Most of us take modern medicine for granted. For example, antibiotics, vaccines, and CT scans are all now recognized as key components of clinical medicine. However, penicillin, the first widely recognized antibiotic, was not clinically available until 1940, and the polio vaccine was not introduced until 1954. It surprises many people to learn that the first CT scans were not clinically available until 1974. These three vital aspects of present-day clinical medicine have contributed to exciting health care advances. Each is also the direct result of research in both the basic sciences and the clinical arena.
The term research appears throughout this book with regard to diagnosing IBS, treating patients, and seeking new therapies. This appendix gives a very brief overview of research, describing basic science and clinical research, explaining how research studies are conducted, and defining some terms commonly used in research.
Studies that explore the basic workings and physiology of the human body constitute basic science research. These research projects may investigate the structure of red blood cells, how nerves regenerate, or how medications are absorbed into the body. Basic science research studies are immensely important in understanding how our bodies and our individual organ systems function. In addition, these studies are critical to understanding why our bodies do not always function normally and why diseases occur. However, the information obtained from basic science research studies may not directly translate into a change in the practice of clinical medicine. Thus, a research study may determine that a unique combination of 13 different chemicals is vital to the development of nerve cells in a tissue culture plate in the laboratory but may not have any practical significance for how patients are treated by their physicians. In contrast, clinical research is designed to answer questions that are directly relevant to patient care, and it often leads to improvements in care.
Clinical research studies generally have four distinct phases. The first phase is to ask a simple question that is clinically relevant, that is, a question whose answer may lead to a change in the practice of clinical medicine. In the field of IBS, this could be as simple as, “What percentage of patients who have IBS also have fructose intolerance?” If it is found that a large percentage of patients who have IBS are fructose intolerant (and patients who have IBS and diarrhea do often suffer from fructose intolerance), then physicians may start counseling their patients who have IBS to avoid fructose, in the hope of ameliorating their symptoms.
The second phase is developing a research protocol, a road map for the research study. It states who can and cannot enter the study, how the study subjects will be evaluated, what specific questions will be asked, what the intervention will be (intervention is defined below), and how the results will be measured. The design of the protocol is critical, as a poorly designed study will inevitably lead to poor results, in the form of data that are either not interpretable or not usable.
The third phase in the research project is to recruit appropriate participants and enter them into the study. This sounds easy, but it can be very difficult. Let’s say that you want to study whether a new medicine improves symptoms in patients who have IBS and diarrhea. You may decide to place an advertisement in the local paper asking for people who have IBS and diarrhea to call your office. During the initial part of the screening process, you would exclude patients who have IBS and constipation or IBS and alternating symptoms, because they might develop side effects from the new medication, fail to respond to it, or develop an adverse effect. You would also need to make sure that the patients truly have IBS, not an infectious diarrhea and not inflammatory bowel disease (patients frequently get IBS and IBD mixed up). Participants would need to meet other inclusion and exclusion criteria as well. Finally, potential participants need to be informed about the risks and benefits of entering a research study, and they must agree to show up for all of the visits. In some research studies, at least 5 to 10 people need to be screened to identify one patient who is appropriate to enter the study.
In the fourth phase, the data from the study are collected, analyzed, and recorded in a written report. Only then can it be made available to clinicians and other researchers. It can take years from the time a study first begins until the data are analyzed and written up. As you can tell from this description, clinical research often moves very slowly.
Many patients who have IBS express interest in becoming involved in a research study. People enter research studies for a variety of reasons. Some are seeking a personal benefit; some are motivated by a desire to help others. Here are the most common reasons people enroll in a research study.
Some people enter a research study to learn more about the specific disease being studied. During the study, participants interact with professionals who are very interested in the disorder under investigation. They are often able to provide information about the condition that is not easily accessible elsewhere.
People sometimes join a research study to get better health care than would otherwise be available to them. Participants in a research trial usually undergo a comprehensive history and physical examination and simple laboratory tests or diagnostic studies. This process occasionally uncovers medical problems that the patient was not previously aware of.
Often, patients enroll in a research study in hopes of receiving the new medication being tested. It is important to know that nearly all clinical research studies involve a randomization process (described below); this means that some volunteers will receive a placebo (defined below) rather than the active medication. However, many research studies are now constructed so that patients who were randomized to placebo during the trial have the opportunity to get the active medication at the end of the study.
If a research study is looking for volunteers to undergo testing for a particular condition and a person is concerned that he or she might have that condition, because a relative did, that person may decide to enroll in the study so that he or she can have a specialized test that would likely not be ordered by a primary care provider or paid for by insurance companies.
Fortunately, many people enroll in research studies because they know that by doing so they are helping advance the field of medicine and contribute to human knowledge. They realize that they may not receive any direct benefits, but the knowledge that they have directly contributed to an advance in science is very rewarding.
Lastly, some people enter research trials because they are paid to participate. Being paid may serve as an added incentive and a compensation for people who have other reasons, too, or it may be the principal motivation. For many college students, participating in research studies can be an easy source of extra income.
After reading all of the reasons that people enter research studies, you may feel motivated enough to call your local medical center and sign up for one of the research studies being conducted there. However, it is important to realize that there are potential disadvantages to participating in a research study. First of all, some studies require multiple visits and multiple tests, and these can consume a large amount of time. Second, because most studies place people in groups randomly, you may be put into the placebo group, so if you are entering a study in the hopes of receiving a new medication for your medical problem, you may be disappointed. Most importantly, although safeguards are in place to protect research subjects, there is always the very small possibility that you will have a severe side effect or suffer a complication from entering a research study. Such risks are always carefully explained to research subjects, and complications are very uncommon, but they do happen. I am a strong believer in conducting research and have been a subject in a few research studies myself (as a medical student), but it is important that you understand both sides of the issue.
Most academic medical centers, university hospitals, and large private practice medical groups have ongoing research projects. A research coordinator is usually employed to help manage these projects, and that person can provide you with a list of active research projects at that institution.
If you are interested in research or in joining a research study, you will encounter certain terms with very specific meanings. Below are definitions of some commonly used terms.
Hypothesis. In scientific studies, the hypothesis is a modified version of the question being asked in the study. However, a hypothesis is much more specific and involves some type of a test or experiment with measurements of the outcome. The hypothesis in a research project is often stated so that one condition is compared to another with reference to statistical significance. For example, the original study question may have been, “Will drug X improve symptoms in patients who have IBS?” The study hypothesis, however, might be: “Can drug X improve symptoms of constipation at least 25 percent better than a placebo?”
Observational study. An observational research study simply observes and records behavior; there is no intervention during the study. Thus, an observational research study could involve recording the use of alternative medications over time. Observational studies can be useful for recording the behaviors of large groups of people, but they don’t attempt to provide insight into the actual behavior.
Experimental study. An experimental study always involves an intervention or action. First, symptoms, behaviors, or patterns are observed and recorded, and then an intervention or experiment, such as a medication, a diagnostic test, a change in diet, or the introduction of a new exercise routine, is imposed. Then, the participants are monitored for a specified time period, to see if the intervention had an effect.
Variables. In research studies, a variable is a trait or characteristic that can be altered. Ideally, in each experiment, only one variable is changed at a time, while others are held constant (intentionally not changed). This provides the most precise measurement of how the changed variable has affected the outcome. For instance, if you wanted to refine your favorite cake recipe, you would not alter the amount of several ingredients at once. Rather, you might first change the amount of salt you add and then see whether you like that change. The next time, you might alter the amount of baking powder, and again note the effect on the result.
Longitudinal study. In longitudinal studies, measurement over a period of time is critical to the design of the study (compare to cross-sectional study). Longitudinal studies continue for days, weeks, months, or even years.
Cross-sectional study. Cross-sectional studies involve measurement at a specific point in time. A cross-sectional study might study a single type of behavior in multiple populations, but only on a single day.
Retrospective study. Retrospective research studies ask a research question but try to asnwer it with data that have already been collected. Retrospective studies do not involve collecting new data, recruiting or enrolling new subjects, or studying new medications. They are often used to answer broad-based research questions that involve large populations of people. For example, one could ask what the prevalence of IBS was in the Medicare population during the 1970s and 1980s. Researchers can easily obtain data from government sources regarding patients who were covered by Medicare during these two decades, and then can identify those patients who were diagnosed with IBS. Note that this type of study is not very good at answering questions like why certain patients develop a disorder.
Prospective study. Prospective studies are designed to collect new information over a specific period of time and, in clinical research, generally involve following patients who have a specific medical condition. Prospective studies may track blood pressure, cholesterol levels, or body weight or measure the influence of a new medication. Clinical research studies investigating new medications are always prospective in nature. The best clinical research employs prospective methods.
Subjects. Participants in a research study are called subjects. They may have no health problems, or they may have a particular type of medical problem.
Inclusion criteria. The inclusion criteria are the characteristics that potential participants must have to be eligible to be included in the research study.
Exclusion criteria. Exclusion criteria are the characteristics—symptoms, signs, diagnoses, or test results—that will exclude a person from being eligible to enter the research study.
Bias. In research studies, bias is defined as “systematic error.” The goal of all research studies is to design the study as carefully as possible, so that errors (bias) will not be unintentionally introduced. Bias can occur at any level of a research study, which is why so much time and effort is needed to design a good clinical research study. Let’s consider a simple research study that involves measuring the incidence of IBS symptoms among patients at a specialized research center in the Midwest. This center sees only patients referred by gastroenterologists. The center publishes a study stating that 100 percent of their patients have severe daily abdominal pain that cannot be treated in any way. Are the results applicable to all patients who have IBS? The answer is no. The results of this study are flawed because there is significant bias (systematic error) in the study. There are many biases, and some are easy to recognize. One, it is a specialized research center and its patients are likely not representative of all patients who have IBS. Two, the center is in the Midwest, and its patients may differ in some relevant ways from patients who have IBS in, say, the South or the Southwest. Three, these patients were referred by gastroenterologists for specialized attention, which means they had probably already not improved with over-the-counter medications, therapy provided by their primary care provider, and medical therapy by their gastroenterologist. Thus, there was a “selection bias” that guaranteed a study population who had more persistent symptoms than the population of all people who have IBS.
Intervention. In research studies, an intervention is an action that is imposed on the research participants; the consequences are then measured. An intervention can take many forms. It might be watching a videotape on the benefits of wearing seat belts, to see if it would change the viewers’ behavior. In IBS studies, interventions often include using medications, and then measuring the change in the patients’ symptoms.
Randomized. In a randomized trial, study participants are selected at random to enter one of the treatment groups, either the active medication group or the placebo group (see below). This means that participants are not selected based on race, sex, age, weight, or any other characteristic. Random selection occurs purely by chance. Most research studies now use computer-generated tables to randomly assign study participants to the different treatment groups. Randomization prevents bias, which can either falsely exaggerate or minimize study results.
Placebo. A placebo is essentially a sugar pill. It looks exactly like the study medication (same color, size, shape, texture). The placebo does not have any active ingredients in it, only inactive ones. A good research study always compares the intervention (for example, a medication) to a placebo.
Placebo effect. This is a fascinating topic that was not well understood for many years. In the distant past, when patients were enrolled in a research study, they were all given the study medication and then evaluated for their response. Many research studies were published showing that the medication had remarkably good benefits. However, when the medicine became available to the public, the impressive results from the research studies were not duplicated. This caused a great deal of confusion until scientists and physicians realized that there is a phenomenon called the “placebo effect.” When patients are given a study medication, they often feel better just because they are involved in a research trial: they are having more frequent doctor visits, being seen by more experienced clinicians, often being provided with additional advice and counseling, or may just be convinced that the medication is going to work. If we are told that a medication is going to make us feel better, we are more likely to feel better, regardless of whether we receive the actual medication or a sugar pill. This is the placebo effect. The placebo effect can be as high as 50 percent in some studies; that is, 50 percent of patients who received the placebo feel better. This shows the importance of designing the study so as to compare the intervention, whether it is surgery or a medication, to a placebo, as a control against inflating the benefits of the intervention.
Placebo controlled. Studies that are placebo controlled have a placebo included in the study design. Many studies compare a single intervention (Drug A) to a placebo; others compare several types of intervention (Drug A at ¼ dose, Drug A at ½ dose, Drug A at full dose) to a placebo. Placebo-controlled studies are the best type of clinical research study, because placebos help adjust the results for the “placebo effect” (see above).
Blind and double blind. Blinding in a study means that participants are unaware of whether an active medication (the study drug) or a placebo (a sugar pill) is being administered. In some studies, only the subjects are “blinded,” whereas in other studies only the practitioners dispensing the substance and evaluating the patients are “blinded.” In a double-blind study, both subjects and practitioners are unaware of who is receiving the active drug and who is getting the placebo. Blinding is important because it helps the true effects of the medication to be properly judged. If the study subjects are blinded, they will not know whether they are receiving active medication or a placebo. All of the pills will look identical in color, size, and shape and will be taken in the same number and at the same time each day. If the physicians dispensing the substances don’t know whether their patients are taking the active medication or the placebo, the physicians can’t unintentionally influence their patients with subtle signals about whether the medication is going to work or not. Such subliminal messages can inflate the real effects of the medication. Double blinding thus minimizes bias from both patients and providers.