institutions to ensure that every researcher has had RCR training, specifically on data retention policies. Greater resources could be devoted to monitoring and enforcing existing RCR regulations. Institutions could be required to provide mentorship training. Agencies could support the development of better systems to ensure data stewardship and transparency.
- Possible University Actions. The quality of RCR training could be made part of regional accreditation. Universities could improve training programs for their own investigators. Universities could perform due diligence on collaborating institutions and investigators. They could specify research misconduct investigation procedure in all agreements.
- Societies and Journals. Journals could establish and strengthen requirements that data supporting published work be made public.