Ethical Considerations When Preparing a Clinical Research Protocol

Evan DeRenzo, Eric A. Singer, Joel Moss

• Publisher: Academic Press
• Year: 2020
• ISBN:
  9780123869548
• Edition: 2
• Topic: Medicine
• Subtopic: Medical Theory, Practice & Reference

Ethical Considerations When Preparing a Clinical Research Protocol, Second Edition, provides a foundation for improving skills in the understanding of ethical requirements in the design and conduct of clinical research. It includes practical information on ethical principles in clinical research, how to design appropriate research studies, how to consent and assent documents, how to get protocols approved, special populations, confidentiality issues, and the reporting of adverse events. The book's valuable appendix includes a listing of web resources about research ethics, along with a glossary, making it an invaluable resource for scientists collaborating in clinical trials, physician investigators, clinical research fellows, and more.

• Walks investigators and trainees through the identification of the ethical aspects of each section of a clinical research protocol
• Includes case histories that illustrate key points
• Contains information on conducting clinical research within the pharmaceutical industry
• Includes internet resources and worldwide web addresses for important research ethics documents and regulations
• Contains a chapter on Study Design and Methodology that is purposely expanded to explicitly address biostatistical considerations

Section B

Preparing the protocol

Chapter Four

Designing a clinical research study

Abstract

The heart of study design is asking a question worth asking or testing a hypothesis worth testing. Shaping an acceptable question or hypothesis begins with an idea that interests an investigator or a sponsor. The idea or interest is ordinarily followed by a thorough literature review. It is critically important to ascertain that the question has not already been answered or that the hypothesis has not already been adequately tested. Conducting a sufficient literature review can be a substantial task.

Keywords

Study design; Clinical research; Literature review; Clinical trial; Drug; Protocol

1 Shaping the study question or hypothesis

The heart of study design is asking a question worth asking or testing a hypothesis worth testing. Shaping an acceptable question or hypothesis begins with an idea that interests an investigator or a sponsor. The idea or interest is ordinarily followed by a thorough literature review. It is critically important to ascertain that the question has not already been answered or that the hypothesis has not already been adequately tested. Conducting a sufficient literature review can be a substantial task.

Although most scientific literature is accessible electronically, the history of medicine is full of examples of important information discovered and then lost for years, generations, or centuries. Even when searching the electronically catalogued sources, sometimes data related to a question or hypothesis are hard to find because they were published in low-visibility or obscure journals. The core medical and medical research journals may not include relevant early or obscure data. This problem is not merely academic but is also of grave concern for the protection of human participants.

The problem of connecting seemingly unrelated and/or temporally distant literature citations may surface only in the wake of a clinical research disaster. An example is the research tragedy of the deaths and emergency liver transplants that occurred in a trial of fialuridine, or FIAU (Straus, 2002; see also case discussion in Chapter 15). Although the cause of the toxicity that was finally identified had been reported in the literature before the FIAU participants became ill, the literature pointing toward the problem was sporadic and not easily understood to predict the adverse events that ultimately occurred. To make the good faith effort needed to review all the possible relevant literature, the responsible investigator will have to reach beyond the electronic citation cataloguing system and scour the records systems for relevant information from multiple related disciplines and, perhaps, from literature published before electronic cataloguing started. Engaging the help of an information specialist and/or pharmacist might facilitate review of the literature and can strengthen the investigator's and IRB's confidence that all pertinent data was reviewed. This is what would be expected, ethically, to think one has done one's job to understand potential risks for human research participants. But even if a vigorous and thorough literature review does not reveal concerning information about the proposed study, other reasons may exist that indicate the proposed question or hypothesis is not appropriate for clinical research. The study may simply not be important enough.

Central to the ethical conduct of clinical research is the requirement that unless important information can be reasonably expected from the completion of the study, it is unethical to subject anyone to even mere inconvenience. Importance is used in its broadest sense. Adding to the world’s knowledge about human and/or animal health, welfare, disease, or suffering and learning more about how humans and animals function in their environment may well advance medical progress. The breadth of important knowledge to be obtained is wide. Nonetheless, ethical and scientific justification for the study and the knowledge expected to be gained are critical parts of the process of shaping a clinical research question or hypothesis appropriate for involvement of human research participants.

Answering even an important question, or testing an important untested or inadequately tested hypothesis, does not ensure the appropriate design of a study question or hypothesis. Timing is another critical factor. It may be that extending the research to humans is premature. The risk of harming a subject may be too great because available knowledge is insufficient to ensure a safety level acceptable for human participants. Balance between the ends of science and protection of the rights and welfare of human participants must be tipped in the right direction.

A sad example of the perils of attempting to answer research questions with human participants too quickly is the study of fetal brain tissue transplantation in Parkinson’s disease (Albin, 2002; Clark, 2002; London and Kadane, 2002). All one needs to know about the history of this research is that the promise demonstrated in rats in the late 1970s and in nonhuman primates in the early 1980s may have been followed too quickly by the attempted application to humans in the 1990s. The transition from bench to bedside moved so rapidly that clinical trials in the late 20th century saw the development of unanticipated side effects in human participants characterized as catastrophic on the front page of the New York Times (Kolata, 2001). The article’s conclusion was to return to the bench before considering further fetal tissue research in humans with Parkinson’s disease. (See Chapter 15 for a full discussion of this case.) Once that the particular approach has been paused or completely scrapped, other scientific and perhaps more ethically sound approaches can be explored. If the scientific findings continue to be promising, investigators will arrive at the original studies' hoped for success (Venkatesh and Sen, 2017). Preventing risks to human participants is critically important, even if they are lined up to volunteer. It is the ethical responsibility of the investigator to maintain "intellectual honesty" and temper one's enthusiasm for one's own work. Remembering that rushing can put research volunteers at risk is the investigator's responsibility. That is true, also, for being self-reflective about the ethics of the science as the science, itself.

Moving from bench to bedside must be done carefully and deliberately. Only in this way can the rights and welfare of human participants be effectively protected. Urgency to find treatments or cures for horrible human diseases is not an acceptable ethical justification for conducting human studies poorly. The prospect of serious unanticipated outcomes must always be a prime concern of a researcher who is shaping a question or hypothesis for investigation in human participants. From the very beginning of the development of a clinical research study, the science and ethics of human trials are intricately and inexorably intertwined. Losing sight of this interdependent relationship is a harbinger of disaster.

To craft an appropriate study question or hypothesis, the following questions can be used to guide the process:

• • What is it that I want to study? What am I most interested in learning?
• • What does the literature say about this topic? How widely has this topic been studied in the past?
• • What was the scientific path leading to the current results?
• • Logically, what is the next question to be answered or hypothesis to be tested to advance the knowledge and understanding of this phenomenon?
• • Does the state of the science justify initiating studies involving human participants? Are additional animal, bench, and/or computer simulation, and/or modeling studies needed?

By asking these questions, and discussing answers with colleagues, the clinical researcher will be on his or her way to shaping a scientifically and ethically sound question for clinical research.

2 Selecting the study design

Study design is a complex process including selection of the optimal population for study, data points for analysis, and analysis strategies. Although selection of a study population ought to be concurrent with considerations of the study’s strategy, it is important that the clinical researcher be familiar with the range of study designs from the outset of the planning process.

2.1 Distinctions between hypothesis-testing and hypothesis-generating clinical research

First is the matter of whether the study is intended to test or to generate an hypothesis. This choice will be determined in large part by findings from the literature. Is the available knowledge about the chosen phenomenon sufficient to shape a plausible hypothesis? Today, the preference is for hypothesis-testing research—research that is designed to produce statistical support or refutation of a formally articulated research hypothesis. Funding sources and journal reviewers like to see statistically validated findings. But it is important to remember that a clinical researcher cannot produce theoretically plausible hypotheses if the information available is not adequate. It is just as important to the advancement of science to investigate previously unexplored areas to begin the data accumulation process needed to develop hypotheses. Although some reviewers will consider a study proposal that is not of a hypothesis-testing statistical design as merely a fishing expedition, studies that seek to generate significant new information are ethically and scientifically justifiable. Further, it may be ethically and scientifically required to perform exploratory analyses if data sufficient to build a credible hypothesis simply do not exist. The justification can be as strong for hypothesis-generating as for hypothesis-testing research.

An investigator is interested in looking for genetic variation or attempting to find genetic markers for diseases of the central nervous system that produce debilitating motor weakness. This is an important area of clinical research that is in its infancy. The investigator decides that there is not enough information to shape an hypothesis on which to focus. Rather, he or she decides to design an hypothesis-generating study in which a genetics component is added to every study at the researcher’s institution involving persons with a disease or condition that meets the diagnostic criteria. The investigator obtains permission to take blood or tissue that is left from clinically indicated procedures on the study participants and is going to scan their genetic material for variations of interest.

A contemporary example of how such hypothesis-generating work can produce great breakthroughs. We now have a single case report of a woman in Scotland who does not experience pain the way most individuals do. Because a researcher became interested in this unusual case, we may someday have novel pain therapies and the work that comes out of this research may change the whole way researchers consider the importance of human genetic material, what is presently called pseudogenes, that have previously been thought of as detritus. Further, this work is thought, perhaps, to someday make a dent in the opioid crisis (Habib et al., 2019).

Another kind of hypothesis-generating study is demonstrated by the natural history study of patients with a disease for which there is no treatment. As long as there continues to be no treatment, observing their disease course is an ethically acceptable way to learn more about the disease so insights into potential treatments and a cure may arise. At the point at which a treatment emerges, it will no longer be ethically acceptable to conduct a study watching the natural progression of the disease. This, setting aside consideration of all the deceptions and racist problems built into the protocol from the outset, the point we are making here is that a central ethical failing of the Tuskegee Syphilis study was that when penicillin was found to be an effective drug against infection that was the time at which the study should have been terminated and study participants treated, clinically, with penicillin. Once a disease that was previously untreatable has some effective intervention, that will be time to transition to hypothesis-testing studies.

2.2 Basic versus applied research

In clinical research, the difference between hypothesis-generating and hypothesis-testing research can often be thought of as the difference between research conducted to understand basic molecular structure and function thought of as basic research and what is called applied biomedical research, research that is at the hypothesis-testing stage. Research conducted to learn about the structure and function of basic biologic structures, depending on the degree of knowledge about the structure, is often at the stage of hypothesis-generating research. Although medical science understands much about the structure and function of human and nonhuman animal physiology, it still has much to learn. By the time research studies have advanced to the point of injecting novel substances into humans or testing new devices on cardiac surgery patients, for example, they have reached the level of hypothesis-testing. But sometimes human participants will be needed for both types of testing. It is important to remember that both are clinical research studies. A common mistake is to think that basic research, where there is no human involved and only materials from human beings are studied at the bench, is not clinical research, but that clinical research entails only those studies that involve putting an experimental agent or device into a living person. This is incorrect. Clinical research is research that involves humans or parts of humans. The distinction between basic and applied research in the clinical research sphere is the distinction between early stage human participants research (i.e., basic), where fundamental mechanisms are being explored and later stage research in which the area of interest is more focused and directed toward some explicit therapeutic intervention (i.e., applied), after sufficient data have been collected on which to build credible hypotheses. Examples of basic clinical research include the genetic studies just envisioned as well as brain imaging studies designed to learn more about how the brai...

Table of contents

• Cover image
• Title page
• Table of Contents
• Copyright
• Preface
• Section A: The basics: What you need to know before starting human participants research
• Section B: Preparing the protocol
• Section C: Procedures, methods, statistics, data management, and record keeping
• Section D: Special ethical issues
• Glossary
• References
• Index